Abstract The emphasis on business size has become more overt in recent years. However, it is not clear how company size influences consumers’ evaluations. Five experiments investigate the effect of size on consumers’ expectations and evaluations of company behaviors. Consumers expect higher communion from small compared to large companies, and consequently, small relative to large companies garner lower evaluations when they exhibit low communion behaviors. These high communion expectations are driven by the relatively lower marketplace power of small companies. While study 1 provides real-world evidence for the effect of company size on evaluations of company behaviors, studies 2A and 2B demonstrate that perceptions of power underlie the effect of company size on expectations for communion. Studies 3A and 3B indicate that when a company engages in low communion behavior, small relative to large companies garner lower evaluations and this effect is driven by consumers’ perceived violation of expectations. Incorporating additional studies, two meta-analyses conducted with four studies for consumer expectations and six studies for consumer evaluations provide confirmatory evidence in support of our hypotheses. This research demonstrates that how companies are perceived in terms of size and power creates meaning for consumers that drives their expectations and subsequent evaluations. company size, power, expectations, communion, expectancy violations, firm transgressions Went in at a busy time. The girl who took my order was stressed out and barked a harsh, “What can I get you!” It took me a couple of seconds to get over the shock, which only seemed to make her more anxious and upset. —Yelp review for Peet’s Coffee & Tea This Yelp review came from an unhappy customer visiting Peet’s Coffee & Tea in the Boston area, where Peet’s is seen as a small brand, having only six store locations in Massachusetts. Understandably, the customer gave the rude service a one-star Yelp rating. If this customer had had the same experience in San Francisco, where Peet’s is seen as a larger brand—with 183 locations in California—would he have given the same one-star rating or would he have been more forgiving? We argue that, intuitively, it feels very wrong when a small business mistreats and berates its customers, much more so than when a large, multibillion-dollar global corporation does the same. We demonstrate that this sense of “wrongness” occurs because consumers expect small (but not large) companies to be warm and friendly, so any behavior that violates this expectation leads to a strong negative reaction. However, because large companies are not expected to be particularly warm and friendly, behavior that is cold or unfriendly does not result in a similarly strong negative response. More specifically, we propose that because small companies are perceived as possessing low power (especially compared to their larger-sized counterparts) and because low power engenders communal behavior (Conway, Pizzamiglio, and Mount 1996; Rucker, Galinsky, and Dubois 2012), consumers expect small companies to be more communal. Therefore, when a small company exhibits low communion, consumers experience disproportionately greater expectancy violations and punish the small company correspondingly. In turn, these detrimental effects can manifest in lower overall evaluations, purchase intentions, and Yelp star ratings (as we highlight in the example above). However, when it comes to another important determinant of consumers’ evaluations—agency—company size and perceived power no longer influence consumers’ expectations. Because agency is tied to core company attributes valued by consumers (e.g., competence and effectiveness predict product quality), consumers tend to expect high agency from all companies, mitigating any effect company size and power might have in this domain. Therefore, agency-related expectancy violations are not affected by company size, and the evaluation of large and small companies is no different when these companies exhibit low-agency behavior. Our findings add to the growing body of work that studies how consumers interact with and perceive brands in ways similar to how they interact with and perceive people in social situations. This work goes beyond prior research examining the way power shapes expectations for how people behave (Rucker, Hu, and Galinsky 2014) to suggest that perceived power also influences consumers’ expectations of how companies behave. Moreover, instead of studying factors beyond those related to consumers’ prior experiences with individual firms (e.g., brand personality, Aaker, Fournier, and Brasel 2004; brand relationships, Aggarwal 2004), we suggest that consumers can form meaningful expectations based on superficial characteristics (i.e., company size) that guide responses to negative company behaviors even in the absence of prior firsthand experiences. Finally, while prior work has shown that consumers’ personal beliefs and values can shape whether they choose to purchase from small versus large companies (Paharia, Avery, and Keinan 2014; Thompson and Arsel 2004), our research examines how company size directly impacts consumers’ expectations and perceptions of a company’s power. THEORETICAL DEVELOPMENT Although the distinction between small mom-and-pop stores and their much larger competitors has always existed, the emphasis on business size has become much more overt in recent years, making it a salient and significant factor in consumers’ purchase decisions. Taking advantage of this sentiment, American Express created Small Business Saturday, which has established itself as a major shopping holiday (Tozzi 2012). Tax deductions and other financial incentives for small businesses were also central issues in the 2016 US presidential elections. Reflecting this increased focus on company size, recent research has explored the underlying factors driving consumers’ support for small versus large companies. For instance, Thompson and Arsel (2004) found that consumers’ fervent support of small, locally owned coffee shops is often driven by their desire to support the “little guy” and their discomfort with the power wielded by the multinational coffee company Starbucks. Exploring this further, Paharia, Avery, and Keinan (2014) showed that consumers are more likely to support small companies when a large, dominant competitor is made salient, because doing so allows them to express their anti-global-conglomerate views and influence the marketplace. In this research, we continue to explore how the competitive environment in which a company resides influences consumers’ purchase decisions. Rather than explore the reasons why consumers may be motivated to support small versus large companies, we examine how the perceived power disadvantage of small companies compared to their larger competitors might shape consumers’ expectations for how a company should behave. Ironically, the disadvantaged position of small companies that often draws consumer support under normal circumstances, as noted by prior research (Paharia et al. 2014; Thompson and Arsel 2004), may also make small relative to large companies more vulnerable when certain transgressions and failures occur. Next, we discuss prior research on social power and company perceptions to explain why company size leads to different expectations of a company’s communion but not agency, which in turn differentially affect consumer responses to negative firm behaviors. Power and Expectations of Communion/Agency Prior research in the social power and stereotypes literatures suggests that low power is associated with high communion while high power is tied to high agency. Power is generally defined as having the ability to influence others by controlling important resources (Keltner, Gruenfeld, and Anderson 2003). Communion refers to an emotional, interpersonal orientation that considers others’ thinking and needs, while agency reflects an assertive, instrumental orientation associated with effective, decisive behavior aimed at getting things done (Bakan 1966). Prior research notes that low power fosters a communal orientation because the powerless have little control over their own outcomes and lack the ability to act independently and pursue their own goals, so they must pay close attention to the behaviors and needs of others to survive (Galinsky, Gruenfeld, and Magee 2003; Keltner et al. 2003). In contrast, high power fosters agentic behavior because the powerful can remain independent from the influence of others and have the freedom to pursue their own interests (Kipnis 1972; Rucker et al. 2012). Because others’ behaviors are typically attributed to dispositional rather than situational factors (Jones and Harris 1967), the behaviors of low- and high-power groups are attributed to dispositional characteristics such that low- and high-power groups are expected to be communal and agentic, respectively (Conway et al. 1996). Accordingly, for various social stereotypes, the weaker, low-power group is linked to communion while the stronger, high-power group is tied to agency, such as stereotypes for women versus men (Eagly, Wood and Diekman 2000), the poor versus rich (Christopher and Schlenker 2000), and small versus large countries (Poppe and Linssen 1999). We note that different labels have been used to denote the two fundamental content dimensions related to social judgment. For instance, the power literature primarily employs a distinction between “communion” and “agency” (Rucker et al. 2012) while the stereotypes literature prefers to distinguish between “warmth” and “competence” (Fiske et al. 2002). In an editorial, researchers studying these two fundamental dimensions from various perspectives note that the communion/warmth dimension is defined and operationalized very similarly across different research areas, and hence the two labels are often seen as interchangeable (Abele et al. 2008). However, the researchers note that while conceptualizations of both agency and competence include traits related to ability (e.g., effective, organized; Abele et al. 2008), research using the term agency additionally emphasizes an assertive, instrumental orientation aimed at getting things done (e.g., decisive, persistent; Abele and Wojciszke 2007). In this way, the construct of agency covers a wider range of traits, encompassing those related to competence as well as those related to goal achievement (Abele et al. 2008). Our theorizing draws heavily from the power literature that uses the terms communion and agency. Moreover, the focus on instrumentality and getting things done emphasized by prior conceptualizations of agency reflects how many companies behave in the competitive marketplace. Therefore, we choose to use the terms communion and agency rather than warmth and competence in our conceptualization. Expectations of Companies Diverging from prior work focused on people’s perceptions of other people, our research is focused on understanding how perceptions of size and power might influence consumers’ expectations of companies—a context that has not been examined in previous research to the best of our knowledge. We posit that when it comes to the perception of companies, company size and the consequent perceptions of power are likely to shape expectations of the company’s communion. However, consumers may expect all companies, regardless of size and power, to possess similarly high levels of agency. We next explain the reasoning behind our predictions. Expectations of Communion We predict that consumers will come to expect higher levels of communion from small versus large companies because these firms differ significantly in marketplace power. Small firms typically have fewer resources than their larger counterparts and tend to be at a competitive disadvantage in terms of price, product variety, brand awareness, and other factors that drive consumer choice (Paharia et al. 2014). To navigate the competitive environment, small companies must pay close attention to other, more powerful firms (Ailawadi et al. 2010; Shankar 1999). Moreover, to attract and keep customers, small companies must pay close attention to customers’ needs and build brand loyalty through warm, trust-based relationships (Malone and Fiske 2013). In contrast, large companies often have asymmetric control over valued resources (e.g., money, knowledge, supplier relationships), which gives them a disproportionate amount of power and allows them to act with little consideration of others (Scherer and Ross 1990; Schmalensee 1989). Large companies can pressure and demand more of their business partners, as well as aggressively pursue their competitors in the marketplace (Mukherji et al. 2011; Neumark, Zhang, and Ciccarella 2008). We posit that, much like how they develop stereotypes applied to people as discussed earlier, consumers develop the expectation that small companies are more communal than large companies by observing how power shapes the behaviors of small and large companies in the marketplace. As high communion is inherently positive (Abele and Wojciszke 2014; Suitner and Maass 2008), behaviors that violate these expectations will result in outcomes that are more negative than anticipated. Since favorable expectations create frames of reference against which consumers make comparative judgments of a company’s actions, the aversiveness of negative outcomes is enhanced by the unanticipated nature of expectancy-violating behaviors (Mellers et al. 1997). Negative expectancy violations are surprising and disruptive because they signal uncertainty and suggest an inability to predict others’ behaviors and avoid negative outcomes (Bettencourt et al. 1997; Olson, Roese, and Zanna 1996). In consumer-brand relationships, violations of brand personality traits (Aaker et al. 2004) and relationship norms (Aggarwal 2004) can also negatively impact consumers’ purchase intentions and evaluations. Therefore, to the extent that exhibiting low communion is more inconsistent with consumers’ expectations of small versus large companies, we predict that a small company will be penalized more harshly than a large company for engaging in the same low-communion behavior. Furthermore, based on our conceptualization, we believe that a perceived violation of expectations should drive these differential reactions to the low-communion behaviors of small versus large companies. Expectations of Agency Extensive research in psychology finds that people tend to expect high agency from high-power social groups (Conway et al. 1996). Since large companies are often powerful players in the marketplace, one might predict that consumers expect higher agency from large relative to small companies. However, we examine perceptions of companies that are commercial in nature and inherently focused on competing in the marketplace. In this context, we propose that consumers expect high levels of agency from all companies, regardless of size, for three main reasons. First, research on underdog brands suggests that small companies can be associated with high agency. By refusing to be held back by their disadvantaged positions and persevering in the face of daunting obstacles (Paharia et al. 2011), underdogs demonstrate many agentic traits, such as ambition, persistence, and determination. Given that a brand’s underdog status can be inferred from its small size and weak market standing (Hoch and Deighton 1989), consumers may also expect small companies to exhibit high agency. Second, to survive and remain profitable in a highly competitive market, all companies must possess agentic traits, such as competence and efficiency, to effectively deliver quality products that consumers value (Slater and Narver 1994). Put differently, possessing competence and effectiveness (i.e., agentic qualities) can signal to consumers a company’s high-quality product offerings (Aaker et al. 2010; Goldsmith, Lafferty, and Newell 2000). If consumers believe agency is necessary for a company’s survival and profitability in the marketplace, they may also expect that all companies, small and large, possess high levels of agency. Finally, competence, a set of traits that falls under the agency domain, is a stronger predictor of consumer satisfaction and purchase behaviors than warmth and morality, which fall under the communion domain. When a service provider is inefficient or makes mistakes related to agency, having a positive attitude or a smiling face has little effect on customer quality perceptions and satisfaction (Grandey et al. 2005). Moreover, when choosing between service providers, consumers are more likely to choose a competent service provider over a warm or moral one (Kirmani et al. 2017). A company’s perceived competence also has a greater impact on purchase behaviors than its perceived warmth, and this stronger influence of competence on consumer preferences holds for both large, well-known companies (e.g., Mozilla) as well as small, lesser-known companies (e.g., World of Good or GDitty; Aaker et al. 2010). To the extent that agency-related traits are closely tied to the core attributes consumers value in companies, we reason that consumers are likely to take a company’s agency for granted, expecting high agency from all companies. Therefore, consumers may be less likely to differentiate between lower-power small and higher-power large companies in their expectations of agency. Consequently, if consumers have high expectations of agency for both small and large companies, then low-agency behavior will be similarly (un)expected regardless of company size, resulting in comparable consumer responses. In other words, when a company exhibits low agency, size may not influence consumers’ evaluations or purchase intentions. OVERVIEW OF STUDIES Across five studies, we test our proposed conceptualization. In study 1, we examine the implications of size-based expectations for real customer reactions by showing that being perceived as a small relative to a large company can lead to lower Yelp star ratings for low-communion behaviors, but similar ratings for low-agency behaviors. Then we conduct controlled experiments to test our theoretical framework, depicted in figures 1A and 1B. First, we demonstrate that company size influences perceptions of marketplace power, which in turn drive expectations of a company’s communion (figure 1A). However, consumers hold high expectations of agency for both small and large companies, mitigating any effect company size and power may have in this domain. These effects hold regardless of whether consumers consider small versus large companies at the group level (study 2A) or individual firm level (study 2B). Next, we show that low-communion (but not low-agency) behavior from a small company is seen as a greater violation of expectations compared to the same behavior from a large company, resulting in lower overall evaluations of the small relative to large company (figure 1B). This effect holds for both consumers’ brand evaluations (study 3A) and purchase intentions (study 3B). FIGURE 1A View largeDownload slide COMPANY SIZE-BASED EXPECTATIONS DRIVEN BY POWER FIGURE 1A View largeDownload slide COMPANY SIZE-BASED EXPECTATIONS DRIVEN BY POWER STUDY 1: THE EFFECT OF COMPANY SIZE IN THE FIELD To examine whether size-based expectations affect actual consumer responses to low-communion behaviors in the marketplace, we compared the Yelp ratings of Peet’s Coffee & Tea in two markets: Boston, where it is a relatively smaller player, and San Francisco, where it is a relatively larger player. We used the Yelp rating system as a proxy for overall brand evaluation and coded user-written reviews for mentions of communion and agency. We predicted that Yelp reviews mentioning low-communion behavior would have lower corresponding ratings when Peet’s was perceived as relatively small versus large. However, we predicted that perceptions of company size would not affect Yelp ratings for reviews describing low-agency behavior. Method Of Peet’s 239 coffee shops, 183 are located in California, with its remaining stores spread across eight other states, where the number of stores ranges from three to 14. We reasoned that Peet’s would be perceived as a relatively large company (with its stronger store presence) in the California market but a relatively small company in other states. For the purposes of comparison, we focused on two cities: San Francisco, California, where Peet’s has 26 locations, and Boston, Massachusetts, where Peet’s has only six locations.1 Yelp Review Sample We sought to extract Yelp reviews (text and star ratings, with five stars being most favorable) posted over the three-year period from June 1, 2013, to June 8, 2016. Two Boston locations with the lowest total number of reviews (seven and 17) did not have reviews that went as far back as June 2013, so reviews for these locations were not collected. Therefore, four Peet’s locations in the Boston market made up the stores for the small-company condition. For the large-company condition, four locations in the San Francisco market were randomly selected such that these locations had comparable average star ratings and total number of reviews. For these eight Peet’s store locations across the two cities, there were 243 reviews total (113 for Boston, 130 for San Francisco) covering the three-year period, all of which were extracted for this study. The average star rating for our Boston sample of reviews was neutral (M = 3.29, SD = 1.37); 30% were negative (one to two star), 19% were neutral (three star), and 51% were positive (four to five star). The average star rating for our San Francisco sample was also neutral (M = 3.43, SD = 1.37); 26% were negative, 16% were neutral, and 58% were positive. The Boston and San Francisco locations in our sample did not differ in average star rating (F(1, 241) = .62, NS).2 Coding Two independent coders blind to our hypothesis were instructed to determine whether each Yelp review mentioned behavior related to communion and agency, which were defined for them by specific traits drawn from prior research (Abele and Bruckmüller 2011; Abele and Wojciszke 2014). For each review, coders classified the review as mentioning low communion, high communion, or neutral. An analogous classification was done for agency (i.e., low agency, high agency, or neutral; see appendix A for traits used and review excerpts). Other than the review text, no other information was provided to our coders. Intercoder reliability was reasonable (rCommunion = .81, rAgency = .72), and any disagreements were resolved by a third independent coder. For the communion classification, 17% of reviews were classified as low communion, 48% as high communion, and 35% as neutral. For the agency classification, 29% of reviews were classified as low agency, 56% as high agency, and 15% as neutral. Results and Discussion Yelp Ratings We predicted that when Yelp reviewers mentioned low communion, they would give lower ratings if the store was perceived as small (i.e., located in Boston) compared to when it was perceived as large (i.e., located in San Francisco). We did not expect perceived company size to influence Yelp ratings for any other condition. We fully crossed our three independent variables in a regression model: perceived company size (small = 0, large = 1), communion classification (low = –1, neutral = 0, high = 1), and agency classification (low = –1, neutral = 0, high = 1). Yelp rating served as the dependent variable. Since we predicted perceived company size would have an effect only when low communion was mentioned, we anticipated that the size × communion classification interaction would be significant. Analysis revealed significant main effects of communion (b = .84; t(235) = 6.04, p < .0001) and agency classification (b = .76, SE = .11; t(235) = 6.87, p < .0001). Importantly, the key size × communion interaction was significant (b = –.38, SE = .19; t(235) = –2.02, p < .05). No other effects were significant. When reviewers mentioned behavior related to low communion, stores perceived as small received lower Yelp ratings than those perceived as large (t(235) = 2.21, p < .03; see table 1). Perceived company size had no effect on Yelp ratings when behavior related to low agency was mentioned (t(235) = .74, NS) or for any other classification (ps > .25). TABLE 1 STUDY 1: YELP RATING AS A FUNCTION OF COMPANY SIZE AND BEHAVIOR Low communion Neutral High communion Low agency Neutral High agency Small company (Boston) 1.88 (.75) 2.99 (1.17) 3.64 (.94) 2.13 (.98) 2.68 (1.44) 3.70 (.82) Large company (San Francisco) 2.58 (1.26) 3.02 (1.41) 3.52 (1.11) 2.30 (1.18) 3.08 (1.39) 3.74 (1.04) Low communion Neutral High communion Low agency Neutral High agency Small company (Boston) 1.88 (.75) 2.99 (1.17) 3.64 (.94) 2.13 (.98) 2.68 (1.44) 3.70 (.82) Large company (San Francisco) 2.58 (1.26) 3.02 (1.41) 3.52 (1.11) 2.30 (1.18) 3.08 (1.39) 3.74 (1.04) Note.— Standard deviations indicated in parentheses. TABLE 1 STUDY 1: YELP RATING AS A FUNCTION OF COMPANY SIZE AND BEHAVIOR Low communion Neutral High communion Low agency Neutral High agency Small company (Boston) 1.88 (.75) 2.99 (1.17) 3.64 (.94) 2.13 (.98) 2.68 (1.44) 3.70 (.82) Large company (San Francisco) 2.58 (1.26) 3.02 (1.41) 3.52 (1.11) 2.30 (1.18) 3.08 (1.39) 3.74 (1.04) Low communion Neutral High communion Low agency Neutral High agency Small company (Boston) 1.88 (.75) 2.99 (1.17) 3.64 (.94) 2.13 (.98) 2.68 (1.44) 3.70 (.82) Large company (San Francisco) 2.58 (1.26) 3.02 (1.41) 3.52 (1.11) 2.30 (1.18) 3.08 (1.39) 3.74 (1.04) Note.— Standard deviations indicated in parentheses. Discussion The results of this study provide initial support for our conceptual model, with Yelp ratings as a measure of customer evaluations and naturalistic manipulations of company size and behavior. Only when customer reviews indicated that Peet’s exhibited low communion did company size have an effect: customers provided lower ratings when Peet’s was perceived as small versus large. Since Peet’s was founded in California, arguably it may be a more beloved brand in San Francisco than in Boston, making brand strength and familiarity potential confounds in this field study. However, this alternative explanation would predict overall higher Yelp ratings for San Francisco (large) versus Boston (small) locations, regardless of behavior type or valence; however, we did not observe a main effect of perceived company size. In any case, we conducted a number of controlled experiments to provide direct evidence for our proposed conceptualization and the underlying process. STUDY 2A: COMMUNION AND AGENCY EXPECTATIONS Study 2A was designed to demonstrate that company size drives perceptions of power, which in turn influence consumers’ expectations, but only in the communion domain. We had participants rate the extent to which communion and agency traits were typical of small (or large) companies. In addition, to provide a benchmark for participants’ communion and agency expectations, we also asked the same participants to rate the extent to which the same communal and agentic traits were typical of companies in general. Doing so also allowed us to determine whether consumers had especially high expectations of communion from small companies, or especially low expectations of communion from large companies, or both. Method Amazon Mechanical Turk (MTurk) workers residing in the United States (N = 198, Mage = 38.08, 44% female) were randomly assigned to answer questions regarding their opinions of small or large companies (sample size was preset at approximately 200). They were presented with traits related to low communion (cold, unfriendly, dishonest; α = .96), high communion (ethical, kind, understanding; α = .91), low agency (incompetent, ineffective, inefficient; α = .93), and high agency (capable, decisive, organized; α = .80; Abele and Bruckmüller 2011; Abele et al. 2008). Participants were asked “how typical is it that small (large) companies possess each of these characteristics” (1 = very atypical, 9 = very typical). All participants, regardless of size condition, also provided typicality ratings for “companies in general” for the same traits. The order in which small (large) companies and companies in general were rated was counterbalanced. Following prior conceptualizations of power (Keltner et al. 2003; Rucker et al. 2012), we developed a six-item power score to tap into perceptions of a company’s ability to control others’ outcomes and to behave independently in the marketplace (α = .91; see appendix B). The order in which the trait and power measures were completed was counterbalanced. At the end of this and all studies, participants reported their gender and age, which had no significant effects on any of our results and will not be discussed further; no participants were dropped from this or any other analysis. Results and Discussion Factor Analysis We conducted exploratory factor analysis with varimax rotation on all dependent variables, revealing five factors with eigenvalues greater than one, which consisted of low communion, high communion, low agency, high agency, and power (see web appendix A for factor loadings). Given this factor pattern, we analyzed ratings of low and high communion and agency separately. Companies in General We first examined participants’ ratings of companies in general to gauge baseline expectations of communion and agency. Although all participants rated companies in general regardless of size condition, we included company size as a factor to ensure participants rating small versus large companies did not rate companies in general differently. Therefore, we conducted a 2 (company size condition: small vs. large, between-subjects) × 2 (trait dimension: communion vs. agency, within-subjects) × 2 (trait level: low vs. high, within-subjects) mixed design ANOVA that revealed only a main effect of trait level (F(1, 588) = 84.05, p < .0001) and a trait dimension × trait level interaction (F(1, 588) = 143.78, p < .0001; see figure 2). Participants rated high-communion traits as less typical than low-communion traits (MHigh = 4.73, SD = 1.57 vs. MLow = 5.10, SD = 1.92; F(1, 588) = –2.00, p < .05). While ratings of high communion were lower than the scale midpoint of five (t(197) = –2.36, p < .05), ratings of low communion were not significantly different (t(197) = .74, NS). In the agency domain, participants rated high-agency traits as more typical than low-agency traits (MHigh = 6.40, SD = 1.52 vs. MLow = 3.68, SD = 1.65; F(1, 588) = 14.96, p < .0001). Notably, ratings of high agency were above (t(197) = 12.99, p < .0001), while ratings of low agency were below, the scale midpoint (t(197) = –11.30, p < .0001). These results suggest that participants have moderate expectations of communion but high expectations of agency for companies in general. FIGURE 1B View largeDownload slide CONSUMER REACTIONS TO COMPANY TRANSGRESSIONS DRIVEN BY EXPECTANCY VIOLATIONS FIGURE 1B View largeDownload slide CONSUMER REACTIONS TO COMPANY TRANSGRESSIONS DRIVEN BY EXPECTANCY VIOLATIONS Small versus Large Companies The same mixed-design ANOVA on ratings for small versus large companies revealed the following effects: company size (F(1, 588) = 3.44, p = .06), trait level (F(1, 588) = 269.32, p < .0001), company size × trait dimension (F(1, 588) = 98.98, p < .0001), and trait dimension × trait level (F(1, 588) = 39.27, p < .0001). These effects were qualified by the significant three-way interaction (F(1, 744) = 137.26, p < .0001; see figure 3). Supporting our prediction that small companies are expected to be more communal than large companies, high-communion trait ratings were higher (MSmall = 6.69, SD = 1.58 vs. MLarge = 4.38, SD = 1.51; F(1, 588) = 85.56, p < .0001) and low-communion trait ratings were lower for small versus large companies (MSmall = 2.83, SD = 1.50 vs. MLarge = 5.70, SD = 2.08; F(1, 588) = 131.56, p < .0001). Moreover, small companies received above-midpoint ratings for high communion (t(197) = 10.64, p < .0001) but below-midpoint ratings for low communion (t(197) = –14.38, p < .0001). Large companies received the opposite, with below-midpoint ratings for high communion (t(197) = –4.09, p < .0001) and above-midpoint ratings for low communion (t(197) = 3.34, p < .01). In contrast, small and large companies did not receive different high-agency trait ratings (MSmall = 6.25, SD = 1.26 vs. MLarge = 6.54, SD = 1.52; F(1, 588) = 1.35, NS) or low-agency trait ratings (MSmall = 3.53, SD =1.51 vs. MLarge = 3.61, SD = 2.03; F(1, 588) = .12, NS). Mirroring ratings for companies in general, both small and large companies received above-midpoint ratings for high agency (small: t(197) = 9.89, p < .0001; large: t(197) = 10.10, p < .0001) and below-midpoint ratings for low agency (small: t(197) = –9.74, p < .0001; large: t(197) = –6.83, p < .0001). In sum, participants expected high communion from small companies and moderately low communion from large companies, but expected high agency from both small and large companies. FIGURE 2 View largeDownload slide STUDY 2A: TRAIT TYPICALITY OF COMPANIES IN GENERAL AS A FUNCTION OF COMPANY SIZE CONDITION, TRAIT DIMENSION, AND TRAIT LEVEL NOTE.— THE DOTTED LINE DENOTES THE SCALE MIDPOINT OF 5. FIGURE 2 View largeDownload slide STUDY 2A: TRAIT TYPICALITY OF COMPANIES IN GENERAL AS A FUNCTION OF COMPANY SIZE CONDITION, TRAIT DIMENSION, AND TRAIT LEVEL NOTE.— THE DOTTED LINE DENOTES THE SCALE MIDPOINT OF 5. Small and Large Companies versus Companies in General Next, we determined whether expectations of small and large companies differed from those of companies in general. Relative to companies in general, small companies were rated higher on high communion (MDifference = 2.02, SD = 1.93; t(98) = 10.38, p < .0001) and lower on low communion (MDifference = –2.25, SD = 2.02; t(98) = –11.07, p < .0001), confirming that small companies are strongly expected to exhibit high communion and also strongly expected not to exhibit low communion. In contrast, relative to companies in general, large companies were rated lower on high communion (MDifference = –.42, SD = 1.22; t(98) = –3.44, p < .001) and higher on low communion (MDifference = .59, SD = 1.66; t(98) = 3.51, p < .001), suggesting that there are some expectations for large companies to exhibit moderately low levels of communion. Analysis of agency ratings indicated that expectations of both small and large companies mapped closely to those of companies in general. Neither small (high agency: MDifference = –.14, SD = 1.74; t(98) = –.83, NS; low agency: MDifference = –.14, SD = 1.82; t(98) = –.77, NS) nor large companies (high agency: MDifference = .13, SD = 1.04; t(98) = 1.29, NS; low agency: MDifference = –.07, SD = 1.37; t(98) = –.51, NS) were rated differently on low- or high-agency traits compared to companies in general. To compare small versus large companies, we applied the same mixed ANOVA as above to relative trait ratings (which we calculated by subtracting ratings for companies in general from ratings for small or large companies). Lower-order effects were qualified by the significant three-way interaction (F(1, 588) = 121.11, p < .0001, see figure 4). Relative high-communion trait ratings were higher (F(1, 588) = 96.04, p < .0001) and relative low-communion trait ratings were lower for small versus large companies (F(1, 588) = 129.96, p < .0001). Small and large companies did not receive different levels of relative high- or low-agency trait ratings (Fs < 1.25, ps > .20). FIGURE 3 View largeDownload slide STUDY 2A: TRAIT TYPICALITY AS A FUNCTION OF COMPANY SIZE, TRAIT DIMENSION, AND TRAIT LEVEL NOTE.— THE DOTTED LINE DENOTES THE SCALE MIDPOINT OF 5. NOTE.— THE DOTTED LINE DENOTES THE SCALE MIDPOINT OF 5. FIGURE 3 View largeDownload slide STUDY 2A: TRAIT TYPICALITY AS A FUNCTION OF COMPANY SIZE, TRAIT DIMENSION, AND TRAIT LEVEL NOTE.— THE DOTTED LINE DENOTES THE SCALE MIDPOINT OF 5. NOTE.— THE DOTTED LINE DENOTES THE SCALE MIDPOINT OF 5. Mediation by Power Next, we examined whether perceptions of power accounted for consumers’ higher expectations of communion for small companies as well as their lower expectations of communion for large companies. A one-way ANOVA on power confirmed that small relative to large companies were perceived as possessing less power (MSmall = 4.56, SD = 1.34 vs. MLarge = 7.22, SD = 1.17; F(1, 249) = 220.90, p < .0001). Consistent with our predictions, mediation analysis revealed that power mediated the effect of company size (small = 0, large = 1) on relative ratings of high communion (B = –.62 CI95: –1.19, –.11) and low communion (B = 1.18, CI95: .53, 1.86) such that the lower power attributed to small (vs. large) companies led to higher relative ratings of high communion and lower relative ratings of low communion (model 4; Preacher, Rucker, and Hayes 2007). We found no mediation by power for relative ratings of high agency (B = .14, CI95: –.40, .68) or low agency (B = –.15, CI95: –.66, .40). Supporting our conceptualization, study 2A shows that company size influences perceptions of power, which in turn drive consumers’ expectations of communion but not agency. Follow-up Study 2B We conducted a follow-up study 2B to demonstrate that these size-based expectations also held for consumers’ assessments of a specific small or large company rather than for just the abstract conceptualization of small or large companies as was operationalized in study 2A (see web appendix B for full description). The focal company Lee & Sons (a luggage company) was made to appear much smaller or larger in comparison to its competitors. We also included a control condition in which Lee & Sons appeared similar in size relative to its competitors. Furthermore, for this study, we employed an expanded list of items to assess power, communion, and agency. Results showed that, as expected and consistent with the results of study 2A, Lee & Sons was rated higher on communion when it was perceived as small versus large. Furthermore, compared to the control condition, being perceived as small increased expectations of communion, while being perceived as large decreased expectations of communion. Mediation analysis found that perceptions of power drove the extent to which the small-company condition garnered higher communion ratings than both the large-company and control conditions. In contrast, Lee & Sons was rated equally high on agency regardless of size perceptions. Discussion Studies 2A and 2B demonstrate that when consumers consider companies both at the group and individual levels, perceptions of company size drive perceptions of power, which in turn shape expectations of communion. In both studies, expectations for the high communion of small companies appeared to be the strongest, which suggests that consumers would experience the greatest expectancy violations and respond the most negatively when a small company exhibits low communion. These studies also suggest that expectations of agency for both small and large companies are fairly high, and independent of company size. Therefore, when a company exhibits low agency, size should not affect consumers’ perceived expectancy violations and their subsequent responses. In our remaining studies, we test these implications directly by exploring the consequences of companies exhibiting low-communion and low-agency behavior and the mediating role of consumers’ expectancy violations. STUDY 3A: EVALUATIONS OF LOW COMMUNION AND LOW AGENCY Having established that consumers expect high communion from small but not large companies, we investigate how consumers respond when a small versus large company violates these expectations. We predict that when a company exhibits low-communion behavior, the company will be penalized more severely if it is perceived as small relative to large. Because consumers hold different size-based expectations only in the communion domain, we do not anticipate a size effect when a company exhibits low-agency behavior. In study 3A, we manipulated relative size perceptions and examined participants’ evaluations of a hypothetical pet food retailer, PawStop, that received customer reviews indicating either low-communion or low-agency behavior. Moreover, this study tested whether consumers’ more negative responses to the low-communion behaviors of a small company were driven by expectancy violations. Method We recruited 199 MTurk workers residing in the United States (sample size preset at approximately 200; Mage = 36.32, 47% female). The study used a 2 (company size: small vs. large) × 2(behavior: low communion vs. low agency) between-subjects design. Participants were told that they would be assigned to evaluate one of three different pet food retailers, but all participants evaluated PawStop. Size Manipulation We sought to change perceptions of relative size rather than create absolute differences across conditions. According to prior research, features distinguishing the evaluation target from others (e.g., a woman in an otherwise all-male group) are likely to be used as a basis for categorization, and once categorization has occurred, relevant expectations become activated and can influence subsequent judgments (Fiske and Neuberg 1990; Pettigrew and Martin 1987; Taylor et al. 1978). Using this insight, we had participants read business profiles of three pet food retailers, which included numerous details about each company’s philosophy and products. While all participants were informed that PawStop had a revenue of $3.1 million, net income of $800,000, and 60 employees, these statistics for the other two companies (Chewy and EntirelyPets) were changed such that PawStop appeared larger or smaller in comparison (Paharia et al. 2014). In the large-company condition, the other two companies were presented as being much smaller (e.g., $505,000 in revenue, $99,000 in net income, four employees), whereas in the small-company condition, the other two firms were presented as being much larger (e.g., $505 million in revenue, $3.9 million in net income, 200 employees). In this way, we manipulated the perceived size of PawStop while exposing participants across all conditions to the same information regarding this focal company. Web appendix C provides details on stimuli and results of a pretest showing that the size manipulation was successful and activated higher expectations of communion but not agency in the small- versus large-company conditions. Communion/Agency Manipulation Incorporating the common consumer practice of consulting online reviews to help make purchase decisions (Kupor and Tormala 2018), we presented participants with quantitative customer ratings ranging from one to five stars (five stars = most favorable). Filler ratings on purchase policies were identically high across all conditions. The key manipulation was contained in the low customer service ratings. In the low-communion condition, PawStop received low ratings on friendliness, sincerity, and helpfulness, while in the low-agency condition, PawStop received low ratings on effectiveness, competence, and organization (see appendix C for stimuli and web appendix C for a pretest). Measures Participants completed four items assessing their evaluations of PawStop by indicating how likeable, positive, good, and appealing PawStop was (e.g., 1 = very bad, 9 = very good; α = .96). Then they completed a measure of expectancy violations by rating how much the company’s behavior was expected, was typical, and fit their beliefs (e.g., 1 = very atypical, 9 = very typical; α = .90; Bettencourt et al. 1997). For ease of interpretation, these items were reverse-scored such that higher values reflected greater expectancy violations. Participants also completed a size manipulation check (1 = very small in size, 9 = very large in size) and familiarity ratings (1 = not at all familiar, 9 = very familiar) for their perceptions of PawStop.3 Results and Discussion Checks For size ratings, there was only the expected main effect of company size (MSmall = 3.56, SD = 1.79 vs. MLarge = 5.58, SD = 2.05; F(1, 195) = 54.41, p < .05). For familiarity ratings, there were no significant effects (Fs < .27); consistent with PawStop being a hypothetical company, familiarity ratings were low (M = 1.82, SD = 1.71). Brand Evaluations We tested our prediction that a small relative to large company would receive lower evaluations for exhibiting low communion but not low agency. We conducted a 2 (company size: small vs. large) × 2(behavior: low communion vs. low agency) ANOVA on participants’ brand evaluations. There were no main effects of company size (F(1, 195) = .92, NS) or rating (F(1, 195) = .18, NS) but a significant two-way interaction (F(1, 195) = 3.89, p < .05; see figure 5). As predicted, when PawStop exhibited low communion, brand evaluations were lower when PawStop was perceived as small versus large (MSmall = 3.66, SD = 1.66 vs. MLarge = 4.38, SD = 1.80; F(1, 195) = 4.19, p < .05). However, when PawStop exhibited low agency, brand evaluations were no longer affected by company size (MSmall = 4.25, SD = 1.89 vs. MLarge = 4.00, SD = 1.55; F(1, 195) = .53, NS). Moreover, when PawStop was perceived as small, the company garnered marginally lower evaluations for exhibiting low communion versus agency (F(1, 195) = 2.87, p = .09); however, when PawStop was perceived as large, evaluations were not significantly different (F(1, 195) = 1.20, p = .27). FIGURE 4 View largeDownload slide STUDY 2A – RELATIVE TRAIT TYPICALITY AS A FUNCTION OF COMPANY SIZE, TRAIT DIMENSION, AND TRAIT LEVEL FIGURE 4 View largeDownload slide STUDY 2A – RELATIVE TRAIT TYPICALITY AS A FUNCTION OF COMPANY SIZE, TRAIT DIMENSION, AND TRAIT LEVEL FIGURE 5 View largeDownload slide STUDY 3A – BRAND EVALUATIONS AND EXPECTANCY VIOLATIONS AS A FUNCTION OF COMPANY SIZE AND BEHAVIOR FIGURE 5 View largeDownload slide STUDY 3A – BRAND EVALUATIONS AND EXPECTANCY VIOLATIONS AS A FUNCTION OF COMPANY SIZE AND BEHAVIOR Mediating Role of Expectancy Violations We predicted that expectancy violations would mediate the effect of company size on evaluations when PawStop exhibited low communion but not low agency. We first conducted the same ANOVA as above to test whether exhibiting low communion elicited stronger expectancy violations when the company was small relative to large. There were no main effects of company size (F(1, 195) = 1.84, NS) or behavior (F(1, 195) = 1.07, NS) but a significant two-way interaction (F(1, 195) = 8.91, p < .01; see figure 5). When PawStop exhibited low communion, participants indicated greater expectancy violations when it was perceived as small versus large (MSmall = 6.38, SD = 1.54 vs. MLarge = 5.35, SD = 1.58; F(1, 195) = 9.81, p < .01). However, when PawStop exhibited low agency, there was no longer a company size effect (MSmall = 5.43, SD = 1.76 vs. MLarge = 5.81, SD = 1.73; F(1, 195) = 1.36, NS). Moreover, when PawStop was perceived as small, participants indicated greater expectancy violations when the firm exhibited low communion versus low agency (F(1, 195) = 8.02, p < .01); when the firm was perceived as large, expectancy violations were not significantly different for low-communion versus low-agency behavior (F(1, 195) = 1.92, p = .17). Next, showing discriminant validity between our two measures, an exploratory factor analysis with varimax rotation yielded two factors that loaded separately on brand evaluations and expectancy violations (see web appendix C). Then we conducted moderated mediation analysis where company size served as the independent variable (small = 0, large = 1), expectancy violations as the mediator, behavior type as the moderator (low communion = 0, low agency = 1), and brand evaluations as the dependent variable. Bootstrap analysis (model 8; Preacher et al. 2007) found a significant positive indirect effect of company size on brand evaluations through expectancy violations but only when PawStop exhibited low communion (B = .58, 95% CI: .23, 1.01) and not when it exhibited low agency (B = –.22, 95% CI: –.64, .15). The index of moderated mediation was also significant (95% CI: –1.43, –.26), indicating that the indirect effect of company size was significantly different in the two behavior type conditions. Follow-up Study 3B To ensure that this pattern of effects was not specific to the product category or manipulations used in study 3A, we tested whether these results held for a different sample (undergraduate students), product category (chocolate), behavior manipulation (qualitative reviews), and dependent variable (purchase intentions; see web appendix D for full description). We used the same relative size manipulation to introduce participants to the focal company, Dagoba Chocolate. Participants read customer reviews modeled off Amazon.com, which included a qualitative assessment of the company in addition to a quantitative star rating. Dagoba was described as exhibiting low communion (disregarding customer needs) or low agency (having inconsistent quality). As predicted, purchase intentions were lower when Dagoba was perceived as small versus large but only when Dagoba exhibited low communion and not when it exhibited low agency. Again, expectancy violations mediated the effect of company size on purchase intentions. Therefore, this additional study replicated the negative effect that low-communion behavior has on consumers’ evaluations of a small versus large company. Discussion Consistent with our predictions, we find that exhibiting low communion elicits stronger expectancy violations and lower evaluations for a small relative to a large company. However, when a company exhibits low-agency behavior, company size no longer affects expectancy violations or brand evaluations. By using a relative manipulation of company size, we held constant the information related to the focal company, indicating that it is the perception of company size (rather than absolute size) that drives our effects. We also manipulated perceptions of communion and agency using customer reviews that approximated real consumer experiences, enhancing the external validity of our studies. Importantly, by showing that the higher levels of expectancy violations elicited by the low communion of the small relative to large company drove participants’ evaluations, these findings further demonstrate that consumers’ expectations of high communion from small relative to large companies have important consumer and marketplace consequences. SINGLE-PAPER META-ANALYSES To attain a more precise estimate of our effects, we performed a series of single-paper meta-analyses (SPMs; McShane and Bockenholt 2017). The first SPM included four studies (studies 2A, 2B, and two additional studies) and examined whether consumers expected higher levels of communion but equal levels of agency from small versus large companies, as is proposed by the model in figure 1A. Significant differences were found across small versus large companies for ratings on communion traits (d = 1.31, SE = .30, 95% CI: .72, 1.90) but not agency traits (d = –.20, SE = .30, 95% CI: –.78, .38). Condition estimates confirmed that small relative to large companies were rated higher on communion (MSmall = 7.05 vs. MLarge = 5.74) but equally high on agency (MSmall = 6.70 vs. MLarge = 6.90). The second SPM examined six studies (studies 3A and 3B, plus four additional studies). We tested whether low-communion but not low-agency behavior garnered lower evaluations when a company was perceived as small versus large, as is illustrated in figure 1B. When companies engaged in low-communion behavior, significant differences were found between ratings of small versus large companies on company evaluations (d = –.72, SE = .14, 95% CI: –1.00, –.44), such that small companies received lower evaluations than large companies (MSmall = 3.80 vs. MLarge = 4.52). However, when companies exhibited low agency, the effects of company size were not significant (d = .27, SE = .15, 95% CI: –.02, .56), though large companies did directionally receive lower evaluations than small companies (MSmall = 4.40 vs. MLarge = 4.13). Web appendix E provides details on the studies used in these SPMs and the results. GENERAL DISCUSSION We demonstrate that small relative to large companies garner lower evaluations for engaging in low-communion behaviors because consumers expect small companies to be more communal than large companies. While consumers’ higher expectations for small versus large companies in the communion domain stem from perceived power differences, these same power differences do not result in differential expectations in the agency domain. Therefore, when a company is perceived as small versus large, exhibiting low communion garners lower Yelp ratings, less favorable evaluations, and lower purchase intentions. Given the weaker effect of company size on expectations of agency, no significant effects of size emerge on consumer evaluations when a company exhibits low agency. These results are robust across various product categories (i.e., coffee, pet supplies, and chocolate) and levels of brand familiarity (i.e., well-known Peet’s, lesser-known Dagoba, hypothetical PawStop). Moreover, our studies examine company behaviors by measuring or varying the extent to which online customer reviews mention low-communion or low-agency behavior, a context closely aligned with real customer experiences. Further highlighting the influence of consumer expectations about company traits, heightened expectancy violations elicited by the low communion of a small company drive the disproportionately negative responses from consumers. Broadly speaking, this research contributes to the literature employing the “brand-as-person” metaphor to develop conceptual tools that illuminate the nature of consumer-brand interactions (Aggarwal 2004; Fournier 1998). While past research has identified and delineated the content of consumers’ expectations for different types of companies (Aaker et al. 2010; Kervyn et al. 2012), our research explains why consumers develop these expectations and explores how consumers respond when these expectations are violated. In this research, we explore when expectations about companies develop in a manner like those for people and when these expectations diverge. Aligning with the social perceptions of people in the communion domain, expectations are based on how much power consumers believe companies possess, such that the lower perceived power of small versus large companies drives higher expectations of communion. Diverging from the social perceptions of people in the agency domain, the effects of company size and perceived power are mitigated because consumers generally take a company’s high agency for granted such that large versus small companies are expected to be equally agentic. Our findings also speak to research examining how marketplace dynamics affect consumers’ support for individual companies. Prior research has looked to consumers’ personal beliefs and demonstrated that support for a small versus large company can be driven by alignment between the symbolic meaning of company size and consumers’ personal values. For instance, consumers support small, underdog brands because they personally identify with the underdog narrative (Paharia et al. 2011), while consumers support small, local coffee shops to reinforce their anticommercialization beliefs (Thompson and Arsel 2004). Diverging from these perspectives, we suggest that consumer support for small companies can also be guided by expectations for how small companies ought to behave, regardless of consumers’ personal beliefs. We find that while perceiving small companies as low in power leads consumers to believe small companies possess higher levels of communion, this also makes these companies especially vulnerable to consumer backlash when they violate consumers’ expectations by exhibiting low communion. We also contribute to the literature examining consumer reactions to firm transgressions. Prior research has primarily focused on how consumers’ one-to-one interactions and connections with a brand or company shape their reactions to firm transgressions. For instance, product failures elicit stronger consumer backlash for brands with sincere versus exciting brand personalities (Aaker et al. 2004), while poor product performance accompanied by polite and respectful interactions garners less negative reactions for brands fostering communal versus exchange relationships with consumers (Aggarwal and Larrick 2012). Although we also take an interpersonal approach to understanding consumers’ reactions, our perspective has an advantage in that an ongoing consumer-brand relationship is not assumed or required. In fact, we find that our effects occur even when consumers have had no prior interactions with a hypothetical company. Here, we find that consumers rely on seemingly superficial but easily observable characteristics like company size to inform their expectations, which meaningfully shape their reactions to negative firm behaviors. Finally, while company size is readily observed and measured, our research highlights how perceptions of company size can also be easily shaped through comparisons between a focal company and salient competitors in the marketplace, providing practitioners with powerful insights and marketplace nudges. Given that perceptions of size drive expectations of company traits (i.e., communion), companies may have more control over consumers’ expectations than previously thought. For instance, a company may alleviate the disadvantage of being small when low-communion transgressions occur by strategically positioning itself relative to much smaller competitors to appear larger and alter consumer expectations. For instance, Samuel Adams is one of the largest players in the US beer market but often positions itself as being a small brewery relative to competitors like Miller and Budweiser. If Samuel Adams is faced with bad publicity regarding communion, it may face backlash from consumers. To preempt this, Samuel Adams could position itself relative to much smaller craft breweries to appear larger in size, thereby reshaping consumers’ expectations. We also uncover an uncanny advantage for companies perceived as large and powerful: the ability to engage in low communion without incurring extra punishment (relative to smaller competitors). As large companies like Amazon enter new product categories, they may be more resilient in the face of potential brand failings related to communion. Even though consumers may perceive Amazon as less warm and friendly than its smaller competitors, they may punish Amazon less than its smaller competitors when the inevitable low-communion transgression occurs. Our research also raises interesting questions for future work. We find that consumers’ expectations for the high communion of small companies seem to be the strongest and most extreme. Expectations of communion for small companies deviated the greatest from those of companies in general, and the low-communion behavior of small companies elicited the greatest expectancy violations when we compared across all conditions. While our conceptualization does not directly address why the communal expectations for small companies are so strong, the social power literature may provide insights into why this might have occurred. Since people are less motivated to form accurate and individuated impressions of low- relative to high-powered individuals (Dépret and Fiske 1999; Fiske and Neuberg 1990), consumers may rely more heavily on their size-based expectations when forming impressions of low-powered small companies because doing so requires fewer cognitive resources. If this is the case, then the communal expectations tied to small companies may become stronger and more polarized as consumers repeatedly rely on their expectations, rather than individuating information, when forming their impressions of small companies (Fiske and Neuberg 1990). We also note that while our findings suggest that consumers strongly associate small firms with communion, there may be exceptions. For instance, startups are well known for their cutthroat, aggressive culture (Manjoo 2017), so consumers may come to expect startups, despite their small size, to be low in communion. In the context of social stereotypes, people often reconcile the unexpected behavior of a subset of group members by creating a subtype that acknowledges the exception to the rule but also preserves the overarching stereotype (Kunda and Oleson 1995). For example, while women are seen as high in communion, professionally ambitious women are seen as low in communion (Rudman and Glick 1999). If consumers’ expectations of small companies are also hierarchical in nature, then small startups may constitute a subtype of small companies that are expected to be low on communion but exceptionally high on agency. Future research could also explore whether these expectations protect startups from being penalized disproportionately for low-communion behavior, but the “protective” ability of low-communion expectations may be limited to more moderate transgressions like the ones examined in our studies. In conclusion, a growing body of research supports the notion that consumers perceive brands and companies in ways that both parallel and diverge from their perceptions of people and social groups (Aaker et al. 2010; Kervyn et al. 2012). This research demonstrates that perceptions of a company’s relative size influence perceptions of its power, which in turn drive consumers’ expectations and subsequent preferences. In doing so, this research enriches our understanding of what drive consumers’ expectations as well as how they react to expectancy-violating behaviors. DATA COLLECTION INFORMATION Data for study 1 (Yelp reviews and star ratings) were collected by a research assistant in June 2016 under the supervision of the first author. Data for study 2A (March 2018), study 2B (May 2015), the study 2B pretest (January 2015), study 3A (December 2017), the study 3A pretests (February 2018), and the study 3B pretests (April 2016) were collected online via MTurk by the first author. Data for study 3B (April 2016), the Thrive Home Furnishings study (September 2015), the PawStop study (April 2014), and the Sherwood Coffee Co. study (March 2012) were collected at the University of North Carolina Charlotte by research assistants under the supervision of the first author. Data for the Casemate study (June 2016) were collected in the Management Department Behavioral Lab at the University of Toronto by research assistants under the supervision of the second author. The first author analyzed all the data in consultation with the second author. The authors thank Jim Bettman, Mary Frances Luce, and Ann McGill for their helpful comments on previous versions of this article and Amber Hill for her assistance with data entry and coding. The financial support from the Social Sciences and Humanities Research Council of Canada is also gratefully acknowledged. Supplementary materials are included in the web appendix accompanying the online version of this article. Appendix A: Study 1—Traits and Review Excerpts Communion traits: ethical, friendly, honest, kind, understanding, warm Agency traits: competitive, decisive, effective, efficient, forceful, organized Low communion “i get it they were busy but that’s no excuse for treating customer in a bad way. This girl (white girl maybe 5.2 tall) don’t know her name but she was the cashier she talked very rude, she was rolling her eyes and was annoyed by what i was asking.” Low agency “Came in here around 6:30. Ordered a coffee-free drink with almond milk. They made my drink with coffee not once, but twice. I threw the drink in the garbage and walked out the door. Never coming back. The employees here need to learn their menu.” Low communion and low agency “Coffee is ok. But the barista is super rude. Same girl, yelling at other customers & making the wrong drink. Twice for me. Not returning for sure.” High communion “The staff here are always energetic and happy. They recognize faces and orders after a while, and there is very little turnover.” High agency “Quick and knowledgeable service! Best coffee in the city and love my morning stops on my way to work.” High communion and High agency “Great flavorful coffee (yes better flavor than Starbucks and Dunkin). Friendly helpful staff. Love coming here.” Neutral communion and agency “I got my standard cup of black coffee. I wanted a blonde roast, but the barista told me Peet’s only does bolder dark roasts. I settled for it and found it to be satisfactory. I wouldn’t say Peet’s is better than Starbucks, just different. It’s just a matter of taste.” Appendix B: Study 2A—Power Measure All items were answered via nine-point scales with 1= not at all, 9 = very much. How much influence do [small/large] companies have in the marketplace? How likely is it for [small/large] companies to become leaders in the marketplace? To what extent can [small/large] companies determine what their competitors do (e.g., what product to launch, at what price, etc.)? To what extent can [small/large] companies behave the same way regardless of what competitors are doing? To what extent can [small/large] companies choose and pursue their own goals and interests? To what extent can [small/large] companies make others (e.g., suppliers, customers, etc.) do things they would not otherwise do? Appendix C: Study 3A—Behavior Manipulation Footnotes 1 All Peet’s store locations in Massachusetts are in Boston. A pretest using Amazon Mechanical Turk (N = 86, Mage = 35.86, 42% female) confirmed that consumers living in California perceived Peet’s as a larger company (1 = very small in size and 9 = very large in size) relative to consumers living in Massachusetts (MCA = 5.77, SD = 1.91, MMA = 4.95, SD = 1.72; F(1, 83) = 3.96, p < .05). 2 Across all 32 Peet’s store locations (i.e., the total population in our two chosen cities: six stores in Boston and 26 in Los Angeles), there was also no difference in average star ratings for the San Francisco (M = 3.76, SD = .34) versus Boston locations (M = 3.75, SD = .38; F(1, 30) = .01, NS). 3 An attention check asked, “In the customer service ratings for PawStop, which attribute received the lowest rating?” with choices 1) Return Policies, 2) Customer Service Sincerity, 3) Customer Service Competence, 4) Sales and Coupons. This attention check was failed by 17% of participants; however, omitting participants who failed the check did not alter the pattern of our results, so we retained all participants in our analyses. References Aaker Jennifer L. , Fournier Susan , Brasel S. Adam ( 2004 ), “ When Good Brands Do Bad ,” Journal of Consumer Research , 31 ( 1 ), 1 – 16 . Aaker Jennifer L. , Vohs Kathleen D. , Mogilner Cassie ( 2010 ), “ Nonprofits Are Seen as Warm and For-Profits as Competent: Firm Stereotypes Matter ,” Journal of Consumer Research , 37 ( 2 ), 224 – 37 . Abele Andrea E. , Bruckmüller Susanne ( 2011 ), “ The Bigger One of the ‘Big Two’? Preferential Processing of Communal Information ,” Journal of Experimental Social Psychology , 47 ( 5 ), 935 – 48 . Abele Andrea E. , Cuddy Amy J. C. , Judd Charles M. , Yzerbyt Vincent Y. ( 2008 ), “Fundamental Dimensions of Social Judgment ,” European Journal of Social Psychology , 38 ( 7 ), 1063 – 5 . Abele Andrea E. , Uchronski Mirjam , Suitner Caterina , Wojciszke Bogdan ( 2008 ), “ Towards an Operationalization of the Fundamental Dimensions of Agency and Communion: Trait Content Ratings in Five Countries Considering Valence and Frequency of Word Occurrence ,” European Journal of Social Psychology , 38 ( 7 ), 1202 – 17 . Abele Andrea E. , Wojciszke Bogdan ( 2007 ), “ Agency and Communion from the Perspective of the Self versus Others ,” Journal of Personality and Social Psychology , 93 ( 5 ), 751 – 63 . Abele Andrea E. , Wojciszke Bogdan ( 2014 ), “Communal and Agentic Content in Social Cognition: A Dual Perspective Model,” in Advances in Experimental Social Psychology , Vol. 20 , ed. Zanna Mark P. , Olson James M. , Burlington, MA : Academic Press , 195 – 255 . Aggarwal Pankaj ( 2004 ), “ The Effects of Brand Relationship Norms on Consumer Attitudes and Behavior ,” Journal of Consumer Research , 31 ( 1 ), 87 – 101 . Aggarwal Pankaj , Larrick Richard P. ( 2012 ), “ When Consumers Care About Being Treated Fairly: The Interaction of Relationship Norms and Fairness Norms ,” Journal of Consumer Psychology , 22 ( 1 ), 114 – 27 . Ailawadi Kusum L. , Zhang Jie , Krishna Aradhna , Kruger Michael W. ( 2010 ), “ When Wal-Mart Enters: How Incumbent Retailers React and How This Affects Their Sales Outcomes ,” Journal of Marketing Research , 47 ( 4 ), 577 – 93 . Bakan David ( 1966 ), The Duality of Human Existence: Isolation and Communion in Western Man , Chicago : Rand McNally . Bettencourt B. Ann , Dill Karen E. , Greathouse Scott A. , Charlton Kelly , Mulholland Amy ( 1997 ), “ Evaluations of Ingroup and Outgroup Members: The Role of Category-Based Expectancy Violation ,” Journal of Experimental Social Psychology , 33 ( 3 ), 244 – 75 . Christopher Andrew N. , Schlenker Barry R. ( 2000 ), “ The Impact of Perceived Material Wealth and Perceiver Personality on First Impressions ,” Journal of Economic Psychology , 21 ( 1 ), 1 – 19 . Conway Michael , Pizzamiglio M. Teresa , Mount Lauren ( 1996 ), “ Status, Communality, and Agency: Implications for Stereotypes of Gender and Other Groups ,” Journal of Personality and Social Psychology , 71 ( 1 ), 25 – 38 . Dépret Eric , Fiske Susan T. ( 1999 ), “ Perceiving the Powerful: Intriguing Individuals versus Threatening Groups ,” Journal of Experimental Social Psychology , 35 ( 5 ), 461 – 80 . Eagly Alice H. , Wood Wendy , Diekman Amanda B. ( 2000 ), “Social Role Theory of Sex Differences and Similarities: A Current Appraisal,” in The Developmental Social Psychology of Gender , ed. Eckes Thomas , Trautner Hanns Martin , Mahwah, NJ : Lawrence Erlbaum , 123 – 74 . Fiske Susan T. , Cuddy Amy J. C. , Glick Peter , Xu Jun ( 2002 ), “ A Model of (Often Mixed) Stereotype Content: Competence and Warmth Respectively Follow from Perceived Status and Competition ,” Journal of Personality and Social Psychology , 82 ( 6 ), 878 – 902 . Fiske Susan T. , Neuberg Steven L. ( 1990 ), “A Continuum of Impression-Formation, from Category-Based to Individuating Processes: Influences of Information and Motivation on Attention and Interpretation,” in Advances in Experimental Social Psychology, Vol. 23 , ed. Zanna Mark P. , Waltham, MA : Academic Press , 1 – 74 . Fournier Susan ( 1998 ), “ Consumers and Their Brands: Developing Relationship Theory in Consumer Research ,” Journal of Consumer Research , 24 ( 4 ), 343 – 73 . Galinsky Adam D. , Gruenfeld Deborah H. , Magee Joe C. ( 2003 ), “ From Power to Action ,” Journal of Personality and Social Psychology , 85 ( 3 ), 453 – 66 . Grandey Alicia A. , Fisk Glenda M. , Mattila Anna S. , Jansen Karen J. , Sideman Lori A. ( 2005 ), “ Is ‘Service with a Smile’ Enough? Authenticity of Positive Displays during Service Encounters ,” Organizational Behavior and Human Decision Processes , 96 ( 1 ), 38 – 55 . Hoch Stephen J. , Deighton John ( 1989 ), “ Managing What Consumers Learn from Experience ,” Journal of Marketing , 53 ( 2 ), 1 – 20 . Jones Edward E. , Harris Victor A. ( 1967 ), “ The Attribution of Attitudes ,” Journal of Experimental Social Psychology , 3 ( 1 ), 1 – 24 . Keltner Dacher , Gruenfeld Deborah H. , Anderson Cameron ( 2003 ), “ Power, Approach, and Inhibition ,” Psychological Review , 110 ( 2 ), 265 – 84 . Kervyn Nicolas , Fiske Susan T. , Malone Chris ( 2012 ), “ Brands as Intentional Agents Framework: How Perceived Intentions and Ability Can Map Brand Perception ,” Journal of Consumer Psychology , 22 ( 2 ), 166 – 76 . Kipnis David ( 1972 ), “ Does Power Corrupt? ” Journal of Personality and Social Psychology , 24 ( 1 ), 33 – 41 . Kirmani Amna , Hamilton Rebecca W. , Thompson Debora V. , Lantzy Shannon ( 2017 ), “ The Differential Effect of Underdog Positioning on Moral and Competent Service Providers ,” Journal of Marketing , 81 ( 1 ), 103 – 17 . Kunda Ziva , Oleson Kathryn C. ( 1995 ), “ Maintaining Stereotypes in the Face of Disconfirmation: Constructing Grounds for Subtyping Deviants ,” Journal of Personality and Social Psychology , 68 ( 4 ), 565 – 79 . Kupor Daniella , Tormala Zakary ( 2018 ), “When Moderation Fosters Persuasion: The Persuasive Power of Deviatory Reviews,” Journal of Consumer Research , 45 ( 3 ), 490 – 510 . Malone Chris , Fiske Susan T. ( 2013 ), The Human Brand: How We Relate to People, Products, and Companies , San Francisco : Jossey-Bass . Manjoo Farhad ( 2017 ), “Uber’s Lesson: Silicon Valley’s Start-Up Machine Needs Fixing,” New York Times, https://nyti.ms/2tOSBHc. McShane Blakeley B. , Bockenholt Ulf ( 2017 ), “Single-Paper Meta-Analysis: Benefits for Study Summary, Theory Testing, and Replicability,” Journal of Consumer Research , 43 , 1048 – 63 . Mellers Barbara A. , Schwartz Alan , Ho Katty , Ritov Ilana ( 1997 ), “ Decision Affect Theory: Emotional Reactions to the Outcomes of Risky Options ,” Psychological Science , 8 ( 6 ), 423 – 9 . Mukherji Prokriti , Sorescu Alina , Prabhu Jaideep C. , Chandy Rajesh K. ( 2011 ), “ Behemoths at the Gate: How Incumbents Take on Acquisitive Entrants (and Why Some Do Better than Others) ,” Journal of Marketing , 75 ( 5 ), 53 – 70 . Neumark David , Zhang Junfu , Ciccarella Stephen ( 2008 ), “ The Effects of Wal-Mart on Local Labor Markets ,” Journal of Urban Economics , 63 ( 2 ), 405 – 30 . Olson James M. , Roese Neal , Zanna Mark P. ( 1996 ), “Expectancies,” in Social Psychology: Handbook of Basic Principles , ed. Higgins E. Tory , Kruglanski Arie W. , New York : Guilford , 211 – 38 . Paharia Neeru , Avery Jill , Keinan Anat ( 2014 ), “ Positioning Brands against Large Competitors to Increase Sales ,” Journal of Marketing Research , 51 ( 6 ), 647 – 56 . Paharia Neeru , Keinan Anat , Avery Jill , Schor Juliet B. ( 2011 ), “ The Underdog Effect: The Marketing of Disadvantage and Determination through Brand Biography ,” Journal of Consumer Research , 37 ( 5 ), 775 – 90 . Pettigrew Thomas F. , Martin Joanne ( 1987 ), “ Shaping the Organizational Context for Black American Inclusion, ” Journal of Social Issues , 43 ( 1 ), 41 – 78 . Poppe Edwin , Linssen Hub ( 1999 ), “ In-Group Favouritism and the Reflection of Realistic Dimensions of Difference between National States in Central and Eastern European Nationality Stereotypes ,” British Journal of Social Psychology , 38 ( 1 ), 85 – 102 . Preacher Kristopher J. , Rucker Derek D. , Hayes Andrew F. ( 2007 ), “ Addressing Moderated Mediation Hypotheses: Theory, Methods, and Prescriptions ,” Multivariate Behavioral Research , 42 ( 1 ), 185 – 227 . Rucker Derek D. , Galinsky Adam D. , Dubois David ( 2012 ), “ Power and Consumer Behavior: How Power Shapes Who and What Consumers Value ,” Journal of Consumer Psychology , 22 ( 3 ), 352 – 68 . Rucker Derek D. , Hu Miao , Galinsky Adam D. ( 2014 ), “ The Experience versus the Expectations of Power: A Recipe for Altering the Effects of Power on Behavior ,” Journal of Consumer Research , 41 ( 2 ), 381 – 96 . Rudman Laurie A. , Glick Peter ( 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 – 10 . Scherer Frederic M. , Ross David ( 1990 ), Industrial Market Structure and Economic Performance , Boston : Houghton Mifflin . Schmalensee Richard ( 1989 ), “ Intra-industry Profitability Differences in US Manufacturing 1953–1983 ,” Journal of Industrial Economics , 37 ( 4 ), 337 – 57 . Shankar Venkatesh ( 1999 ), “ New Product Introduction and Incumbent Response Strategies: Their Interrelationship and the Role of Multimarket Contact ,” Journal of Marketing Research , 36 ( 3 ), 327 – 44 . Slater Stanley F. , Narver John C. ( 1994 ), “ Market Orientation, Customer Value, and Superior Performance ,” Business Horizons , 37 ( 2 ), 22 – 8 . Suitner Caterina , Maass Anne ( 2008 ), “ The Role of Valence in the Perception of Agency and Communion ,” European Journal of Social Psychology , 38 ( 7 ), 1073 – 82 . Taylor Shelley E. , Fiske Susan T. , Etcoff Nancy L. , Ruderman Audrey J. ( 1978 ), “ Categorical and Contextual Bases of Person Memory and Stereotyping ,” Journal of Personality and Social Psychology , 36 ( 7 ), 778 – 93 . Thompson Craig J. , Arsel Zeynep ( 2004 ), “ The Starbucks Brandscape and Consumers’ (Anticorporate) Experiences of Glocalization ,” Journal of Consumer Research , 31 ( 3 ), 631 – 42 . Tozzi John ( 2012 ), “Small Business Saturday Grows Up,” https://www.bloomberg.com/news/articles/2012-11-26/small-business-saturday-grows-up. © The Author(s) 2018. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Journal of Consumer Research – Oxford University Press
Published: Apr 1, 2019
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera