The effect of traditionally marginalized groups in advertising on consumer response

Abstract

Increasingly, national brands have cast people from marginalized groups in advertising. It is important to understand the elements that influence consumers’ responses to advertisements featuring groups who have been traditionally excluded from advertising campaigns. Although consumers may wish to buy brands that support their own views on human rights and equality, we propose that consumers in the target market may be uncomfortable if the group portrayal contradicts their beliefs about the topic, concept, or social groups in the ad. Across two studies, we show that when an ad creates this type of internal contradiction within consumers, it may elicit a more negative response than an ad portraying a more traditional model. However, we also find preliminary evidence that using models from marginalized groups can be more effective than using more traditional models—as long as such portrayals do not violate certain target consumers’ schemas. The implications of these results will be discussed.

Marketers use advertising as an agent of change. Whereas some brands take a political or social stand to effect change, ultimately, marketers aim to change consumers’ propensity to purchase their products by creating brands that directly relate to consumers’ core beliefs, values, and needs (Paharia et al. 2014; Sandikci and Ekici 2009). As consumers’ values, lifestyles, and beliefs evolve, brands must also evolve in how they speak to these consumers if they are to maintain consumer loyalty. In an effort to stay abreast of cultural shifts, or even possibly to help bring them about, brands and marketers have recently used advertising that shatters long-held stereotypes—not just to break through and drive interest but to speak to evolving customs and new, growing cohorts of potential consumers (Jaffe 2015). Some observers may view the greater use of traditionally marginalized groups such as same-sex parents and multiracial families as “activist advertising.” However, the end goal for marketers is inherently to connect with consumers in order to enrich brand perceptions and drive sales (e.g., Bonchek and France 2016). In the current research, we explore how the use of more diverse and non-traditional portrayals in advertising affects consumer reactions to a brand. Specifically, we seek to determine whether their use creates a positive or negative effect, particularly when the ad’s social cues seem incongruous with consumers’ sense of identity.

Gone are the days when brands could simply provide a message. Today, consumers expect engagement. To relate to consumers, brands must create a relationship and be part of their consumers’ everyday conversation (e.g., Bonchek and France 2016; Fournier 1998). To create engagement, brands are increasingly creating mass-market campaigns that show individuals and groups from various backgrounds (Zmuda 2014) and possibly, to effect even greater change. All-American brands like Cheerios and Chevrolet have not shied away from seeking to be more inclusive in their advertising despite receiving significant backlash on social media. In 2013, Cheerios aired a commercial showing an interracial couple with their bi-racial daughter. In 2014, Chevrolet aired a commercial with the tagline “The new us,” featuring various family vignettes, including same-sex and interracial couples.

Some brands have gone a step further by not simply depicting marginalized groups in their advertising, but also using ads with these groups to broaden and define their brands’ values. For example, a 2014 Honey Maid commercial features a two-dad family, and other non-traditional families, all of whom are portrayed as “wholesome.” This ad not only shows a diverse range of families but also directly challenges the belief that non-traditional families are less “wholesome” or “family-like” (Kille and Tse 2017; Weigel 2008). In this way, some brands are using their advertising platform to promote positive messages about traditionally unrepresented groups that actively counter negative stereotypes about these groups. In hopes of attracting socially liberal consumers, these ads do not merely reflect societal changes already in place—they reflect consumers’ desire that brands effect change and move society forward.

Despite the increased prevalence of such non-traditional portrayals in advertising, the practice has received relatively little scholarly attention. As a result, consumer reactions to this approach are not well understood. Prior related research has shown that cause marketing, or marketing that explicitly supports a societal issue, is related to purchase activism and actual purchasing behavior (e.g., Ballings et al. 2018). Purchase activism can be defined as the desire to purchase items from brands that embody values similar to one’s own (Paharia et al. 2014). Marketing efforts that reflect values related to societal issues or helping others have been shown to positively influence motivations to purchase from brands (Ellen et al. 2006). Furthermore, purchase activism has been shown to serve as a driving motivation linking competitive brand framing to purchase intentions (Paharia et al. 2014). Specifically, Paharia and colleagues found that when small brands (e.g., local coffee shop) were framed as being in competition with large, well-known brands (e.g., Starbucks), this increased purchase activism, which in turn increased intentions to purchase from smaller brands.

Research on cause marketing has shown that consumers are driven to support brands that stand for issues they deem important; however, the issues used in previous research have tended to focus on specific causes rather than marginalized groups. Research on the use of marginalized and underrepresented groups in marketing has largely focused on how these various groups are portrayed in advertising (e.g., Bailey 2006; Lee and Joo 2005) and how prejudice toward the portrayed groups drives negative responses to the ad (Angelini and Bradley 2010; Bhat et al. 1998; Puntoni et al. 2011). Our goal in the current research is to extend beyond previous literature to explore other factors that drive the effectiveness of non-traditional portrayals in ads.

Marketers who want their brands to (1) become part of the cultural dialogue, (2) capitalize on changing social trends, and (3) counter stereotypes need to understand how their approach influences their target consumer’s cognitive framework (schema) as it relates to target consumers’ most basic values. Although the Honey Maid ad was a commentary on how nuclear families have evolved, it also had to appeal to its core target market (i.e., families with children). Thus, the ad’s message, imagery, and cues needed be highly relevant to families with children regardless of any additional message they intended to portray (Aaker et al. 2000; Whittler and DiMeo 1991).

Target marketing involves creating advertising that provides cues with which consumers can identify. These cues may include factors like age, ethnicity, socio-economic status, or culturally relevant situations. These cues are intended to trigger schemas aligned with the target consumer’s characteristics and experiences. A schema is a stored framework of knowledge that represents information about a topic, a concept, or a particular social group (Fiske and Linville 1980). Activating relevant schemas by having the ad depict situations that are familiar, personal, and important to the target creates relatability and increases the likelihood that the target identifies with the brand (Brumbaugh 2002; Deshpande and Stayman 1994). The extent to which a brand is perceived as relevant to one’s self-image is positively correlated with one’s satisfaction with the brand and predicts preferences for one brand over others (Jamal and Goode 2001).

However, if the intended schema is depicted incorrectly or contains cues that are incongruous with the target consumer’s schema, the resulting disconnect may elicit negative emotions (Bettencourt et al. 1997; Roese and Sherman 2007) and ultimately reduce the desire to purchase. For instance, when examining the effectiveness of using self-relevant family schemas to target black consumers, Brumbaugh (2002) found that black participants evaluated ads portraying a white family alongside black cultural cues (e.g., map of Africa, Kente cloth) more negatively than similar ads with a black family. Although not directly tested, Brumbaugh posited that this negative reaction was due to the incongruity elicited by embedding white models within the context of black cultural cues. White participants (i.e., non-target consumers) did not demonstrate these strong negative reactions to the incongruent ads, suggesting that schema violations are perceived more aversively by target consumers than non-target consumers.

We propose that consumers targeted by ads with portrayals of marginalized groups will be highly sensitive to the alignment of the ad’s message with their expectations and values. Target consumers may be motivated to respond positively when a non-traditional approach (i.e., use of marginalized groups) aligns with their personal beliefs regarding egalitarianism and/or because it is a socially desirable response (see Plant and Devine 1998). Given the recent socially liberal shift in the USA (Solomon 2014; Zmuda 2014), simply seeing traditionally excluded groups in ads should elicit positive reactions such as increased purchase activism and overall favorable perception of the brand—unless the ads include cues that violate consumers’ schemas. When schema violations occur, we expect similar negative reactions from targeted consumers to those found by Brumbaugh (2002).

In sum, we predict that when models from marginalized groups are shown in a way that is consistent with the consumer’s schema, it will have a positive effect on brand evaluations (driven by the consumer’s purchase activism). However, when models from marginalized groups are shown along with information that is perceived as incongruous with the consumer’s schema, it will have a negative effect on brand evaluations (driven by consumers’ negative emotions). We believe these effects will be the most pronounced for target consumers since they are the most sensitive to an ad’s message and how it aligns with their personal beliefs.

Study 1

Given the increased incidence of same-sex parents in commercials (e.g., Honey Maid, Wells Fargo, Hallmark), we used this demographic to test our predictions. We examined how target (parents) and non-target consumers (non-parents) responded to a print ad showing two fathers as parents relative to a traditional mother/father parental structure.

Participants

One hundred seventy-three US-based participants (53% male, 47% female; Mage = 35.47, SD = 11.25) were recruited through Amazon’s Mechanical Turk and completed the experiment through the web-based survey software Qualtrics.

Procedure

Using a between-subjects design, participants viewed a print ad for Cheerios that featured an image of either a family with two male parents or a family with one male and one female parent (see Appendix for sample stimuli). We identified families with children as the target audience, as market analysis has shown that households with children are more likely to purchase Cheerios than those without children (Anderson 2012). Cues that signal family were used to manipulate self-relevant schema congruity. Specifically, schema congruency was manipulated by positioning Cheerios as offering either “convenient breakfasts for busy families” (schema neutral; control) or “wholesome breakfasts for wholesome families” (schema incongruent).Footnote 1

After viewing the ad, participants completed a three-item measure of brand evaluation using 7-point scales. Participants rated the extent to which the ad (1) changed their interest in purchasing the product (1 = low interest, 7 = high interest), (2) influenced their opinion of the product (1 = very bad, 7 = very good), and (3) influenced the extent to which they liked the advertised brand (1 = dislike very much, 7 = like very much). Negative emotions were measured by asking participants to rate the extent to which they felt the following eight emotions after viewing the ad: disgusted, angry, revolted, irritated, accepting, happy, affectionate, and pleased (last four items reverse-scored; scale ranged from 1 = not at all to 7 = very much; Bhat et al. 1998). Purchase activism was measured with four items from Paharia et al.’s (2014) Purchase Activism Scale. Using a 5-point scale (1 = do not agree at all, 5 = agree completely), participants responded to the following items: (1) I would purchase this product because the brand embodies the ideals that I live by, (2) I would purchase this product because the brand is aligned with my moral beliefs, (3) I believe that purchasing this product would help make a difference in the world, and (4) Cheerios helped make a political statement. To determine target market (i.e., parental status), participants were asked whether they had children. Finally, participants completed Herek’s (1988) Attitudes toward Gay Men Scale and provided demographic information.

Results

We conducted a 2 (ad diversity: gay parents vs. straight parents) × 2 (audience: parents vs. non-parents) × 2 (schema: busy vs. wholesome families) ANOVA on brand evaluations, negative emotions, and purchase activism.Footnote 2 Attitudes toward gay men served as a covariate for all analyses (the covariate did not significantly affect the dependent variable) (see Table 1 for correlations and descriptive statistics for our measured variables).

Table 1 Descriptive statistics and correlations of measured variables in Study 1

Brand evaluations

Analysis revealed a significant three-way interaction (F(1, 164) = 14.18, p < .01; see Fig. 1). Parents (target consumers) reacted significantly less favorably to the gay parents (M = 4.31, SD = 1.38) versus the straight parents ad (M = 5.04, SD = 0.79) when the “wholesome families” schema was activated (F(1, 164) = 5.79, p < .05). However, when the “busy families” schema was activated, the gay parents ad elicited more favorable brand evaluations (M = 5.64, SD = 0.99) than the straight parents ad (M = 4.75, SD = 0.66) among parents (F(1, 164) = 8.76, p < .01). In contrast, non-parents (non-target consumers) responded more favorably to the gay parents ad (M = 5.03, SD = 0.84) versus the straight parents ad (M = 4.54, SD = 0.63) when the “wholesome family” schema was activated (F(1, 164) = 3.57, p < .06). No significant differences were seen based on the family portrayed when the “busy family” schema was activated (Mgay = 4.86, SD = 0.79; Mstraight = 4.85, SD = 1.12; F(1, 164) = 0.01, n.s.).

Fig. 1
figure1

Brand evaluations as a function of audience, ad diversity, and schema in Study 1

Negative emotions

A significant three-way interaction between ad diversity, audience, and schema was found (F(1, 164) = 8.85, p < .01). Parents expressed significantly more negative emotions for the gay parents (M = 3.71, SD = 1.68) versus the straight parents ad (M = 2.48, SD = 1.09; F(1, 164) = 11.54, p < .01) when wholesome family was activated but did when the busy family schema was activated (Mgay = 1.99, SD = 0.95; Mstraight = 2.35, SD = 0.78; F(1, 164) = 1.41, n.s.). For non-parents, the non-traditional casting had no effect on negative emotions in either the wholesome (Mgay = 2.26, SD = 1.01; Mstraight = 2.78, SD = 1.00; F(1, 164) = 2.26, n.s.) or busy family conditions (Mgay = 2.48, SD = 1.30; Mstraight = 2.60, SD = 0.79; F(1, 164) = 0.08, n.s.).

Purchase activism

Analyses showed a significant three-way interaction (F(1, 164) = 3.62, p = .05). Parents expressed significantly higher purchase activism after viewing the gay parents (M = 3.01, SD = 1.44) versus the straight parents ad (M = 1.81, SD = 1.05; F(1, 164) = 9.05, p < .01) when the busy family schema was activated, but when the wholesome family schema was activated (Mgay = 2.28, SD = 1.32; Mstraight = 2.36, SD = 1.23; F(1, 164) = 0.04, n.s.). Non-parents reported greater purchase activism when viewing the gay parents versus the straight parents ad in both the wholesome (Mgay = 2.95, SD = 1.12; Mstraight = 1.96, SD = 0.98; F(1, 164) = 8.71, p < .01) and busy family conditions (Mgay = 2.84, SD = 1.43; Mstraight = 1.97, SD = 1.14; F(1, 164) = 6.82, p < .01).

Mediation analysis

We proposed that the divergent effects of using non-traditional portrayals on consumers’ responses are driven by negative emotions and purchase activism when the “wholesome families” schema is activated, and this should be especially the case for target consumers. To test our predictions, we conducted moderated mediation analysis (Hayes 2013; Model 12) with the use of ad diversity (straight parent ad = 1, gay parent ad = 0) as the independent variable, audience (parent = 0, nonparent = 1) and activated schema (wholesome family = 0, busy family = 1) as moderators, negative emotions and purchase activism as mediators, and brand evaluations as the dependent variable.

For parents viewing the “wholesome families” ads, negative emotions significantly mediated the negative effect of ad diversity on brand evaluations (b = − 0.37; CI − 0.78, − 0.08); however, purchase activism (b = − 0.03; CI − 0.50, 0.42) did not mediate this effect. In contrast, for parents viewing the “busy families” ads, purchase activism (b = 0.57; CI 0.17, 1.00), but not negative emotions (b = 0.11; CI − 0.08, 0.31), mediated the positive effect of ad diversity on brand evaluations. For non-parents, only purchase activism drove the effect of ad diversity on brand evaluations for both the “wholesome families” (b = 0.48; CI 0.23, 0.78) and “busy families” ads (b = 0.42; CI 0.07, 0.81). Negative emotions did not have a significant indirect effect for either the “wholesome families” (b = 0.42; CI 0.07, 0.81) or the “busy families” ads for non-parents (b = 0.42; CI 0.07, 0.81).

Discussion

The results from Study 1 show that using non-traditional portrayals of marginalized groups in ads influences consumers’ attitudes. However, other factors such as schemas also come into play. If the ad was consistent with—or at least not in violation of—their schemas, target consumers expressed greater purchase activism, which drove more favorable brand evaluations for an ad showing two fathers versus traditional parents. However, when the non-traditional portrayal also included elements that were incongruent with the target consumer’s schema, target consumers reported more negative emotions that led to less favorable ad evaluations relative to the more traditional portrayal. Use of the two-father parents did not appear to negatively affect non-target consumers: they reported higher brand evaluations for the two-father ad than for the traditional parent ad but only in the schema-incongruent condition. The responses of non-target consumers appear to map more closely with what marketers are likely expecting when developing ads that defy negative beliefs about marginalized social groups (Zmuda 2014).

Taken together, these findings suggest that consumers respond positively to most portrayals of a non-traditional family with the exception of target consumers in the schema violation condition. The positive effects may have occurred because consumers generally find it refreshing to see non-traditional, diverse portrayals in ads (Solomon 2014). Another possibility is that participants engaged in socially desirable responding to avoid appearing prejudiced (Plant and Devine 1998). However, this may be less likely as there is evidence suggesting that there is no relation between prejudicial attitudes toward gay men and external motivation to respond without prejudice toward gay men (Ratcliff et al. 2006). Finally, the positive findings may be the result of alignment between ads showing non-traditional models and people’s egalitarian values, as we propose. This idea is supported by the higher levels of purchase activism reported by those viewing the non-traditional versus traditional ad, leading to higher levels of brand evaluations (with the exception of target consumers in the schema-incongruent condition).

These findings begin to shed light on why target versus non-target audiences may respond differently to more diverse portrayals in advertising; however, we note two limitations. First, since we only looked at the use of same-sex couples, it is unclear whether similar results would occur with other non-traditional or marginalized groups in advertising. Second, although families were identified as the target audience for this study, it is possible that gay men and lesbians perceived themselves as target audience members and therefore responded to the ads differently than heterosexual individuals. Because gay men and lesbians were not the identified target audience, survey participants were not asked to disclose their sexual orientation. Therefore, it is not possible to say whether the participants’ own sexual orientation influenced the results.

To address these limitations, we conducted Study 2 using the same methodology but used overweight models to target women in advertising for bottled water. Overweight women are typically a marginalized group and are seldom portrayed in advertising (Merrick 2014). In Study 2, we examine how portraying overweight women in situations that violate the target’s schemas can influence the reactions of target consumers (women) as compared to non-target consumers (men). Although water is a good consumed by everyone regardless of demographics, women were selected as the target audience because women have been identified as the target audience for some bottled water campaigns (Newman 2013). The use of a different marginalized group in the ad allows us to address the first limitation noted by ensuring that our findings are not dependent on consumers’ feelings about portrayals of people in the minority regarding sexual orientation. We address the second limitation by gathering information not only on the identified target group (i.e., women) but also on the potential perceived target group (i.e., overweight individuals).

We also extend Study 1 by testing multiple types of incongruent schemas to gain a better understanding of the nature of our effects. Based on stereotypes that overweight individuals are unhealthy, unintelligent, and lacking in willpower (c.f., Puhl et al. 2008), we posited that consumers would perceive the portrayal of an overweight woman as inconsistent with their expectations of a “successful woman” and “active woman.” Thus, portraying an overweight woman as successful or active will elicit negative emotions, and subsequently, more negative brand evaluations for target but not for non-target consumers.

Study 2

Participants

Three hundred seven US-based Mechanical Turk participants were recruited for this study (58% male, 42% female; Mage = 34.16, SD = 10.82).

Procedure

Using a between-subjects design, participants viewed a print ad promoting Dasani water that featured either a thin or overweight woman. A thin model was selected as the schema-congruent representation, as the majority of models seen in advertisements are thin (Merrick 2014). To manipulate schema violation, we positioned Dasani as offering “convenient drinks for busy women” (schema neutral), “premium drinks for successful women” (schema incongruent), or “energizing drinks for active women” (schema incongruent; see Appendix for sample stimuli).Footnote 3 After completing the same brand evaluation, purchase activism, and negative emotion measures used in Study 1, participants also completed the Anti-fat Attitude Scale (Crandall 1994) and provided demographic information, which included height and weight, allowing us to calculate BMI. They also completed manipulation check about the model’s size by responding to 1 item: “How would you describe the woman in the ad,” using a 7-point scale (1 = Very Overweight to 7 = Very Thin).

Results

A manipulation check confirmed that participants viewing the overweight model (M = 2.48, SD = 1.42) perceived her as heavier than participants viewing the thin model (M = 5.0, SD =  1.38; F(1, 293) = 495.28, p < .0001). We tested our predictions by conducting a 2 (audience: men vs. women) × 2 (ad diversity: thin vs. overweight) × 3 (schema: busy vs. successful vs. active) ANOVA on each dependent variable, with anti-fat attitudes and BMI as covariates (which were not significant in analyses) (see Table 2 for correlations and descriptive statistics for our measured variables).

Table 2 Descriptive statistics and correlations of measured variables in Study 2

Brand evaluations

Analysis revealed a significant three-way interaction (F(1, 293) = 3.78, p < .05). For female participants (see Fig. 2), the overweight model ad garnered less favorable evaluations relative to the thin model ad, but only when the “successful woman” schema was activated (MOverweight = 3.82, SD = 1.01 vs. MThin = 4.52, SD = 1.04; F(1, 293) = 4.42, p < .05). Although the overweight model ad garnered directionally more favorable evaluations than the thin model ad when the “busy woman” schema was activated, this contrast did not reach conventional levels of significance (MOverweight = 4.49, SD = 1.44 vs. MThin = 3.87, SD = 1.30; F(1, 293) = 2.26, p = .13). When the “active woman” schema was activated, there was no effect of model weight (MHeavy = 4.35, SD = 0.96 vs. MThin = 4.56, SD = 0.83; F(1, 293) = 0.27, NS).

Fig. 2
figure2

Brand evaluations as a function of audience, ad diversity, and schema in Study 2

In contrast, male participants’ brand evaluations were not affected by model size in the “successful woman” (MOverweight = 4.33, SD = 1.34 vs. MThin = 3.95, SD = 1.38; F(1, 293) = 1.49, NS), “busy woman” (MOverweight = 3.96, SD = 1.31 vs. MThin = 4.13, SD = 1.09; F(1, 293) = 0.35, NS), or the “active woman” conditions (MOverweight = 4.03, SD = 1.19 vs. MThin = 4.04, SD = 1.80; F(1, 293) = 0.01, NS).

Negative emotions

Analysis revealed a significant three-way interaction (F(1, 293) = 3.08, p < .05). For female participants, the overweight model ad garnered more negative emotions relative to the thin model ad, but only when the “successful woman” schema was activated (MOverweight = 3.77, SD = 1.31 vs. MThin = 3.03, SD = 1.05; F(1, 293) = 4.07, p < .05) and not when the “busy woman” (MOverweight = 3.11, SD = 1.27 vs. MThin = 3.61, SD = 1.49; F(1, 293) = 1.53, NS) or the “active woman” schema was activated (MOverweight = 2.81, SD = 1.24 vs. MThin = 2.84, SD = 0.82; F(1, 293) = 0.01, NS). In contrast, male participants’ negative emotional responses were not affected by model size in the “successful woman” (MOverweight = 2.96, SD = 1.18 vs. MThin = 3.31, SD = 1.24; F(1, 293) = 1.35, p = .25), “busy woman” (MOverweight = 3.35, SD = 1.21 vs. MThin = 3.30, SD = 1.25; F(1, 293) = 0.03, NS), or the “active woman” conditions (MOverweight = 2.97, SD = 1.16 vs. MThin = 3.57, SD = 1.34; F(1, 293) = 2.34, p = .13).

Purchase activism

None of our independent variables had a significant effect on participants’ purchase activism (F’s < 2.00, p’s > .14). Female participants’ purchase activism ratings were not affected by model size in the “successful woman” (MOverweight = 2.10, SD = 1.41 vs. MThin = 2.00, SD = 1.20; F(1, 293) = 0.01, n.s.), “busy woman” (MOverweight = 2.34, SD = 1.37 vs. MThin = 1.25, SD = 1.09; F(1, 293) = 1.75, n.s.), or the “active woman” conditions (MOverweight = 2.10, SD = 1.07 vs. MThin = 2.38, SD = 1.20; F(1, 293) = 0.56, n.s.). Similarly, male participants’ purchase activism ratings were not affected by model size in the “successful woman” (MOverweight = 2.10, SD = 1.09 vs. MThin = 1.71, SD = 1.17; F(1, 293) = 1.73, n.s.), “busy woman” (MOverweight = 1.77, SD = 1.08 vs. MThin = 1.85, SD = 1.20; F(1, 293) = 0.08, n.s.), or “active woman” conditions (MOverweight = 1.83, SD = 1.00 vs. MThin = 2.35, SD = 1.58; F(1, 293) = 1.91, n.s).

Mediation analysis

For each schema condition, we used moderated mediation analysis to examine the effects of ad diversity, target audience, negative emotion, and purchase activism on brand evaluations (model 8; female = 0, male = 1; overweight model = 0, thin model = 1). For female participants, results indicated that negative emotions (indirect effect 0.51; 95% CI 0.09, 0.95) but not purchase activism (indirect effect − 0.02; 95% CI − 0.23, 0.17) mediated the effect of ad diversity on brand evaluations in the successful woman schema condition. In contrast, none of the mediation pathways were significant for male participants (negative emotions indirect effect − 0.23; 95% CI − 0.62, 0.18; purchase activism indirect effect − 0.07; 95% CI − 0.27, 0.07). Analogous analyses for the “busy woman” and “active woman” conditions did not yield any significant indirect effects for female or male participants.

Discussion

Results of Study 2 provide further support for our prediction that non-traditional portrayals in advertising will elicit negative responses from consumers if they are incongruent with a consumer’s schema, especially for target consumers. However, the findings also show that violating a schema alone is not enough to prompt negative reactions. Although the overweight model was incongruent with schemas about both successful women and active women, the use of such models only led to increasingly negative emotions and brand evaluations for one incongruent category, the successful woman schema. This finding suggests that the use of schema incongruity in ads is complex and is likely dependent on numerous factors. One potential reason for the negative consequences of the overweight model for the ad depicting the “successful” woman but not the ad showing the “active” woman may be related to perceptions of threat toward one’s self and how participants perceive the controllability of the descriptor used in the ad. Specifically, being active is something that is more directly under an individual’s control; one decides to engage in exercise or does not. Therefore, the ad describing the overweight woman as active may not elicit a threat to one’s own sense of whether or not they can be active. However, the idea of success for women, particularly in the workplace, is not something that may not be perceived as completely within one’s control. Indeed, several external barriers to women’s success (such as discrimination and sexism) have been well documented in the literature (see Heilman and Eagly 2008). Thus, the incongruence between the schema one has for success and the ad showing an overweight woman as successful may have triggered feelings of threat for female participants.Footnote 4 In Section 7, we expand on this and further explore potential reasons why some schema violations may be more damaging to evaluations than others.

We did not find a positive effect when the non-traditional model was portrayed in a schema-neutral manner for either women (target) or men (non-target). Although inconsistent with findings from Study 1 showing positive effects of non-traditional models for non-target consumers in the schema-incongruent condition and target consumers in the schema-neutral condition, the lack of replication is not completely surprising. Some research suggests that the bias against heavy individuals is greater than bias against gay men, lesbians, and bisexuals (Latner et al. 2008). Further, weight stigmatization is ubiquitous and seen even among people who are members of the marginalized group (c.f., Schwartz et al. 2006). Given the strong weight bias, consumers in both the target and non-target groups may not feel the need to show extra support for the portrayal of overweight models. That is, consumers may feel that rewarding brands for portraying heavier models in their advertising is not required for them to feel aligned with their moral values and egalitarian beliefs.

General discussion

The use of non-traditional models in ads has multifaceted implications. These two studies suggest that the use of non-traditional groups in advertising can affect how consumers feel about a brand. Although use of non-traditional models can create positive reactions, several factors, including the non-traditional group displayed, the target consumer market, whether consumers’ schemas are violated, and the type of schema that is violated, mitigate these reactions. A one size fits all approach should not be used when deciding how to best portray non-traditional models in ads; rather, brands must consider a number of factors to achieve the best outcome for the brand and the non-traditional group displayed.

Using different portrayals of non-traditional models and different target consumers, we show that the negative emotions elicited by schema violations for targeted consumers can drive negative reactions, especially in how they evaluate the advertised brand. This effect held when a family with two-father parents violated the “wholesome families” schema and when an overweight model violated the “successful woman” schema. Prior research suggests that negative emotions stemming from schema incongruity can occur for several reasons. It may be the case that consumers become frustrated with their inability to reconcile the schema incongruity. As a result, they react negatively, which in turn affects their brand evaluation (Peracchio and Tybout 1996; Torelli and Ahluwalia 2011). People in the target audience may have felt increased levels of frustration in the schema-incongruent condition because the schema was particularly relevant to their own self-image. However, the lack of negativity, and in some instances the outright positivity, seen from non-targeted consumers in schema-incongruent conditions suggests that this frustration hypothesis cannot completely explain the pattern of results in the current studies.

Another possibility is that consumers use their own schemas to define and evaluate themselves (Markus 1977). By violating these closely-held schemas, an ad may inadvertently elicit threat from targeted consumers. This perspective may help explain why parents viewed the “wholesome” same-sex ad negatively. A family that is “wholesome” is not only judged by how it looks or acts, but also by how it measures up to traditional societal norms of what is moral, pure, clean and innocent. It may also explain why showing an overweight woman as “active” (a schema violation) was not perceived negatively by female participants in Study 2, but showing an overweight woman as successful was viewed negatively. Specifically, being active is within a woman’s control. In contrast, being successful and wholesome is highly influenced by how one is perceived, and thus less controllable than being active. Although people can strive to be successful or wholesome, these goals are more contingent on external barriers (e.g., success for women in the workplace is often hampered in part by gender biases that are prevalent; e.g., Rudman and Glick 1999) or external perceptions that act as barriers (e.g., the concept of wholesomeness is socially constructed in part by people’s views of morality and by religious ideologies; e.g., Goodwin et al. 2014). Thus, attaining success or wholesomeness is not entirely within one’s own control. People face fewer direct barriers from others in their own goal toward being active or not. Therefore, although showing an overweight woman as active violates a schema, this violation may be seen as less problematic and threatening by women because they have more control over whether or not they can actually become active women.

However, targeted consumers may feel more threatened by non-traditional portrayals that challenge identity-relevant schemas related to domains where they have less control and security. If the portrayal does not correspond to something over which they feel they can have primary control—as is the case of being perceived as wholesome or successful—it may create a sense of insecurity and unease. Indeed, people who are insecure in their abilities or position in important social groups may be more likely to feel threatened (Finkelstein and Fishbach 2012; Sommer and Baumeister 2002; Ward and Dahl 2014). Thus, targeted consumers may find inaccurate portrayals of schemas related to self-relevant social groups that have external barriers to membership particularly disconcerting. Although identity threat was not measured in the current study, our findings may provide initial evidence that ads portraying marginalized groups with messages that seem incongruent to the target can elicit negative responses because they call into question how the target sees him or herself. Future research is needed to examine the extent to which identity threat is a factor in leading to target audiences’ negative reactions to schema-incongruent ads portraying diversity and what brands can do to prevent negative reactions from occurring. Future research should more directly examine the role of perceived controllability in the extent to which schema violations influence reactions to marketing ads. Furthermore, future research should examine additional factors that may serve as a mechanism for explaining the relation between marketing ads that violate schemas associated with negative stereotypes about marginalized groups and brand evaluations. Based on our findings, we believe that it is likely that multiple psychological and emotional factors may influence this relationship, and the extent to which different factors (such as purchase activism) drive this relationship varies based on the schema violation that is present.

The complexity of effectively capitalizing on the use of non-traditional models in ads was also demonstrated in the inconsistencies of positive effects seen across Studies 1 and 2. Brands should carefully consider how seemingly positive portrayals of traditionally marginalized groups may be perceived when other historical and societal factors are considered. For instance, in early 2017 Pepsi aired an ad that attempted to show a message of global unity using a “happy” protest theme. Many consumers perceived this as making light of real troubling issues facing marginalized groups where protest is used, such as police brutality in the U.S. Another pertinent example is the October 2017 Facebook ad for Dove featuring an image of a black woman removing a brown shirt and appearing to transform into a white woman removing a similar shirt. Unilever’s stated intent was to represent the “diversity of real beauty” (Wootson 2017). However, many target consumers understood the transformation of black women into white women, particularly in the context of a skin-cleansing product, to be a racist depiction reflective of a history of seeing white skin as clean and black skin as dirty (Wootson 2017). The use of models of different races likely would have generated a positive effect for Dove if presented in a different manner. However, the ad overtly ignored the historical and cultural context of negative depictions of African Americans and was therefore seen as offensive to consumers of all racial and ethnic backgrounds.

In sum, our research suggests that there can be benefits to using more diverse portrayals in advertising, especially for brands trying to gain or maintain consumers within a particular market. However, companies seeking to promote diversity and benefit their organization need to carefully consider how non-traditional portrayals can be best used so as not to alienate the target audience. Future research is needed to gain a better understanding of which variables need to be considered so as to optimize positive reactions and avoid negative reactions.

Notes

  1. 1.

    A pretest with a sample of n = 122 participants indicated that relative to a family with a father and a mother, both parents and non-parents perceived a family with same-sex parents as more incongruent with their schemas related to wholesome families (F(1, 118) = 24.19, p < .001) but not incongruent with their schemas of what represents busy families (F(1, 118) = 0.70, n.s).

  2. 2.

    A manipulation check was included in the study to ensure that participants viewed the people in the ad as a family. In the straight parents condition, all participants believed the ad portrayed a committed couple. In the gay parents condition, 69 participants believed the ad portrayed a committed couple, 11 participants believed the ad portrayed two friends, and one participant chose “other.” All participants were included in the following analyses although omitting participants who failed the manipulation check yielded the same pattern of results.

  3. 3.

    Results from a pretest support the schema incongruity manipulations such that overweight models were rated as less congruent than thin models with participants’ schema of a successful woman (F(1, 155) = 5.56, p < .05) and active woman (F(1, 155) = 82.73, p < .001). No differences were seen for the busy schema between perceptions of overweight and thin models (F(1, 155) = 1.03, n.s).

  4. 4.

    An anonymous reviewer raised the question of how people may have interpreted the description of “successful woman,” and whether it was possible that the overweight model was perceived as active but not successful in weight management. Qualitative data from a second pre-study (sample N = 130) examining people’s schemas suggests that over 50% of respondents characterized a successful woman using terms related career and professionalism, whereas only 10% mentioned descriptors related to health or weight. Specifically, many participants described the “successful woman” schema as someone who dresses professionally. These findings suggest that people’s schema of a “successful woman” focused on professional dress to a much greater extent than weight management. When we combine this information with results from our other pre-study showing that the overweight model was incongruent with participants’ schemas about success, we believe that our primary findings occurred because of differences in how participants process the incongruity of overweight and success versus the incongruity of overweight and active (e.g., differences in perceived controllability), not because there was a misinterpretation in the intended schema violation for the success condition.

References

  1. Aaker, J. L., Brumbaugh, A. M., & Grier, S. A. (2000). Non-target markets and viewer distinctiveness: the impact of target marketing on advertising attitudes. Journal of Consumer Psychology, 9, 127–140. https://doi.org/10.1207/S15327663JCP0903_1.

    Article  Google Scholar 

  2. Anderson, J. (2012). Marketing communications plan: Honey Nut Cheerios. Retrieved August 31, 2018 from https://jessicaandersonresume.weebly.com/uploads/1/4/3/0/14305546/honey_nut_cheerios.pdf.

  3. Angelini, J. R., & Bradley, S. D. (2010). Homosexual imagery in print advertisements: attended, remembered, but disliked. Journal of Homosexuality, 57, 485–502. https://doi.org/10.1080/00918361003608665.

    Article  Google Scholar 

  4. Bailey, A. A. (2006). A year in the life of the African-American male in advertising: a content analysis. Journal of Advertising, 35, 83–104. https://doi.org/10.2753/JOA0091-3367350106.

    Article  Google Scholar 

  5. Ballings, M., McCullough, H., & Bharadwaj, N. (2018). Cause marketing and customer profitability. Journal of the Academy of Marketing Science, 46, 234–251. https://doi.org/10.1007/s11747-017-0571-4.

    Article  Google Scholar 

  6. Bettencourt, A., Dill, K. E., Greathouse, S. A., Charlton, K., & Mullholland, A. (1997). Evaluations of ingroup and outgroup members: the role of category-based expectancy violation. Journal of Experimental Social Psychology, 33, 244–275. https://doi.org/10.1006/jesp.1996.1323.

    Article  Google Scholar 

  7. Bhat, S., Leigh, T. W., & Wardlow, D. L. (1998). The effect of consumer prejudices on ad processing: heterosexual consumers’ responses to homosexual imagery in ads. Journal of Advertising, 27, 9–28. https://doi.org/10.1080/00913367.1998.10673566.

    Article  Google Scholar 

  8. Bonchek, M. & France, C. (2016). Build your brand as a relationship. Harvard Business Review. Retrieved from https://hbr.org/2016/05/build-your-brand-as-a-relationship

  9. Brumbaugh, A. M. (2002). Source and nonsource cues in advertising and their effects on the activation of cultural and subcultural knowledge on the route to persuasion. Journal of Consumer Research, 29, 258–269. https://doi.org/10.1086/341575.

    Article  Google Scholar 

  10. Crandall, C. S. (1994). Prejudice against fat people: ideology and self-interest. Journal of Personality and Social Psychology, 66, 882–894. https://doi.org/10.1037/0022-3514.66.5.882.

    Article  Google Scholar 

  11. Deshpande, R., & Stayman, D. M. (1994). A tale of two cities: distinctiveness theory and advertising effectiveness. Journal of Marketing Research, 31, 57–74. https://doi.org/10.2307/3151946.

    Article  Google Scholar 

  12. Ellen, P. S., Webb, D. J., & Mohr, L. A. (2006). Building corporate associations: consumer attributions for corporate socially responsible programs. Journal of the Academy of Marketing Science, 34, 147–157. https://doi.org/10.1177/0092070305284976.

    Article  Google Scholar 

  13. Finkelstein, S. R., & Fishbach, A. (2012). Tell me what I did wrong: experts seek and respond to negative feedback. Journal of Consumer Research, 39, 22–38. https://doi.org/10.1086/661934.

    Article  Google Scholar 

  14. Fiske, S. T., & Linville, P. W. (1980). What does the schema concept buy us? Personality and Social Psychology Bulletin, 6, 543–557. https://doi.org/10.1177/014616728064006.

    Article  Google Scholar 

  15. Fournier, S. (1998). Consumers and their brands: developing relationship theory in consumer research. Journal of Consumer Research, 24, 343–373.

    Article  Google Scholar 

  16. Goodwin, G. P., Piazza, J., & Rozin, P. (2014). Moral character predominates person perception and evaluation. Journal of Personality and Social Psychology, 106(1), 148–168. https://doi.org/10.1037/a0034726.

    Article  Google Scholar 

  17. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: a regression-based approach. New York: The Guilford Press.

    Google Scholar 

  18. Heilman, M. E., & Eagly, A. H. (2008). Gender stereotypes are alive, well, and busy producing workplace discrimination. Industrial and Organizational Psychology: Perspectives on Science and Practice, 1, 393–398.

    Article  Google Scholar 

  19. Herek, G. M. (1988). Heterosexuals’ attitudes toward lesbians and gay men: correlates and gender differences. The Journal of Sex Research, 25, 451–477. https://doi.org/10.1080/00224498809551476.

    Article  Google Scholar 

  20. Jaffe, C. (2015). How advertising has become an agent of social change. Retrieved from https://medium.com/@moonstorming/how-advertising-has-become-an-agent-of-social-change-148aa0ef303a

  21. Jamal, A., & Goode, M. M. H. (2001). Consumers and brands: a study of the impact of self-image congruence on brand preference and satisfaction. Marketing Intelligence & Planning, 19(7), 482–492. https://doi.org/10.1108/02634500110408286.

    Article  Google Scholar 

  22. Kille, D. R., & Tse, C. T. (2017). Whose family fits? Categorization and evaluation of same-sex and cross-race-parent families. Group Processes & Intergroup Relations, 20, 109–124. https://doi.org/10.1177/1368430215595106.

    Article  Google Scholar 

  23. Latner, J. D., O’Brien, K. S., Durso, L. E., Brinkman, L. A., & MacDonald, T. (2008). Weighing obesity stigma: the relative strength of different forms of bias. International Journal of Obesity, 32, 1145–1152. https://doi.org/10.1038/ijo.2008.53.

    Article  Google Scholar 

  24. Lee, K., & Joo, S. (2005). The portrayal of Asian Americans in mainstream magazine ads: an update. Journalism and Mass Communication Quarterly, 82, 17–30. https://doi.org/10.1177/107769900508200311.

    Article  Google Scholar 

  25. Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63–78. https://doi.org/10.1037/0022-3514.35.2.63.

    Article  Google Scholar 

  26. Merrick, A. (2014). Marketing “Real” Bodies. The New Yorker. Retrieved from https://www.newyorker.com/business/currency/marketing-real-bodies

  27. Newman, A. A. (2013). Nestle adds premium brand in still water arena. The New York Times. Retrieved from http://www.nytimes.com/2013/06/10/business/media/nestle-adds-a-premium-brand-in-still-water-category.html?ref=business&_r=1&.

  28. Paharia, N., Avery, J., & Keinan, A. (2014). Positioning brands against large competitors to increase sales. Journal of Marketing Research, 51, 647–656. https://doi.org/10.1509/jmr.13.0438.

    Article  Google Scholar 

  29. Peracchio, L. A., & Tybout, A. M. (1996). The moderating role of prior knowledge in schema-based product evaluation. Journal of Consumer Research, 23, 177–192. https://doi.org/10.1086/209475.

    Article  Google Scholar 

  30. Plant, E. A., & Devine, P. G. (1998). Internal and external motivation to respond without prejudice. Journal of Personality and Social Psychology, 75, 811–832. https://doi.org/10.1037/0022-3514.75.3.811.

    Article  Google Scholar 

  31. Puhl, R. M., Moss-Racusin, C. A., Schwartz, M. B., & Brownell, K. D. (2008). Weight stigmatization and bias reduction: perspectives of overweight and obese adults. Health Education Research, 23, 347–358. https://doi.org/10.1093/her/cym052.

    Article  Google Scholar 

  32. Puntoni, S., Vanhamme, J., & Visscher, R. (2011). Two birds and one stone: purposeful polysemy in minority targeting and advertising evaluations. Journal of Advertising, 40, 25–42. https://doi.org/10.2753/JOA0091-3367400102.

    Article  Google Scholar 

  33. Ratcliff, J. J., Lassiter, G. D., Markman, K. D., & Snyder, C. J. (2006). Gender differences in attitudes toward gay men and lesbians: the role of motivation to respond without prejudice. Personality and Social Psychology Bulletin, 32, 1325–1338. https://doi.org/10.1177/0146167206290213.

    Article  Google Scholar 

  34. Roese, N. J., & Sherman, J. W. (2007). Expectancy. In A. W. Kruglanski & E. T. Higgins (Eds.), Social psychology: A handbook of basic principles (Vol. 2, pp. 91–115). New York: Guilford Press.

    Google Scholar 

  35. Rudman, L. A., & Glick, P. (1999). Feminized management and backlash toward agentic women: the hidden costs to women of a kinder, gentler image of middle managers. Journal of Personality and Social Psychology, 77(5), 1004–1010. https://doi.org/10.1037/0022-3514.77.5.1004.

    Article  Google Scholar 

  36. Sandikci, O., & Ekici, A. (2009). Politically motivated brand rejection. Journal of Business Research, 62(2), 208–217. https://doi.org/10.1016/j.jbusres.2008.01.028.

    Article  Google Scholar 

  37. Schwartz, M. B., Vartanian, L. R., Nosek, B. A., & Brownell, K. D. (2006). The influence of one’s own body weight on implicit and explicit anti-fat bias. Obesity: A Research Journal, 14, 440–447. https://doi.org/10.1038/oby.2006.58.

    Article  Google Scholar 

  38. Solomon, A. (2014). “Honey Maid and the Business of Love,” The New Yorker. Retrieved from https://www.newyorker.com/business/currency/honey-maid-and-the-business-of-love

  39. Sommer, K. L., & Baumeister, R. F. (2002). Self-evaluation, persistence, and performance following implicit rejection: the role of trait self-esteem. Personality and Social Psychology Bulletin, 28, 926–938. https://doi.org/10.1177/01467202028007006.

    Article  Google Scholar 

  40. Torelli, C. J., & Ahluwalia, R. (2011). Extending culturally symbolic brands: a blessing or a curse? Journal of Consumer Research, 38, 933–947. https://doi.org/10.1086/661081.

    Article  Google Scholar 

  41. Ward, M. K., & Dahl, D. W. (2014). Should the devil sell Prada? Retail rejection increases aspiring consumers’ desire for the brand. Journal of Consumer Research, 41, 590–609. https://doi.org/10.1086/676980.

    Article  Google Scholar 

  42. Weigel, D. J. (2008). The concept of family: an analysis of laypeople’s views of family. Journal of Family Issues, 29, 1426–1447. https://doi.org/10.1177/0192513X08318488.

    Article  Google Scholar 

  43. Whittler, T. E., & DiMeo, J. (1991). Viewers’ reactions to racial cues in advertising stimuli. Journal of Advertising Research, 31, 37–46.

    Google Scholar 

  44. Wootson, C. R. (2017). A Dove ad showed a black woman turning herself white. The backlash is growing. The Washington Post. Retrieved from https://www.washingtonpost.com/news/business/wp/2017/10/08/dove-ad-that-shows-a-black-woman-turning-herself-white-sparks-consumer-backlash/?utm_term=.048c196dbe26

  45. Zmuda, N. (2014). Ad campaigns are finally reflecting diversity of U.S. Advertising Age. Retrieved from http://adage.com/article/news/ad-campaigns-finally-reflects-diversity-u-s/292023/

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Enrica N. Ruggs.

Appendix: Sample Stimuli from Studies 1 and 2

Appendix: Sample Stimuli from Studies 1 and 2

figurea

Study 1 Example Stimuli. On the left, an example of same-sex parents, schema-incongruent stimulus; on the right, an example of heterosexual parents, schema-neutral stimulus.

figureb

Study 2 Example Stimuli. On the left, an example of overweight woman, schema-incongruent stimulus; on the right, an example of thin woman, schema-neutral stimulus.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Ruggs, E.N., Stuart, J.A. & Yang, L.W. The effect of traditionally marginalized groups in advertising on consumer response. Mark Lett 29, 319–335 (2018). https://doi.org/10.1007/s11002-018-9468-3

Download citation

Keywords

  • Diversity
  • Target marketing
  • Advertising
  • Identity
  • Schemas