Coke vs. Pepsi: Brand Compatibility, Relationship Power, and Life Satisfaction

Coke vs. Pepsi: Brand Compatibility, Relationship Power, and Life Satisfaction Individuals often evaluate, purchase, and consume brands in the presence of others, including close others. Yet relatively little is known about the role brand preferences play in relationships. In the present research, the authors explore how the novel concept of brand compatibility, defined as the extent to which individuals have similar brand preferences (e.g., both partners prefer the same brand of soda), influences life satisfaction. The authors propose that when brand compatibility is high, life satisfaction will also be high. Conversely, because low brand compatibility may be a source of conflict for the relationship, the authors propose that it will be associated with reduced life satisfaction. Importantly, the authors predict that the effects of brand compatibility on conflict and life satisfaction will depend upon relationship power. Across multiple studies and methodologies, including experimental designs (studies 2, 3, 5) and dyadic data from real-life couples (studies 1, 4, 6), the authors test and find support for their hypotheses. By exploring how a potentially unique form of compatibility influences life satisfaction, including identifying a key moderator and an underlying mechanism, the current research contributes to the literatures on branding, close relationships, consumer well-being, and relationship power. brand compatibility, close relationships, relationship power, consumer well-being, multilevel modeling, actor-partner interdependence model Consider for a moment some of your favorite brands. For example, perhaps Coke, Starbucks, and Crest are your favorite brands of soda, coffee, and toothpaste, respectively. Now think about your partner’s favorite brands in the same product categories. Perhaps your partner likes the same brands, or, perhaps your partner prefers Pepsi, Dunkin’ Donuts, and Colgate. Does this compatibility, or lack of compatibility, matter? In the present research, we propose that brand preferences within close relationships do matter, and that they can have significant consequences, including influencing life satisfaction. To investigate this proposition, we introduce the term brand compatibility, which we define as the extent to which individuals in close relationships share similar brand preferences, and we examine whether brand compatibility is related to life satisfaction. Importantly, we hypothesize that the link between brand compatibility and life satisfaction will depend upon power in the relationship. Power is typically defined as the ability to control valued resources and the capacity to influence the behavior of others (Anderson and Galinsky 2006; Emerson 1962; French and Raven 1959; Keltner, Gruenfeld, and Anderson 2003). Imagine that you and your partner have different brand preferences. Now imagine that you perceive you have a lot of power in the relationship. Most likely, your partner’s preferences would not affect your day-to-day consumer decisions. You would probably still drink Coke and go to Starbucks. Further, research has shown that high-power individuals are less likely to notice the attitudes and opinions of their lower-power counterparts (Berdahl and Martorana 2006; Fiske 1993), so you might not even be aware that your partner prefers Pepsi and Dunkin’ Donuts. Thus, brand (in)compatibility most likely would not be an issue for you. On the other hand, imagine that you do not have a lot of power in the relationship. In this situation, you are more likely to be aware of your partner’s different brand preferences as you find yourself drinking more Pepsi and stopping at Dunkin’ Donuts more often than you would prefer. Given that these differences are more pronounced for low-power partners, you may view the relationship as having more disagreements or greater conflict, which in turn may lead to lower happiness in your day-to-day life. Thus, we suggest that the link between brand compatibility, conflict, and life satisfaction depends upon whether one is relatively high or low in power within one’s romantic relationship. When one is high in power, then one should be able to control relationship outcomes and get the brands that one prefers. Thus, we predict that for those who are high in relationship power, brand compatibility will not affect life satisfaction presumably because they are able to acquire their preferred brands regardless of their partner’s preferences. Conversely, when one is low in relationship power, one is less likely to get the brands that one prefers. As a result, lower-power partners may be more likely to perceive greater conflict in the relationship. Therefore, we predict that for lower-power individuals, brand compatibility will be related to life satisfaction. Specifically, when brand compatibility is high, life satisfaction will also be high because there is no conflict—lower-power individuals are able to enjoy their preferred brands (which are the same as their partner’s preferred brands). But for low-power individuals, when brand compatibility is low, perceived conflict should be increased and life satisfaction will be reduced. We test these predictions across six studies, using a mix of dyadic studies of real relationship partners and experimental studies that allow us to manipulate perceptions of brand compatibility, and thus provide causal evidence for its importance above and beyond other types of compatibility. In addition, we provide evidence that perception of conflict is the mechanism through which brand compatibility and power interact to influence life satisfaction. The present research makes several contributions. First, we explore a new construct, brand compatibility, and demonstrate that it can have important downstream consequences for life satisfaction. By connecting this consumer construct to a well-established marker of psychological well-being, our findings support the importance of consumer behavioral constructs to the health and well-being literature (Diener and Biswas-Diener 2002; Fredrickson and Joiner 2002). Second, we explore how brand preferences intersect with relationship contexts, contributing to burgeoning efforts to understand how consumer behavior both shapes and is shaped by social relationships (Corfman and Lehmann 1987; Dzhogleva and Lamberton 2014; Lowe and Haws 2014; Luo 2005; Park 1982; Raghunathan and Corfman 2006; Steffel and Le Boeuf 2014). Third, our findings illustrate the importance of brand preferences in relationships. Relationship scholars have long been interested in compatibility (Blossfeld 2009; Bramlett and Mosher 2002; Byrne 1971; Gonzaga, Campos, and Bradbury 2007; Little and Perrett 2002; Mare 1991; Rusbult et al. 2009), but most theorists have argued that similarity of important values such as religion and ethical beliefs is what matters. In the current studies, we show that even mundane, everyday forms of compatibility, such as brand compatibility, can have significant consequences both for the relationship, in terms of perceived conflict, and for the individual, in terms of life satisfaction. Our findings support classic theoretical arguments that social similarities (e.g., race, religion) are likely to be only modestly related to how well couples get along, whereas areas of similarity that affect the day-to-day decisions partners make are likely to be more important (Levinger and Breedlove 1966; Levinger and Rands 1985). Fourth, our findings point to the importance of studying power in close relationships. While most research on power explores organizational and broader social contexts, recently theorists have highlighted the need to reintroduce the power concept to the study of close relationships (Simpson et al. 2014; Simpson, Griskevicius, and Rothman 2012). In particular, the current research challenges the long-held assumption in the relationship literature that compatibility is inherently good (Acitelli, Douvan, and Veroff 1993; Byrne 1971; Montoya, Horton, and Kirchner 2008; Murray et al. 2002); instead, our findings suggest that compatibility may matter primarily for those who perceive having low power in the relationship. THEORETICAL BACKGROUND Brand Compatibility Brands play many roles in consumers’ day-to-day lives. As brands can represent one’s identity, personality, beliefs, social connections, culture, and heritage (Escalas and Bettman 2003; Holt 2002; Muniz and O’Guinn 2001; Ng and Houston 2006; Park and John 2010), individuals may use brands to communicate to others who they are or to represent their self (Belk 1988; Berger and Heath 2007; Escalas and Bettman 2005; Malhotra 1988). Research has shown that individuals may use brands as relationship partners (Aaker, Fournier, and Brasel 2004; Aggarwal 2004; Carroll and Ahuvia 2006; Fournier 1998; Park et al. 2010; Thomson, MacInnis, and Park 2005), or individuals may merely use brands as a heuristic to help guide choice in a world increasingly overrun with options (Aaker and Keller 1990; Hoeffler and Keller 2003; Keller 1993). Regardless of the reason, individuals commonly evaluate, purchase, and consume their preferred brands. Although individuals typically consume the brands that they prefer, there are situations in which they may not be able to do so. For example, workplace lunches provide certain brands of soda; certain shopping centers have only one coffee shop. When brand preferences are constrained by situational factors, consumers tend to feel less satisfied with the experience (Clee and Wicklund 1980; Fitzsimons 2000). We suggest that close relationships, in addition to stockouts and other consumer variables (e.g., time, money), can also constrain the expression of brand preferences. Indeed, close relationships are one of the strongest, most controlling, and long-term types of social situations, so their opportunity for influence is extremely high (Thibaut and Kelley 1959). As close relationships involve repeated, frequent interactions in diverse settings, and strong mutual interdependence whereby the outcomes of one’s decisions affect both the other person and the relationship itself (Berscheid, Snyder, and Omoto 1989; Kelley 2013; Thibaut and Kelley 1959), individuals are often constrained by their partner’s preferences and the demands of the relationship. This constraint should be especially evident and influential in everyday life when the partners have different consumer preferences, including brand preferences, and be most likely to influence life satisfaction. To illustrate why brand preferences may matter, consider the following two scenarios. Imagine you and your partner are looking for a new car. You have always wanted an Audi. In the first scenario, imagine your partner also loves Audi. You buy the Audi and are both happy. In this first scenario, you and your partner have high brand compatibility, defined as the degree to which individuals have similar brand preferences. However, imagine a second scenario in which your partner does not like Audi. Instead, your partner prefers Lexus cars, and in this scenario, you get the Lexus. This is obviously not a terrible situation (it is still a luxury car brand), but it is not the car brand you had always wanted. Now imagine you are driving with your partner to get coffee. First, imagine you and your partner both like Starbucks. There is no conflict. You stop at Starbucks, you both get to enjoy the coffee you prefer, and you are on your way. However, in the second scenario, you like Starbucks, but your partner prefers Dunkin’ Donuts. Where do you stop? Ideally, you stop at both places, but perhaps due to time constraints you can stop at only one. Further, imagine that instead of just one time, this is your daily commute to work. This simple brand decision, which brand of coffee to buy, is part of your everyday life as a couple. Daily, you find yourself driving in the Lexus (the car you did not want), drinking Dunkin’ Donuts (coffee that is not your favorite). Extending beyond the car and coffee example, you, as a couple, may also have to choose between two different brands of soda, beer, or even toilet paper to purchase and consume at home. In the moment, and over time, these small, seemingly trivial brand choices may have consequences. High brand compatibility means that both partners’ desire to satisfy their brand preferences can be easily accomplished, in the moment and on a regular, daily basis, and we conjecture that it may be influential in the couple’s overall well-being, as is fulfillment of other goals and needs in everyday life (Emmons 1986; Emmons and King 1988). On the other hand, over the lifetime of a relationship, low brand compatibility means that a couple is likely to make hundreds or thousands of small decisions in which the two partners cannot both satisfy their preferred outcome. Having dissimilar brand preferences, or low brand compatibility, thus creates multiple opportunities for conflict and presents a challenge to the couple. If you really like Diet Coke, Starbucks, and Audis but find yourself drinking your partner’s favorites, Diet Pepsi and Dunkin’ Donuts, and driving a Lexus, you might perceive greater conflict in the relationship and be less happy. This is what we propose in the current research: that low brand compatibility will be associated with greater perceived conflict and reduced life satisfaction for those partners who fall on the “losing” side of incompatible preferences (i.e., those low in power; see the following section). We propose this will happen both in the moment—that people will temporarily feel less satisfied—and over time, as these experiences accumulate in memory. The potential costs and benefits of similarity in relationships has been a topic of immense interest in the relationships literature for decades (Acitelli et al. 1993; Bramlett and Mosher 2002; Byrne 1971; Dyrenforth et al. 2010; Houts, Robins, and Huston 1996; Montoya et al. 2008; Watson et al. 2004), but compatibility researchers have more typically focused on deeply held values like religious beliefs and political ideologies (Heaton and Pratt 1990) and on personality traits (Blum and Mehrabian 1999; Glicksohn and Golan 2001). Here, we extend the logic of compatibility to study a far more mundane and everyday type of similarity—namely, brand preferences. Importantly (and this is where our research turns away from the standard assumption that compatibility is straightforwardly beneficial), we suggest that these two situations—of high and low compatibility—are of differential importance to the members of the couple, depending on how much power each partner holds in the relationship. Power in Relationships Power is a common and pervasive component of social interactions and relationships (Anderson, John, and Keltner 2012; Galinsky, Gruenfeld, and Magee 2003). Power is commonly defined as having relative control over valued resources and capacity to influence the behavior of others (Anderson and Galinsky 2006; Emerson 1962; French and Raven 1959; Keltner et al. 2003; Magee and Galinsky 2008), while resisting the influence of others over oneself (Cromwell and Olson 1975). Power in close or romantic relationships can be thought of similarly—as the capacity to influence outcomes in the relationship. In the current research, we focus directly on perceived relationship power. We assume that perceived relationship power is related to decisions and experiences within the relationship, including the outcomes of brand decisions. That is, we assume that partners who perceive themselves as high power more reliably obtain and consume their desired brands than do partners who perceive themselves as low power. Although this has not been experimentally demonstrated, it appears to be a safe assumption: by definition, partners who are relatively higher in power are more likely to report obtaining their desired outcomes. Indeed, providing some support for the notion, historical research on gender and family decision making suggested that men reported greater power and reported controlling more of the important purchasing decisions in families (Filiatraut and Ritchie 1980; Kirchler 1993). If this logic extends to minor brand decisions in everyday life, partners who perceive higher power should more frequently consume the brands they want to consume. This tendency for high-power partners to “get what they want” may occur via several interrelated mechanisms. First, and most simply, they may just control the outcomes directly, by demands or pressure on their partners (Simpson et al. 2014). The Starbucks-loving high-power partner may simply scoff at the suggestion of stopping at Dunkin’ Donuts. However, the process may also emerge from subtler and more indirect psychological dynamics in the couple. High-power individuals are more likely to express their opinions to others (Berdahl and Martorana 2006); thus, perhaps high-power partners will simply make their brand preferences known to their partners more often or more clearly. High-power individuals have also been shown to be more impervious to others’ attitudes. For example, they have a reduced tendency for perspective taking and comprehending how others think and feel (Galinsky et al. 2006) and are less motivated to attend to others’ thoughts and behaviors (Fiske 1993; Keltner et al. 2003). Thus, high-power partners are less likely to notice the discrepant preferences of their relationship partners, which likely reduces the chance they would consider those preferences in their brand choices. In the realm of decision making, consumers high in power tended to discount others’ opinions (Mourali and Yang 2013). Third, high-power individuals may be less likely to value their partners’ preferences. For example, consumers led to feel high in power spent more on purchases for themselves, and less on purchases for others, while consumers led to feel low in power showed the reverse pattern, spending more on purchases for others (Rucker, Dubois, and Galinsky 2011). As a result of these dynamics, in everyday life, individuals who perceive greater power in the relationship are more likely to obtain their desired brands regardless of their partner’s preferences. Therefore, we predict that for high-power individuals, brand compatibility is not a particularly important construct in determining their quality of life or sense of well-being. If they have high brand compatibility with their romantic partner, that is great, but if not, that is still good—their own outcomes are relatively unaffected by their partner’s preferences. In other words, differences in brand preferences should not be perceived as a source of conflict for high-power partners. Thus, we predict that for individuals who are high in power, brand compatibility will not affect life satisfaction. Quite in contrast, low-power partners, by definition, are less likely to have control over outcomes within the relationship, including brand choices. Low-power partners are much less likely to attempt to control the outcome directly through pressure or demands (Simpson et al. 2014). In addition, they have been found to express their opinions to others less (Berdahl and Martorana 2006), and they are more likely to pay attention to the preferences, attitudes, and feelings of their partners (Fiske 1993; Keltner et al. 2003). Furthermore, research has shown that they conform to other people’s opinions (Mourali and Yang 2013) and value their outcomes (Rucker et al. 2011). Consequently, they may be more likely simply to know their partner’s preferred brands and to conform to them when making brand choices, even if those preferences do not reflect their own. For these reasons, low-power individuals are relatively less likely to “win” in conflicts over brand preferences in close relationships. As a result, brand compatibility may play a much larger role for low-power partners. Given this reasoning, we hypothesize that the construct of brand compatibility should be of particular importance in predicting the well-being, in terms of relationship conflict and life satisfaction, of low-power individuals. CURRENT RESEARCH In the current research, we investigate how power dynamics and brand preferences in close romantic relationships affect life satisfaction. As high-power partners are less likely to be aware of their partners’ preferences (Berdahl and Martorana 2006; Fiske 1993) and more likely to perceive control over outcomes within the relationship (Galinsky et al. 2003; Simpson et al. 2012), including brand outcomes, we predict that brand compatibility will not affect life satisfaction for high-power individuals. On the other hand, we predict that for low-power individuals, as brand compatibility decreases, so too will life satisfaction. As low-power individuals are more likely to be aware of differences in brand preferences and less likely to perceive control over outcomes, they will have less control over consumption decisions relative to their partner, making the similarity of the two partners’ preferences very important to their outcomes. We further predict that low-power partners will perceive greater conflict in the relationship when brand compatibility is low. This greater conflict will in turn contribute to their reduced life satisfaction. We tested these hypotheses across six studies (and one direct replication). As this is a new construct, in our first study we examined whether brand compatibility between romantic partners was related to life satisfaction and other forms of compatibility. In our second study, we experimentally manipulated perceived brand compatibility and measured relationship power and life satisfaction. In our third study, we experimentally manipulated perceived power in the relationship and measured brand compatibility and life satisfaction. In our fourth study, we simultaneously measured brand compatibility, power, and life satisfaction in a laboratory setting with both members of romantic couples. In dyadic analyses, we used the Actor-Partner Interdependence Model (Kenny, Kashy, and Cook 2006) to further investigate how power influences the effect of brand compatibility on life satisfaction. We explored the possibility of an actor-by-partner interaction, examining whether the strongest effect of brand compatibility on life satisfaction would emerge for low-power individuals with high-power partners. In our fifth study, we examined potential mechanisms of this effect. Using a scenario study in which we were able to manipulate both brand compatibility and relationship power, we explored the role of perceived conflict. Finally, in our sixth study, we again examined both members of romantic couples to examine the dyadic effects, and test the notion that brand compatibility and power’s interactive effect on life satisfaction is mediated by perceptions of conflict within the relationship. If high-power individuals are not aware of or affected by incompatibility, they should be less likely to perceive conflict. If low-power individuals are more aware of their partner’s preferences and more affected by incompatibility, they should perceive greater relationship conflict, and thus, feel less satisfied with their lives. Across all of the studies, we report all data exclusions (if any), all manipulations, and all measures in the study (with the exceptions noted below). First, for each study involving both members of the couple, sample sizes were based on subject availability as well as the needs of unrelated research projects that were run in conjunction with data collection. For studies using online participants, sample sizes were determined a priori. No additional data were collected after data analyses began (with the exception of study 2, in which additional conditions were added in response to reviewer request). In some experiments, as indicated in the text, we included additional irrelevant measures, such as filler items or measures for other research projects. As discussed in the methods section of study 4, data from this study were collected in two waves during multistudy events with other researchers in the lab; we do not have access to all those measures or data. Finally, in study 6, a multiwave, longitudinal dyadic study, not all measures in the study are reported due to the sheer number of variables and measures. All measures related to the hypotheses are described, and only the described measures were analyzed for the current hypothesis. STUDY 1 As this is the first research to examine the role of brand compatibility on life satisfaction, we first sought to explore whether brand compatibility is associated with life satisfaction and other forms of compatibility. In study 1, we survey both partners within a couple. We ask them about various measures that are commonly examined in similarity and compatibility research (race, religion, personality, values, etc.). In addition, we ask each partner to report his/her own brand preferences and measure life satisfaction, or general happiness with how one’s life is going, as the dependent variable. By comparing the participant and the partner on each domain, we are thus able to construct objective measures of these various forms of compatibility using outside raters. Method Participants We recruited participants from the local farmer’s market in a southeastern city. In order to participate, participants had to be in a relationship with their partner for at least six months and living with their partner for at least three months. Both partners had to be present and willing to complete the survey in order to participate. Both partners of 63 romantic relationships (52% female, 1% trans, 1% both, 1% neither) completed the survey. Due to comprehension/incomplete data issues, two participants and their partners were excluded from all analyses. Partners ranged in age from 25 to 70 years with an average age of 40.67 years (SD = 12.94) and had been in a relationship for 13.61 years on average (SD = 12.96). Couples received financial compensation in exchange for their participation. Measures and Procedure After indicating consent, each partner was given a clipboard with a survey and told that there were several parts to the survey. Each partner was instructed to complete the survey separately and to not discuss their responses with their partner. For each of the compatibility factors we created a compatibility score for each couple. (Note: We repeated the analyses using various coding schemes, to ensure that the results are not driven by the use of any specific type of coding scheme.) Brand Compatibility Each partner was asked to list his/her favorite brand in each of the following five categories: coffee, chocolate, car, beer, and soda. We selected these categories because they are common categories wherein most individuals in our sample would have experience with different brands, and in which individuals generally have specific brand preferences. In addition, they are categories in which individuals regularly make choices and wherein a partner is repeatedly exposed to these choices (e.g., every time they drive the household car they are reminded of their partner’s preference, or every time they open the fridge they may see their partner’s brand of soda, or every time they pass a Starbucks they are reminded of their partner’s favorite brand of coffee). Furthermore, these categories vary in terms of price and type (e.g., durable vs. nondurable). Essentially, we simply sought to sample a wide variety of product types. Four undergraduate research assistants blind to the hypothesis of the study rated each of the brand pairs (i.e., how the brand that partner 1 within couple 1 listed as his/her favorite compared with the brand that partner 2 within couple 1 listed as his/her favorite) within a product category on a 1 (completely incompatible) to 5 (completely compatible) scale. Coders were instructed to evaluate the brands on how compatible, or similar, one partner’s brand was with the other partner’s brand. Each coder was given some examples of brand compatibility coding using the category of soda. They were told that the exact same brand response within a couple for a category, such as partner 1 responding with Diet Coke and partner 2 also responding with Diet Coke, would be a 5. Brands that are close but not exact, such as partner 1 responding Coke and partner 2 responding Diet Coke, would be a 4. A 2 or 3 would be given for brand pairs that are not identical, but not competitors. For example, if partner 1 said Coke and Partner 2 said Sprite, they would be given a 3. These brands are not completely compatible, but have an overarching relationship—in this case, that they are owned by the same parent company. Competitor or opposite brands would be coded as a 1. For example, if partner 1 said Coke and partner 2 said Diet Pepsi, they would be given a 1. This coding scheme has the advantage of allowing us to differentiate between responses such as “No preference” from “I don’t like soda,” which would have different implications for the couple. For example, if a participant said he liked Coke but his partner said she had no preference, this would be coded as more compatible than if the participant said he liked Coke, but his partner said she does not like soda. In the former case, purchasing a 12-pack of Coke could satisfy both partners, whereas in the latter case no brand of soda will satisfy both partners.1 In addition, coders were told that this is a subjective rating—that there are no right or wrong answers, but that they should be consistent in their ratings. Coders were told that if they had any questions about the brands, they could look them up online. The four coders’ ratings were averaged to create a brand compatibility score for each brand category (Mbeer = 2.41, SD = 1.22; Mcar = 2.69, SD = 1.20; Mchoc = 2.81, SD = 1.38; Mcoffee = 3.22, SD = 1.34; Msoda = 3.00, SD = 1.47). Coders were highly reliable within each brand category (interrater reliability: beer α = .89; car α = .91; chocolate α = .93; coffee α = .94; soda α = .94). Brand categories were then averaged to create one mean brand compatibility score for each couple, which served as the independent variable. Other Measures of Compatibility The following were included as additional measures of compatibility: age, education, race, religiousness, religion, political orientation, personality, and values. We selected these items and calculated similarity within couples based on previous research in similarity and compatibility (Houts et al. 1996; Watson et al. 2004; see Finkel et al. 2012 for a review). (Please see the web appendix for detailed descriptions regarding the coding of these items.)2 Each partner also completed the Satisfaction with Life Scale (Diener et al. 1985; α = .85). (Note: This measure was completed third, after participants indicated their favorite brands and completed the values measure.) This is a well-established measure of life satisfaction, commonly used in research on well-being (Aknin et al. 2013; Burroughs and Rindfleisch 2002; Cohn et al. 2009; Diener and Biswas-Diener 2002; Diener et al. 2010; Luhmann et al. 2012; Martin and Hill 2012) and has been shown to correlate highly with other important measures of well-being and quality of life, such as health, happiness, and self-determination (Arrindell, Meeuwesen, and Huyse 1991; Diener 2012; Pavot and Diener 1993). The measure asks participants to evaluate their life overall, and provide a general sense of how well things are going. For example, items include “In most ways my life is close to my ideal” and “I am satisfied with my life.” Participants completed additional items, including filler items and demographic items. Results We investigated whether brand compatibility predicted life satisfaction while controlling for other variables. We used a multilevel modeling approach (Kenny et al. 2006), with individuals nested within couples to account for violations of statistical independence. Life satisfaction served as our outcome variable, and brand compatibility, race compatibility, political orientation compatibility, religiousness compatibility, education compatibility, personality compatibility, and value compatibility served as our predictor variables. All of the predictor variables were grand mean-centered (Aiken and West 1991; Kenny et al. 2006). When we controlled for various other forms of compatibility, brand compatibility was significantly and positively associated with life satisfaction (B = .40, t (51.74) = 2.66, p = .010). See table 1 for details about all of the predictor variables, and the web appendix for information and a table regarding correlations of all variables. We note that these results were robust across a wide array of models including various coding schemes of compatibility, such as absolute difference versus difference score approaches. We report the model that offers the best fit (e.g., lowest AIC of all the models, most parsimonious). (For information regarding the other models and correlations, please see the web appendix.) Table 1 Regression Coefficients Predicting Life Satisfaction in Study 1 Variable  Estimate  Standard Error  t-test  p value  Intercept  5.46  0.09  59.28  .000  Brand compatibility  0.40  0.15  2.66  .01  Age  0.02  0.09  0.18  .86  Race  –0.23  0.15  –1.56  .12  Political orientation  0.09  0.09  0.98  .33  Religiousness  0.09  0.08  1.22  .23  Education  –0.06  0.09  –0.66  .52  Personality (Big 5)  –0.01  0.18  –0.07  .95  Values (LOV)  0.15  0.30  0.51  .61  Variable  Estimate  Standard Error  t-test  p value  Intercept  5.46  0.09  59.28  .000  Brand compatibility  0.40  0.15  2.66  .01  Age  0.02  0.09  0.18  .86  Race  –0.23  0.15  –1.56  .12  Political orientation  0.09  0.09  0.98  .33  Religiousness  0.09  0.08  1.22  .23  Education  –0.06  0.09  –0.66  .52  Personality (Big 5)  –0.01  0.18  –0.07  .95  Values (LOV)  0.15  0.30  0.51  .61  Dependent variable: life satisfaction; AIC 312.323. Discussion In the present research, we find that brand compatibility within couples is related to other forms of compatibility, namely education, race, and values. We also find that, when all forms of compatibility are simultaneously entered as predictors, brand compatibility remains an important, and significant, predictor of life satisfaction. One might wonder how similarity in soda (i.e., brand) preferences could be more influential than similarity in age or religious preferences. We have several thoughts as to why this pattern of associations emerged. First, it is important to acknowledge that mixed results on the big compatibility dimensions (e.g., age, race, religion, personality) are the norm in this field (Tidwell, Eastwick, and Finkel 2013). For example, in the study of sociodemographic similarity, although some research has shown that marriages between individuals of the same race, religious denomination, parental wealth, and earned income are longer-lasting and more satisfying (Bramlett and Mosher 2002; Heaton and Pratt 1990; Weisfeld et al. 1992), other research has failed to find similar benefits (Houts et al. 1996; Watson et al. 2004), while still other research has found positive associations for some variables, but weak or inconsistent associations for others (Gaunt 2006). In the study of personality, some researchers have found positive effects for personality similarity (Luo and Klohnen 2005; Robins, Caspi, and Moffitt 2000), while others have found that after they controlled for the main effects of each partner’s personality, similarity between personalities had a weak relation to romantic outcomes (Blum and Mehrabian 2001; Dyrenforth et al. 2010; Glicksohn and Golan 2001; Tidwell et al. 2013). Indeed, some research has even found that personality similarity has a negative predictive effect on marital satisfaction (Shiota and Levenson 2007). Second, we suggest that brand compatibility produced a stronger correlation precisely because it is a relatively mundane, everyday type of compatibility. As early as 1966, Levinger and Breedlove “emphasized the importance of identifying areas of similarity that bear upon the day-to-day decisions partners make” (Houts et al. 1996). It may be the case that small, everyday decisions are the ones that more directly affect couples’ sense of well-being, as compared with broader and overarching variables like personality and religion. Because individuals commonly evaluate, purchase, and consume brands in the presence of close others, over time the significance of brand choices within a relationship accumulates. Due to their symbolic nature plus the consumer tendency to repeatedly purchase brands, brands represent a salient source of similarity (or potential dissimilarity) in close relationships. Thus, it is possible that brand compatibility is a more specific or idiosyncratic measure of compatibility, and therefore, likelier to be more predictive than general measures. The main aim of the study was to investigate the extent of the overlap among brand compatibility and previously studied forms of compatibility. In line with our theorizing, there is evidence to suggest that the various forms of compatibility are relatively distinct, and that brand compatibility does seem to have some unique predictive power. On the question of whether brand compatibility is a better or weaker predictor than other forms of compatibility, we are certainly not suggesting brands are the only important form of compatibility, but merely one such form, due to their ubiquity in everyday life. Importantly, we predict that the effect of brand compatibility on different outcomes will be related to relationship power. We test this prediction in the rest of the studies. STUDY 2 Study 2 tests our prediction that the effects of brand compatibility on life satisfaction will depend upon power in the relationship. Because high-power individuals have been shown to project their attitudes and feelings on to others (Keltner et al. 2003; Overbeck and Droutman 2013), measuring both brand compatibility and power from one partner could result in biased responses. Therefore, in our second study, we manipulate perceptions of brand compatibility. We hypothesize that for high-power partners, there will be no effect of brand compatibility condition on life satisfaction; however, for low-power partners, we predict that low brand compatibility will be associated with decreased life satisfaction, compared with high brand compatibility. Method Participants Three hundred twenty-five participants (46% men) from Amazon’s Mechanical Turk completed the study in exchange for financial compensation. In order to participate, individuals had to be in a romantic relationship for at least six months and in the United States. Participants ranged in age from 19 to 73 years with an average age of 35.41 years (SD = 11.62) and had been in a relationship for 6.76 years on average (SD = 7.99). Measures and Procedure After individuals indicated consent, they were told that there were several parts to the study and that the researchers were interested in different topics. Participants first completed the Relationship Power Measure (α = .91), which we created by adapting items from the Personal Sense of Power scale (Anderson et al. 2012). The scale ranged from 1 (strongly disagree) to 7 (strongly agree). Instructions indicated that participants should consider their current romantic relationship when answering items. Example items include, “I can get my partner to listen to what I say,” “My wishes do not carry much weight,” and “I think I have a great deal of power.” (See the appendix for items.) We manipulated perceived brand compatibility using an ease-of-retrieval manipulation (Schwarz et al. 1991). Specifically, participants in the high-brand-compatibility condition saw two lines in the survey and were asked to list two favorite brands that they had in common with their partner. Participants in the low-brand-compatibility condition saw eight lines in the survey and were asked to list up to eight brands. Participants had to list a response in the space provided, either a brand name or “NA,” in order to move on to the next question in the survey, to maximize the experience of ease versus difficulty in the two conditions. In order to be consistent with wording across the two conditions all participants were told the following: Some people in relationships are very compatible in their brand preferences. In other words, their favorite brand of soda is their partner’s favorite brand of soda. Other couples are less compatible in their brand preferences and their favorite brand in a product category is different from their partner’s. In the spaces below, please list brands that both you and your partner consider to be your favorite in that product category. For example, if you and your partner have the same favorite brand of soda, chocolate, coffee, car, beer, etc., you would enter the brand name in the space below. Please enter as many shared favorite brands that you and your partner have as you can. If you run out of shared favorite brands, then please enter “NA.” Next, participants completed the same Satisfaction with Life Scale (Diener et al. 1985; α = .80) used in the previous study. Finally, participants completed demographic items—age, gender, relationship type, and relationship length. Results Post-Test Manipulation Check We failed to include a manipulation check in the original study, and thus conducted a post-hoc manipulation check. Four hundred fifteen participants (49% men) from Amazon’s Mechanical Turk were recruited. Individuals had to be in a relationship for at least six months and in the United States in order to participate. Participants were randomized to either the same high- or low-brand-compatibility condition as used in the main study. As our measure of compatibility, participants were asked to indicate the extent to which they agreed or disagreed with the following statements on a seven-point scale: “My partner and I like the same brands,” “My partner and I are very compatible in our brand preferences,” and “My partner and I have similar brand preferences.” We combined these items to form one measure of perceived compatibility (α = .95). Two participants indicated generic categories (e.g., car, restaurants) and were excluded from the analyses. Because the majority of participants viewed themselves as compatible or highly compatible with their partner (85% of the average responses were in the upper half of the scale), we log-transformed the data (Bagchi and Cheema 2013; Dzhogleva and Lamberton 2014). As predicted, individuals in the high-brand-compatibility condition reported significantly greater compatibility (M = 5.23, SD = 1.16) than did individuals in the low-brand-compatibility condition (M = 5.01, SD = 1.36; t(411) = 2.08, p < .04). Main Analyses Two participants indicated that they were single, and were therefore excluded from the analyses. In addition, six participants’ mean power scores were more than three SDs below the mean and were excluded (Smith et al. 2008), leaving 317 participants. To investigate our hypothesis that low brand compatibility would be associated with decreased life satisfaction for low (but not high) power partners, we conducted a linear regression, with life satisfaction as the outcome variable, and brand compatibility and mean-centered power as the predictor variables. Results revealed a main effect for power (β = .37, t(313) = 7.09, p < .0001), such that greater power was associated with greater life satisfaction. There was no main effect of brand compatibility condition (β = .049, t(313) = 0.94, p = .35) on life satisfaction. Importantly, in line with our predictions, and as illustrated in figure 1, results revealed a significant interaction (β = .11, t(313) = 2.08 p < .04). Because we were interested in whether the effects of brand compatibility on life satisfaction differ depending upon power in the relationship, we first examined whether life satisfaction differed across the brand compatibility conditions for high-power individuals. In line with our predictions, we found that for high-power participants (+1 SD above the mean), life satisfaction did not differ across the two brand compatibility conditions (t(313) = .81; p = .42; 95% CI: –.19 to .47). However, for low-power participants (–1 SD), the effect of brand compatibility condition on life satisfaction was significant (t(313) = 2.14; p = .03; 95% CI: –.69 to –.03). Specifically, lower-power participants in the low-brand-compatibility condition reported significantly lower life satisfaction than did individuals in the high-brand-compatibility condition (figure 1). In addition, we conducted a floodlight analysis (Spiller et al. 2013). The results revealed that the difference between the brand compatibility conditions on life satisfaction occurred for participants whose relationship power levels were at or below power levels of 4.82 (approximately half a SD below the mean). FIGURE 1 View largeDownload slide RESULTS FROM STUDY 2: MANIPULATED BRAND COMPATIBILITY AND RELATIONSHIP POWER PREDICTING LIFE SATISFACTION NOTE.—Brand compatibility condition did not affect life satisfaction for high-power individuals. For low-power individuals, the low-brand-compatibility condition was associated with significantly decreased life satisfaction. FIGURE 1 View largeDownload slide RESULTS FROM STUDY 2: MANIPULATED BRAND COMPATIBILITY AND RELATIONSHIP POWER PREDICTING LIFE SATISFACTION NOTE.—Brand compatibility condition did not affect life satisfaction for high-power individuals. For low-power individuals, the low-brand-compatibility condition was associated with significantly decreased life satisfaction. Discussion3 In this study, we measured participants’ chronic sense of power in their relationship using a standard measure, and experimentally manipulated participants’ perception of brand compatibility. In line with our predictions, we found no effect of brand compatibility condition on life satisfaction for high-power individuals. In contrast, as predicted, we found that for low-power individuals, low brand compatibility led to reduced life satisfaction. Because we manipulated brand compatibility, these results highlight that brand compatibility is not just a marker of existing conflict in the relationship, but that brand compatibility, as a unique construct, interacts with power to predict feelings of life satisfaction. STUDY 3 In study 2, we measured power and manipulated brand compatibility. In study 3, we aimed to conduct a complementary study, increasing the generalizability of our findings. First, we experimentally manipulated individuals’ sense of power in the relationship in order to increase our ability to draw causal conclusions about the role of power in these effects. In line with the literature on social power, we theorize that power is both a chronic, long-term variable reflecting structural realities, and a situational variable, in flux in many relationships and social settings (Galinsky et al. 2003). As described in Jiang, Zhan, and Rucker (2014), because of its prevalence in everyday life, power is not only a long-term structural construct that can be measured, but also a more variable mindset that can be triggered or activated through common experimental techniques. Research has shown that power can be evoked through episodic recall, role playing, and even semantic priming (Galinsky et al. 2003; Magee, Galinsky, and Gruenfeld 2007). In other words, because it is a mindset, all people can draw on experiences of feeling both high and low power—even within the same relationship. One can think of parent-child or professor-student relationships as an example. Although the parent and professor chronically hold more power, one can probably recall a few times when it felt as though the child and/or the student had more power, which in turn can elicit responses associated with feeling low in relationship power. Research has also shown that chronic and manipulated power affect consumer outcomes, including thought and behavior, in similar ways (Rucker and Galinsky 2008, 2016; Rucker, Galinsky, and Dubois 2012). These previous results suggest that manipulating perceptions of relationship power by asking participants to recall an instance in their current relationship when they had low power should produce similar effects as chronic perceptions of power in the relationship. Therefore, as in study 2, we predict that for individuals in the high-power condition, there will be no effect of brand compatibility on life satisfaction. On the other hand, we predict that individuals in the low-power condition will report lower life satisfaction as brand compatibility decreases. Method Participants One hundred eighty-one individuals (51% men) from Amazon’s Mechanical Turk completed the study. As in the previous study, individuals had to be in a relationship for at least six months and in the United States in order to participate. The participants ranged in age from 18 to 68 years with an average age of 33.93 years (SD = 11.76) and had been in their relationship for an average of 6.30 years (SD = 7.78). Participants were paid in exchange for participation in the study. Measures and Procedure After providing consent, participants were asked to list their partner’s favorite brand in the same five categories as used in prior studies: coffee, chocolate, car, beer, and soda. Participants were then asked to list their own favorite brands in each of the same brand categories. Again, we selected these categories because we wanted a range of categories in terms of price and type, and these are brand categories in which individuals often have a preference. Next, participants were randomized to either a high- or low-power condition using a standard power manipulation, which we modified slightly to reflect romantic relationships (Galinsky et al. 2003). Specifically, participants in the high- (low-) power condition were told the following: Please recall a particular incident in which you had power over your partner (your partner had power over you). By power, we mean a situation in which you (your partner) controlled the ability of your partner (you) to get something they (you) wanted, or were in a position to evaluate your partner (you). Please describe this situation in which you had power—what happened, how you felt, etc. Participants then completed the Satisfaction with Life scale (Diener et al. 1985; α = .81) as used in study 2. They also completed a series of demographic items—gender, age, relationship type, and relationship length. Similar to study 1, we created a measure of perceived brand compatibility from participants’ brand responses. Two undergraduate research assistants blind to the hypothesis of the study rated each of the brand pairs (i.e., how compatible the brand that the participant listed as his/her favorite compared with the brand that the participant listed as his/her partner’s favorite) within a product category on the same 1 (completely incompatible) to 5 (completely compatible) scale used in study 1. Raters were highly reliable within each brand category (interrater reliability: beer α = .88; car α = .87; chocolate α = .90; coffee α=.91; soda α = .91). Brand categories were then averaged to create one mean brand compatibility score (Mbeer = 2.62, SD = 1.59; Mcar = 2.88, SD = 1.45; Mchoc = 2.85, SD = 1.58; Mcoffee = 3.10, SD = 1.68; Msoda = 2.88, SD = 1.67). Results One participant indicated that he was single and was therefore excluded from the subsequent analyses, leaving 180 participants. Manipulation Check Following prior research (Galinsky et al. 2003; Mourali and Yang 2013; Smith and Bargh 2008), an undergraduate research assistant blind to both condition and hypotheses rated participants’ responses for how much power the participant reported having over his/her partner using a seven-point Likert scale. Participants in the high-power condition described themselves having significantly more power over their partner (M = 5.27, SD = 1.00) than did individuals in the low-power condition (M = 2.48, SD = 1.24; t(178) = 16.54, p < .0001). Main Analyses A linear regression, with life satisfaction as the outcome variable and power condition and mean-centered brand compatibility as the predictor variables, revealed a marginally significant main effect for brand compatibility (β = .20, t(176) = 1.71; p < .09), such that greater brand compatibility was associated with greater life satisfaction. There was no main effect of power condition on life satisfaction (β = .06, t(176) = 0.69; p = .49). Results also revealed a marginally significant interaction (β = –.20, t(176) = –1.75; p = .08). Examining the interaction (see figure 2), we found that the pattern of results replicated the findings of study 2. Specifically, for individuals in the high-power condition, there was no effect of brand compatibility on life satisfaction, (t(176) = –0.02; p, NS; 95% CI = –.35 to .34). However, for individuals in the low-power condition, lower brand compatibility was associated with reduced life satisfaction (t(176) = 2.62; p < .01; 95% CI = .10 to .71) (figure 2). FIGURE 2 View largeDownload slide RESULTS FROM STUDY 3: MANIPULATED RELATIONSHIP POWER AND MEASURED BRAND COMPATIBILITY PREDICTING LIFE SATISFACTION NOTE.—For individuals who were temporarily made to feel high in relationship power, there is no effect of brand compatibility on life satisfaction. However, for individuals who were temporarily made to feel low in power, as brand compatibility decreases, so does life satisfaction. FIGURE 2 View largeDownload slide RESULTS FROM STUDY 3: MANIPULATED RELATIONSHIP POWER AND MEASURED BRAND COMPATIBILITY PREDICTING LIFE SATISFACTION NOTE.—For individuals who were temporarily made to feel high in relationship power, there is no effect of brand compatibility on life satisfaction. However, for individuals who were temporarily made to feel low in power, as brand compatibility decreases, so does life satisfaction. Discussion In this study, we expanded upon the results from the previous study by manipulating, instead of measuring, power in order to provide evidence of the causality of power in moderating this association between compatibility and life satisfaction. Our interaction effect was marginal in this study. However, as predicted, we found that for individuals temporarily led to feel high in power in their romantic relationship, perceived brand compatibility did not affect life satisfaction. On the other hand, for individuals who are temporarily led to feel low in power, we found that as brand compatibility decreases, so too did life satisfaction. These results thus conceptually replicated our findings from the previous study and demonstrated that the effects of brand compatibility on life satisfaction depend upon power in the relationship. STUDY 4 In our fourth study, we brought both members of existing romantic couples to the laboratory to examine our hypothesis from a dyadic perspective (Simpson et al. 2014). By studying both members of the couple, we were able to measure power from both partners’ perspectives, meaning that we could explore the possibility for actor-partner interaction effects (Kenny et al. 2006). Furthermore, we sought to measure brand compatibility in a more objective manner, by separately asking both members of the couple about their brand preferences and then having raters code the brands’ compatibility (as done in study 1). We sought to replicate the effects from the experimental studies in this dyadic sample—namely, that for high-power individuals, brand compatibility would not affect life satisfaction, while for low-power individuals, as brand compatibility decreases, so too would life satisfaction. Furthermore, we explored the possibility that the effect of brand compatibility on life satisfaction would be affected by an actor-partner interaction effect (Kenny et al. 2006). We speculated that the effect of brand compatibility would be strongest for low-power participants with high-power partners. Method Participants Because of the difficulty of recruiting both members of romantic couples to the laboratory, we recruited participants in two waves. The first wave (N = 54 couples) completed this study as part of a multistudy event; the second wave (N = 50 couples) completed this study as part of a different multistudy event. All participants received financial compensation in exchange for their participation. Thus, participants included both members of 104 romantic relationships, a total of 208 participants. They ranged in age from 19 to 62 years with an average age of 27.33 years (SD = 7.36) and had been in a relationship for 4.08 years on average (SD = 5.07). Measures and Procedure After indicating consent, partners were told that there were several parts to the study and were instructed to complete the measures individually. They completed the Satisfaction with Life Scale (Diener et al. 1985; α = .83), as used in the previous studies, before the other measures reported here. They also completed the same relationship power measure used in study 2 (α = .83). Each partner was asked to list his/her favorite brand in each of the same five categories used in the previous studies 1 and 3: coffee, chocolate, car, beer, and soda. As in study 1, we created an index of brand compatibility from each partner’s brand responses. Four undergraduate research assistants rated each couples’ responses within each product category (i.e., couple 1 partner 1’s favorite brand of soda compared with couple 1 partner 2’s favorite brand of soda) on the same 1 (not at all compatible) to 5 (completely compatible). As mentioned previously, using this coding scheme allows us to differentiate between responses such as “no preference” from “I don’t drink soda.” The coders’ ratings were again averaged to create a brand compatibility score for each brand category (Mbeer = 3.14, SD = 1.40; Mcar = 2.89, SD = 1.20; Mchoc = 2.71, SD = 1.27; Mcoffee = 3.40, SD = 1.52; Msoda = 3.10, SD = 1.28). Coders were highly reliable within each brand category (interrater reliability: beer α = .94; car α = .92; chocolate α = .93; coffee α = .95; soda α = .93). Brand categories were averaged to create one mean brand compatibility score for each couple, which served as the independent variable. Results Primary Analyses We hypothesized that the effects of brand compatibility on life satisfaction would depend upon power in the relationship. We first conducted analyses using a multilevel modeling approach (Kenny et al. 2006), with individuals nested within couples to account for violations of statistical independence. Life satisfaction served as our outcome variable, and power, brand compatibility, and the interaction between them served as our predictor variables. All of the predictor variables were grand mean-centered (Aiken and West 1991; Kenny et al. 2006). In these analyses, we initially included a factor for wave of data collection. The effect of wave was not significant and did not interact with our predictors; therefore, wave will not be discussed further. All analyses were conducted using the Mixed Models procedure in SPSS, with all predictors as fixed effects, and power as a level-one (participant-level) predictor and brand compatibility as a level-two (dyad-level) predictor. There was a significant main effect of power on life satisfaction, such that greater relationship power was associated with greater life satisfaction (B = .58, t(187.26) = 7.93, p < .0001), and a nonsignificant positive association between brand compatibility and life satisfaction (B = .22, t(98.22) = 1.55, p < .13). In line with our predictions, and replicating the findings in the two experimental studies, results revealed a significant interaction between power and brand compatibility on life satisfaction (B = –.31, t(180.80) = –2.40, p < .02). We found that for high-power individuals (+1SD), there was no effect of brand compatibility on life satisfaction (B = –.06, t(155.02) = –0.31; p = .76). On the other hand, for low-power individuals (–1 SD), the association of brand compatibility with life satisfaction was significant (B = .48, t(178.29) = 2.60; p = .01). Specifically, for low-power individuals, reduced brand compatibility was associated with significantly lower life satisfaction (figure 3). FIGURE 3 View largeDownload slide RESULTS FROM STUDY 4: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING LIFE SATISFACTION NOTE.—For high-power partners (1 SD above the mean), there is no effect of brand compatibility on life satisfaction. For low-power partners (1 SD below the mean), as brand compatibility decreases, so does life satisfaction. FIGURE 3 View largeDownload slide RESULTS FROM STUDY 4: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING LIFE SATISFACTION NOTE.—For high-power partners (1 SD above the mean), there is no effect of brand compatibility on life satisfaction. For low-power partners (1 SD below the mean), as brand compatibility decreases, so does life satisfaction. Actor-Partner Interdependence Model Analyses In order to explore the possibility that the effect of brand compatibility on life satisfaction would be particularly strong for individuals who saw themselves as low in power and who had partners who saw themselves as high in power, we conducted additional analyses using the actor-partner interdependence model (APIM; Kenny et al. 2006). These analyses separately estimate actor and partner effects within a multilevel modeling framework, and allow us to test the effects of brand compatibility on life satisfaction within couples of varying power combinations (i.e., high-high, high-low, low-high, and low-low). Actor and partner effects are estimated controlling for each other’s influence. As in our previous analyses, and following the norms in the field (Aiken and West 1991; Kenny et al. 2006), all predictor variables were grand mean-centered. APIM analyses regressing life satisfaction on brand compatibility, actor’s power, partner’s power, and all interaction terms revealed significant main effects for both actor power (B = .67, SE =  .07, t(185.84) = 9.11, p < .001) and partner power (B = .21, SE = .07, t(185.84) = 2.88, p = .004), and a nonsignificant trend for brand compatibility (B = .19, SE = .13, t(98.00) = 1.44, p = .15). A significant three-way interaction emerged (B = –.55, SE = .21, t(98.00) = –2.68, p = .009) (figure 4). In line with our theory, there was no effect of brand compatibility on life satisfaction (i.e., the slopes were not significantly different from zero) for high-power actors (1 SD above the mean), regardless of whether their partner was high power (1 SD above the mean) or low power (1 SD below the mean) (slopes 1 and 2 in figure 4; slope 1: B = .11, t(98.00) = 0.52, p = .61; slope 2: B = .19, t(165.12) = .74, p = .46). Further, there was not a significant difference between the slopes for high-power actors (i.e., slopes 1 and 2 were not significantly different from each other: B = –.04, t(133.07) = –0.22, p = .83). There was a negative trend for the effect of brand compatibility on life satisfaction for low-power actors (1 SD below the mean) with low-power partners (slope 4: B = –.57, t(98.00) = –1.80, p = .08); however, given the extremely small number of couples (N = 4 couples) in this cell, it is hard to interpret this result. Interestingly, and in line with our theorizing about the possible actor-partner interaction effect, there was a significant effect of brand compatibility on life satisfaction for low-power actors with high-power partners (slope 3: B = 1.03, t(165.12) = 4.05, p < .0001). Furthermore, the slope for low-power actors with high-power partners was significantly different from the slope for low-power actors with low-power partners (i.e., slope 3 vs. 4: B = .92, t(118.45) = –2.68, p < .0001). FIGURE 4 View largeDownload slide ACTOR-PARTNER INTERDEPENDENCE MODEL (APIM) RESULTS FROM STUDY 4 NOTE.—APIM analyses revealed that for low-power individuals (1 SD above the mean) with high-power partners, as brand compatibility decreases, so too does life satisfaction (slope 3). FIGURE 4 View largeDownload slide ACTOR-PARTNER INTERDEPENDENCE MODEL (APIM) RESULTS FROM STUDY 4 NOTE.—APIM analyses revealed that for low-power individuals (1 SD above the mean) with high-power partners, as brand compatibility decreases, so too does life satisfaction (slope 3). Discussion Study 4 examined in a dyadic laboratory study the association between brand compatibility, which we measured objectively by asking both members of romantic couples for their preferences, and life satisfaction in romantic couples. In line with our predictions, and replicating the findings of the two experimental studies, we found that the effects of brand compatibility on life satisfaction depended upon power in the relationship. We found that for high-power individuals, brand compatibility had no effect on life satisfaction. In other words, people with high relationship power reported similar levels of life satisfaction whether or not the couple had similar brand preferences among common consumption categories. In contrast, for low-power individuals, as brand compatibility decreased, so too did life satisfaction. In general, for these individuals, the compatibility of brand preferences was meaningful, such that those with low brand compatibility reported lower life satisfaction. Thus, these dyadic analyses replicated the findings of our prior experimental studies. In APIM analyses, it appeared that these effects of decreased brand compatibility on reduced life satisfaction were specific to low-power individuals with high-power partners. For low-power actors with high-power partners, the lower-power actor’s life satisfaction was strongly dependent on brand compatibility. On the other hand, for high-power participants, regardless of partner reports of power, life satisfaction was independent of brand compatibility. Unexpectedly, when both partners reported that they had low power, there was a trend for a negative association between brand compatibility and life satisfaction. Although interesting, we hesitate to overinterpret these patterns, as there were very few couples who were low-low power. Speculatively, when each partner is low power, it may suggest an overall level of dysfunction, as neither partner feels able to control the outcomes in the relationship. Such partnerships may be rare, and when they do occur, may be unstable. Either way, this finding highlights the lack of research on these low-low-power pairs (Simpson et al. 2014). Overall, the findings from study 4 contribute to the current research by replicating the experimental findings of studies 2 and 3 and demonstrating the interaction of brand compatibility and power on life satisfaction in both members of real-world romantic couples. STUDY 5 We propose that brand compatibility and power may lead to differences in perceptions of conflict in the relationship. Merriam-Webster defines conflict as an “opposing action of incompatibles” or a “mental struggle resulting from incompatible or opposing needs, drives, wishes, or external or internal demands” (Merriam-Webster.com). Previous research has shown that high-power partners are less likely to notice, appreciate, or consider others’ preferences, including the preferences of their low-power partners (Fiske 1993; Galinsky et al. 2006; Keltner et al. 2003). Furthermore, by definition high-power partners are more likely to control outcomes in the relationships, suggesting that they most likely do not perceive “an opposing action of incompatibles.” Consequently, they may be unaffected by differences in brand preferences within the relationship. For high-power partners, brand compatibility should not be a source of conflict. On the other hand, previous research has found that low-power partners are not only less likely to express their opinions to others (Berdahl and Martorana 2006), but also more likely to pay attention to the preferences, attitudes, and feelings of their partners (Fiske 1993; Keltner et al. 2003). When brand compatibility is low, low-power partners may therefore be more likely to perceive a “struggle resulting from incompatible or opposing needs, drives, [and] wishes.” They are also more likely to “lose out” to their higher-power partners, which may lead to viewing these decisions as unfair. In order to test whether low-power partners view decisions as unfair when brand compatibility is low, we conducted an online study in which we manipulated perceptions of power (high vs. low) and brand compatibility (high vs. low) and measured perceived fairness as the outcome variable. In line with predictions, individuals in the low-power/low-brand-compatibility condition reported significantly less (at the p < .05 level) perceived fairness with how decisions would be made in the relationship compared with the other three conditions (see the web appendix for additional details regarding this study). As perceived unfairness has been shown to be an underlying driver of relationship conflict (Grote and Clark 2001), we propose that brand compatibility and power interact to influence perceptions of conflict in the relationship. Specifically, we predict that perceptions of conflict will be greatest when brand compatibility and power are both low. We test this hypothesis—that low brand compatibility is more likely to be a source of perceived conflict for low-power partners—in the current study. In keeping with much prior research on power, we utilized a scenario paradigm. Research has shown that manipulating perceptions of power can influence perceptions of other dynamics in social contexts (e.g., high-power individuals have been shown to project their attitudes and feelings on to others; Keltner et al. 2003; Overbeck and Droutman 2013). A scenario method allows us to more cleanly manipulate both perceptions of power and of brand compatibility. In addition, we included a measure of autonomy threat to rule out an alternative process explanation. For parsimony, we present the findings related to conflict here. (Findings related to autonomy threat are provided in the web appendix.) Method Participants Six hundred nine participants (40% men) from Amazon’s Mechanical Turk completed the study in exchange for financial compensation. Participants ranged in age from 18 to 74 years with an average age of 35.43 years (SDage = 11.46). Measures and Procedure After indicating consent, participants were asked to read a scenario and imagine that they were in it. They were told they would be asked questions about the scenario. Participants were randomized to one of four conditions, in which we varied brand compatibility and relationship power, and were presented with a scenario that corresponded to that condition. We manipulated brand compatibility by telling participants they either liked the same or opposing brands as their partner. We manipulated relationship power by including statements from the modified power measure used in the previous studies (Anderson et al. 2012), which we presented as if the participant reported feeling that way about the described relationship. (See the web appendix for the conditions.) While imagining themselves in the scenario, participants were next asked to indicate to what extent they agreed or disagreed with several statements. All participants rated their agreement on a seven-point Likert scale (anchored with Strongly Disagree and Strongly Agree). As our measure of perceived relationship conflict, participants answered to what extent they agreed with the following two items: “My partner and I probably argue a lot” and “My partner and I probably have a lot of conflict” (α = .95). Participants completed other items, including autonomy threat (see the web appendix), filler items, and demographic variables. Results Eight participants indicated they had taken the survey multiple times and were excluded from the following analyses. We hypothesized that participants in the low-brand-compatibility/ low-power condition would report the greatest levels of perceived conflict. In order to test this hypothesis, we conducted an ANOVA with brand compatibility condition, power condition, and their interaction as the predictor variables and conflict as the outcome variable. Results revealed a main effect for brand compatibility condition (F(1, 597) = 30.52, p < .0001), such that the low-brand-compatibility condition reported greater conflict, and a main effect for power (F(1, 597) = 82.37, p < .0001), such that the low-power condition reported greater perceived conflict. Importantly, results also revealed a significant interaction (F(1, 597) = 3.87, p < .05). Planned contrasts indicated that each of the groups was significantly different from the others, with the low-power/low-brand-compatibility condition reporting significantly greater perceived conflict (M = 4.10) than not only the high-power conditions (vs. high-power/low-brand-compatibility condition M = 3.26, t(597) = 4.96, p < .0001; vs. high-power/high-brand-compatibility condition M = 2.47, t(597) = 10.37, p < .0001), but also the low-power/high-brand-compatibility condition (M = 3.68, t(597) = 2.52, p = .01) (figure 5). (Note: Because all of the other studies involve individuals currently in a relationship, while here we asked even those outside of a relationship to imagine themselves in the scenario, we repeated these analyses controlling for relationship status, and the results remain the same.) FIGURE 5 View largeDownload slide RESULTS FROM STUDY 5: PERCEIVED RELATIONSHIP CONFLICT AS A FUNCTION OF MANIPULATED BRAND COMPATIBILITY AND MANIPULATED RELATIONSHIP POWER NOTE.—As predicted, the low-power/low-brand-compatibility condition reported significantly greater perceptions of relationship conflict than the low-power/high-brand-compatibility condition and both of the high-power conditions. FIGURE 5 View largeDownload slide RESULTS FROM STUDY 5: PERCEIVED RELATIONSHIP CONFLICT AS A FUNCTION OF MANIPULATED BRAND COMPATIBILITY AND MANIPULATED RELATIONSHIP POWER NOTE.—As predicted, the low-power/low-brand-compatibility condition reported significantly greater perceptions of relationship conflict than the low-power/high-brand-compatibility condition and both of the high-power conditions. Discussion In a recent article, conflict was predicted by spouses’ differing financial preferences (Rick, Small, and Finkel 2011); extending that work here, we find evidence to support the notion that incompatibility in brand preferences also predicts perceived conflict. These findings suggest that the consistent pattern of evidence across studies demonstrating that low brand compatibility decreases life satisfaction for low-power partners may be related to relationship conflict. In other words, perceived conflict may be the mechanism through which brand compatibility and power interact to influence life satisfaction.4 STUDY 6 In our sixth study, a dyadic study, we again examined brand compatibility, power, and life satisfaction in both members of romantic couples, and we further explore the role of perceived conflict. In line with study 5’s findings, we hypothesized that for low-power partners, low brand compatibility would be associated with increased reports of perceived conflict, given that these individuals are unable to control outcomes, and thus, brand choices likely create some notable problems for their everyday consumption experiences. Previous research has demonstrated that perception of relationship conflict has been shown to predict reductions in general satisfaction (Gordon and Chen 2016; Gustavson et al. 2016). Therefore, we further hypothesized that for low-power individuals, increased perceived conflict would, in turn, drive reduced life satisfaction. Method Participants Study 6 used data from a multiwave study examining romantic relationships conducted at a large public university in Canada. Participants were recruited through advertisements placed in the university newspaper, flyers posted around campus, and presentations in large undergraduate classes. Couples were eligible to participate in the study if both partners indicated they were in an exclusive heterosexual romantic relationship, had been involved for at least six months, and were 18 years of age or older. All participants received financial compensation in exchange for their participation; for completion of the entire multipart longitudinal study, individuals received $100. Participants were both members of romantic relationships (N = 139 couples; Mage = 20.80, SD = 2.16). Reflecting the ethnic diversity of the university campus, participants were majority Caucasian (56.5%) and Asian (33.5%), with 10% reporting mixed or other races. Measures and Procedure Partners took part in a multiwave study with three main components: an initial online battery of questionnaires, an in-lab session with videotaped conversations that occurred one to two weeks later, and then three online follow-up questionnaires, each approximately one month apart, beginning one month after the lab session. Most of the materials and procedures of this study are irrelevant to the current hypothesis, as the study was designed to explore goal dynamics over time and how goals are related to relationship processes (Fitzsimons, Finkel, and vanDellen 2015). All the relevant measures to test our hypothesis about brand compatibility are described here (see the web appendix for any related measures that we did not use in our analyses). During the first online follow-up questionnaire (taken two months after the lab session), we included measures to test the current hypotheses—that is, measures of brand compatibility, relationship power, perceived conflict, and life satisfaction. Because retention is challenging in longitudinal studies, especially ones that require both members of a dyad, we used shortened versions of the relationship power, conflict, and life satisfaction scales. (Note: The measures were completed in that order and the brand compatibility measure was collected after the other variables.) Relationship power was measured by three items taken from the same relationship power measure used in prior studies. The items included: “I can get my partner to do what I want,” “I think I have a great deal of power,” and “My ideas and opinions are often ignored” (reverse-scored) (α = .66). Perceived conflict was measured by five items that measured the degree of conflict partners perceived in their relationship over the course of the past month (Braiker and Kelley 1979). The first two items—“How often did you and your partner argue with each other this month?” and “How often did you feel angry or resentful toward your partner this month?”—were measured on a 1 (never) to 6 (constantly) scale. The other three items—“When you and your partner argued, how serious were the problems or arguments?” “To what extent did you wish you could change things about your partner this month?” and “To what extent did you communicate negative feelings toward your partner (e.g., anger, dissatisfaction, frustration) this month?”—were measured on a 1 (not at all) to 7 (extremely) scale. Items were standardized before being combined into one measure of conflict (α = .83). Life satisfaction was measured with three items taken from the same Diener et al. (1985) measure used in prior studies, “In the last month, in most ways, my life has been close to ideal,” “The conditions of my life were excellent this month,” and “I am satisfied with my life right now,” that were rated on a 1 (strongly disagree) to 7 (strongly agree) scale (Diener et al. 1985; α = .86). As in studies 1 and 4, each partner within a couple was asked to list his/her favorite brand in the same brand categories: coffee, chocolate, car, beer, soda. (In addition, in this study, partners were also asked about their favorite brand of toothpaste. To be consistent, we report the analyses using the same five brand categories as in the other studies. Results remain the same when toothpaste is included.) As in the previous studies, we created a measure of brand compatibility from each pairing of partners’ brand responses. Four undergraduate research assistants rated the brands on the same 1 (not at all compatible) to 5 (completely compatible) scale as used in the previous studies. The coders’ scores were again averaged to create a brand compatibility rating for each brand category (Mbeer = 2.63, SD = 1.46; Mcar = 2.48, SD = 1.16; Mchoc = 2.63, SD = 1.02; Mcoffee = 3.56, SD = 1.52; Msoda = 2.59, SD = 1.26). Raters were highly reliable within each brand category (interrater reliability: beer α = .92; car α = .89; chocolate α = .84; coffee α = .95; soda α = .92). As in the previous studies, brand categories were then averaged to create one mean brand compatibility score. Results In this study, we investigate the role that brand compatibility and power have on conflict, which we predict will in turn affect life satisfaction. Specifically, we hypothesize that for low-power partners, as brand compatibility increases, conflict will decrease. Additionally, we hypothesize that conflict will in turn predict life satisfaction, suggesting that for low-power partners, as brand compatibility increases, conflict decreases and the reduction in conflict is associated with greater life satisfaction. We test our predictions in several steps. First, we examine whether power affects the link between brand compatibility and conflict. Because this is dyadic data, and thus violates assumptions of independence required for standard regression analyses, we again conducted a multilevel analysis (Kenny et al. 2006), with individuals nested within couples. Conflict served as our outcome variable, and power, brand compatibility, and the interaction between them served as our predictor variables. As in the previous study, all of the predictor variables were grand mean-centered. All analyses were conducted using the Mixed Models procedure in SPSS, with all predictors as fixed effects, and power as a level-one (participant-level) predictor and brand compatibility as a level-two (dyad-level) predictor. The results revealed a significant negative association between partners’ reports of power and conflict (B = –.12, t(186.51) = –2.70, p = .008), indicating that as partners felt less power in their relationship, they perceived more conflict. Importantly, as predicted, results revealed a significant interaction between power and brand compatibility on perceived conflict (B = .16, t(181.29) = 2.14, p = .03). Examining the interaction, we find that for high-power partners (+1 SD), brand compatibility is nonsignificantly positively related to conflict (B = .15, t(217.90) = 1.33, p = .19). On the other hand, for low-power partners (–1 SD), brand compatibility is negatively related to perceived conflict (B = –.17, t(218.90) = –1.50, p = .14) (figure 6). Further examining the interaction, we find that low brand compatibility is significantly associated with greater perceived conflict for partners whose power scores are less than 3. FIGURE 6 View largeDownload slide RESULTS FROM STUDY 6: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING PERCEIVED CONFLICT NOTE.—Results from study 6: For high-power (+1 SD) partners, as brand compatibility increases, conflict increases. For low-power (–1 SD) partners, as brand compatibility increases, conflict decreases. FIGURE 6 View largeDownload slide RESULTS FROM STUDY 6: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING PERCEIVED CONFLICT NOTE.—Results from study 6: For high-power (+1 SD) partners, as brand compatibility increases, conflict increases. For low-power (–1 SD) partners, as brand compatibility increases, conflict decreases. Next, we examined whether conflict was related to life satisfaction. A multilevel analysis (with individuals nested within couples) revealed a significant, negative relationship between conflict and life satisfaction (B = –.41, SE = .11, t(243.43) = –3.73, p < .0001), indicating that as conflict in the relationship increased, life satisfaction decreased. In order to test our prediction that an indirect pathway would exist between brand compatibility and power, conflict, and life satisfaction, we conducted an analysis using the Monte Carlo Method for Assessing Mediation (Selig and Preacher 2008). Previous research has indicated that the Monte Carlo method is the better method for assessing indirect effects over the Sobel test and is about as good as nonparametric bootstrapping (Hayes and Scharkow 2013). We used the coefficients and the standard errors from the multilevel model parameters for the interaction term (B = .16, SE = .07) and the mediator term (B = –.40, SE = .11). The 95% confidence interval for the distribution of the indirect effect did not contain zero [–.1396, –.007558], thus supporting the existence of an indirect pathway from brand compatibility and power to life satisfaction through conflict. Following Zhao, Lynch, and Chen (2010), and because this is multilevel data that violates statistical assumptions of independence, we investigated the effect of the predictor variables to the mediator, and the effect of the mediator to the outcome variable. In line with predictions, the reported indirect pathway does support our overall theorizing that the repeated, day-to-day aspect of brand compatibility is driving the increased conflict for low-power partners and leading to the reduction in life satisfaction. However, we also tested the direct pathway on life satisfaction. Although brand compatibility (B = .43, t(118.73) = 3.10, p = .002) and relationship power (B = .27, t(203.53) = 3.27, p = .001) had significant positive effects on life satisfaction, the interaction between brand compatibility and power on life satisfaction was not significant (B = .08, t(200.30) = 0.61, p > .25). To investigate why the direct effect might not have been as strong as in previous studies, we conducted an internal analysis where we examined whether there were differences in relationship type in this study compared to the other studies. In all other studies, participants were older and the couples were likely to be living with their partner (e.g., 100% were living together in study 1, 73% in study 2, 66% in study 3, 63% in study 4). This particular study was run at a university using an undergraduate sample that was much younger (median age 20.0 years) than the other samples. Interestingly, 26% of the couples were in long-distance dating relationships. Given that our proposed theory suggests that for brand compatibility to matter, there should be repeated interactions at the day-to-day level, this may play a role. In an analysis that includes only the couples who are not long distance, the interaction of brand compatibility and power on life satisfaction is marginally significant (B = .29, t(141.03) = –1.89, p = .06); and the effect of the interaction of brand compatibility and power on conflict is significant (B = .29, t(126.99) = 3.72, p < .0001). Of course, this post-hoc speculation needs to be further explored in future research, but it does provide additional insight to our theory. Discussion In our final study, using both members of the couple, we tested our predictions that brand compatibility and power are related to conflict within the relationship. It is important to note that conflict and brand compatibility were not directly correlated (B = –.01, t(122) = –.17, p = .87), suggesting that brand compatibility is not just another measure of conflict in the relationship. Importantly, and in line with our hypotheses, we found that brand compatibility and power significantly interacted to predict conflict. Results from the mediation analysis further suggested the presence of an indirect pathway from the interaction of compatibility and power through conflict to life satisfaction. In other words, high-power partners do not perceive an increase in conflict as a result of low brand compatibility; however, for low-power partners, when brand compatibility is low, conflict is increased, and this increase in perceived conflict is associated with reduced life satisfaction. In addition, in this study, we found evidence in our internal analysis to suggest that it is the repeated day-to-day interactions of couples in terms of brand compatibility and power that is influencing relationship conflict and life satisfaction. GENERAL DISCUSSION This set of studies presents a new construct connecting consumer experiences with close relationships—brand compatibility—and explores its significance for psychological well-being. Across several studies, examining both individuals and couples, we find that brand compatibility within a close relationship predicts life satisfaction, but, importantly, only for individuals who experience low power in their relationship. In other words, the effect of brand compatibility on life satisfaction depends upon power in the relationship. For high-power individuals, brand compatibility is not related to life satisfaction, while for low-power individuals, brand compatibility positively predicts life satisfaction: when brand compatibility is higher, life satisfaction is higher; however, when brand compatibility is low, life satisfaction is reduced. A strength of the studies is that they use different methods to manipulate and measure both brand compatibility and power, allowing us to draw causal conclusions about their roles in predicting life satisfaction, and to provide more assurance of the generalizability of the effects. Brand compatibility was measured as a perception (study 3), was manipulated (studies 2 and 5), and was measured dyadically to obtain a clear and objective indicator of compatibility (studies 1, 4, and 6). Similarly, power was manipulated (studies 3 and 5) and measured (studies 2, 4, and 6). Contributions and Implications Our findings examine how consumer behavior within the context of close relationships affects overall life satisfaction. As close relationships are an integral part of life that predict many meaningful outcomes including mortality rates, financial well-being, health, and happiness (Berkman 1995; Cohen 2004; Keltner et al. 2003; Kiecolt-Glaser, Gouin, and Hantsoo 2010; Kiecolt-Glaser et al. 2005; Liu and Reczek 2012; Marmot 2004; Waite and Gallagher 2002), it is beneficial for consumer researchers to explore links between close relationships and consumer variables. This research makes several contributions to the study of relationships. It is the first to look at how compatibility—of any kind—interacts with power in close relationships. Surprisingly, no research on compatibility in close relationships has examined how it may be more important for some partners than others, instead highlighting the general benefits of compatibility (Bramlett and Mosher 2002; Decuyper, De Bolle, and De Fruyt 2012; Gaunt 2006; Heaton and Pratt 1990; Lucas et al. 2004b; Murray et al. 2002) or questioning the value of compatibility for all partners (Dyrenforth et al. 2010; Houts et al. 1996; Montoya et al. 2008; Tidwell et al. 2013). In this work, we suggest that compatibility is crucial for partners who are low in power, but far less so for those who are high in power. Using the context of brand preferences, we found that compatibility has a predictive influence for individuals who chronically feel low in power and for those who were temporarily led, as a result of an experimental manipulation, to feel low in power. This work thus contributes to prior work on compatibility in close relationships by highlighting the importance of considering other relationship variables, like power, when studying the role of compatibility. Furthermore, the findings suggest that even seemingly trivial forms of compatibility—whether partners report liking the same soda and coffee brands—can affect important psychological phenomena. Life satisfaction is a complex evaluation, but most research studying it has looked at certainly more profound predictors, such as marriage quality, religion, health, self-esteem, job satisfaction, and unemployment (Adams, King, and King 1996; Diener and Diener 2009; Diener et al. 2010; Heller, Watson, and Ilies 2004; Lim and Putnam 2010; Lucas et al. 2004a; Mroczek and Spiro 2005). Here we show that for some people—those low in power—even something as seemingly mundane as brand compatibility within a close relationship influences people’s satisfaction with their lives. Although we studied these ideas only within the brand context, we suggest that they may also be worth studying with other low-level forms of day-to-day compatibility, such as activity, spending, and parenting preferences. If couples are incompatible on dimensions that matter in day-to-day life, our findings here suggest that low-power partners will suffer as a result. In taking a dyadic perspective on consumers’ preference for brands, the studies begin to answer calls for a more social and interpersonal perspective on consumer behavior (Simpson et al. 2012), and for the study of power in consumer behavior more specifically (Brick and Fitzsimons 2016; Rucker et al. 2012). Although research on social power has long been a large and influential topic in psychology, organizational behavior, sociology, and other fields in social science (see Magee and Galinsky 2008 for a review), power in the consumer context has historically received less attention. As Rucker and colleagues noted, “Despite the long-recognized value and experimental study of power across the social sciences, the construct of power has been largely absent from efforts to understand consumer behavior. This is somewhat surprising given that different degrees of power exist and arise in consumers’ everyday activities” (Rucker et al. 2012). Indeed, recent research supports the relevance of experiences of social and interpersonal power for consumer behavior (Jiang et al. 2014; Kim and McGill 2011; Rucker and Galinsky 2008). Our findings thus contribute to the literature on power, and more specifically relationship power, within consumer behavior. This research also has implications in terms of brand loyalty. Individuals’ brand preferences may shift as a function of their close relationships and close relationship status. One could imagine being loyal to particular brands when one is single, and shifting or changing those brand preferences once one enters into a relationship. In addition, as individuals change their relationship status, moving from single to engaged to married, brand preferences and brand loyalties may change. It would be interesting to examine how an individual’s brand preferences shift over the course of his or her lifetime as a function of relationship status. Furthermore, what happens when an individual ends a close, romantic relationship if he/she has shifted brand preferences? Does the individual go back to using the brands he/she preferred from before the relationship, or continue using the brands that he/she used while in a relationship? On a practical note, the current findings have implications for matchmaking, including online dating services. Online dating is now a major multibillion-dollar industry. For example, IAC/InterActiveCorp., the company that owns Match.com, reported over $1.2 billion in revenue last year for its dating websites alone (JMP Securities Initiation Report, http://ir.iac.com/results.cfm; see Supplemental Financial Information and Metrics). Recent research has highlighted the ways in which online dating differs from conventional offline dating (Finkel et al. 2012), and taking consumer preference information into account may be another way. Very few, if any, of these online dating websites ask about everyday consumer preferences and use this information in their matchmaking algorithms. However, our research suggests that one aspect of consumer preferences, namely brand preferences, can affect relationship conflict and well-being. We are not suggesting that brand preferences are the only form of consumer behavior that matchmakers should take into account, but the present research suggests that when designing future questions and surveys, matchmakers, including online services, should consider including items to assess consumer behavior at the day-to-day level, such as brand preferences, to better meet their customers’ needs. For those who will ultimately be high in power, consumer preferences like brand compatibility may be irrelevant, but for those who will ultimately be low in power, it would be beneficial for them to consider. Limitations and Future Research There are several limitations to the current work. One important limitation is that we only ask about brand preferences and do not specifically examine brand use. A potential problem with this could be that couples trade off or engage in multibrand use within the household, and thus their preferences really do not matter for day-to-day life. To explore this, we conducted a simple online study in which we asked individuals who live with their partner to count the total number of brands that are currently in their household across various product categories. Specifically, 406 participants (45% men) were given the same definition of a brand as used throughout the current research. They then counted and listed each of the brands currently in or used by their household within the following product categories: soda, coffee, beer, chocolate, car, toothpaste, toilet paper, laundry detergent, grocery store, orange juice, ketchup, and paper towels. The mean number of brands within each product category was 1.06 (SD = 0.27). The product categories with the highest mean were grocery store (M = 1.83, SD = 0.71) and car (M = 1.54, SD = 0.73), and for all categories except grocery store, the modal response was 0 or 1. Overall, this survey suggests that, on average, at least for the product categories studied, couples tend to purchase at most one brand for the household. Because couples are generally choosing just one brand for the household, brand preferences do matter. Over time, incompatible brand preferences may indeed be a source of conflict for couples, particularly if one partner is continually getting his or her way at the expense of the other partner. Another limitation of the current work may be the robustness of the results. In a couple of studies, the p values were relatively weak. Thus, readers may be concerned about the strength of the evidence for the key interaction. Readers can be reassured that all reporting was fully transparent, as explained in the introduction. All measures and conditions were reported (with exceptions noted), and all attempts were made to follow best-practice guidelines for transparency in research collection and reporting. Because of the relatively weak p values, we also sought additional evidence in the form of a direct replication (study 2B). Despite these efforts, it is reasonable to question the strength of the evidence provided here. We acknowledge this as a potential limitation of the current work. That said, we do believe that the consistent pattern of findings across studies, using multiple methods, including measuring and manipulating brand compatibility and power using various coding schemes and analyses, supports our hypothesis. Further, we hope that our novel findings encourage future research on this important topic. Finally, because it is difficult to recruit both partners of a couple to come to the lab together, we used data in studies 4 and 6 from multistudy events examining romantic relationships. Using data from multistudy events limits control over the order of administration of focal measures. In addition, it also increases the number of other measures to which participants are exposed. One direction for future research could be to further study couples who are both low-low or high-high in relationship power. In our research, we focused on perceptions of relationship power. Accordingly, and as discussed in study 4, this means that within a couple there can be two members who are high-high in power and two members who are low-low in power. What does it mean for both partners to be high in perceptions of relationship power? On the one hand, it could mean that one partner is oblivious. On the other hand, if power is defined as the ability to influence others (Anderson and Galinsky 2006; Emerson 1962; Keltner et al. 2003; Magee and Galinsky 2008), it suggests that both partners believe that they have control in the relationship. If both partners perceive control, or power, it suggests the relationship is an equitable one, and close relationships research has highlighted that the most satisfying relationships are indeed those viewed as equitable (Gray-Little and Burks 1983). Perhaps, then, couples who are high-high in relationship power are happy because they believe they have equal control in the relationship, regardless of the actual breakdown and perhaps regardless of the specific domain. Future research could also investigate whether couples who are high-high in relationship power pursue complementary goal strategies. In other words, do they alternate control across domains that are of differing importance or do they alternate within a domain? In terms of low-low relationships, there were very few couples who were low-low power, and we hesitate to overinterpret this pattern. However, this finding does suggest another area for future research. Overall, our results indicate that perceived power in relationships is not always zero-sum, which is also an interesting area for future research. Future research could also further investigate the role of relationship power in consumer behavior. Recent research has demonstrated that power influences purchasing decisions (Rucker et al. 2012; Rucker, Hu, and Galinsky 2014). For example, Jiang et al. (2014) found that increased perceptions of power result in greater switching behavior. This research, like much of the previous research, has focused on the level of the individual, whereas the present research focuses on partners within romantic relationships. Thus, it would be interesting to examine the role of power in relationships and switching behavior. Although the previous research has found that high-power individuals are more likely to engage in switching behavior (Jiang et al. 2014), a close relationships perspective suggests that low-power partners may be more likely to switch and adopt their high-power partners’ preferences. Longitudinal research could examine how brand preferences shift over the course of a relationship. A related area for future research could examine how brand preferences change over the course of one’s lifetime as a result of relationship status and power. Previous research has indicated that high-power individuals tend to have more stable self-concepts over time (Kraus, Chen, and Keltner 2011), and so it might be that individuals who tend to have power in relationships do not shift their brand preferences. However, for those who generally yield power to their partners, perhaps their brand preferences shift back to the brands they preferred before they entered into the relationship. Alternatively, maybe they do not change brand preferences until they enter into a new relationship. Future research could examine this topic and how it relates to brand loyalty over a consumer’s lifetime. The manner in which consumers make decisions with other people can be different from the manner in which consumers make decisions by themselves (Park 1982). The current research supports this idea and suggests that the brands people consume when they are with their romantic partner may be different from ones that they select when they are alone, especially for low-power partners. One question that arises is whether evaluations of brands change when a partner, especially a high-power partner, is present. Would partners who are low power, and who generally prefer a different brand, change their evaluations to be in line with their partner’s views when their partner is present? In other words, would lower-power partners “go along to get along”? Previous research suggests that they would (Fiske 1993; Keltner et al. 2003; Simpson et al. 2014), and our research suggests that they may subsequently be less happy. Another question is how do brand compatibility and relationship power interact to influence consumers’ brand evaluations depending upon whether only one or both partners like the brand? Future research could tease apart differences in brand preferences, evaluations, and selections when low-power individuals are with versus without their partners, and whether the decision itself is shared or made by one member of the couple. Conclusion Individuals often use and consume brands in the presence of their close partners. Yet very little is known about the role that brand preferences play in close relationships. In the present research, we seek to provide a preliminary examination of this topic by examining how brand compatibility, or the extent to which individuals have similar brand preferences as their partners, influences life satisfaction. For low-power partners, it seems that brand compatibility can reliably predict how satisfied they feel with their lives. For those who tend to yield power to their romantic partners, the brand preferences of their partners take on a rather surprising significance, suggesting the importance of brand preferences in everyday life. If you tend to yield power to your romantic partners, perhaps you should first ensure that they share your love of Coke, or you may find yourself drinking Pepsi yet again. DATA COLLECTION INFORMATION For each study involving both members of the couple, sample sizes were based on subject availability as well as the needs of unrelated research projects that were run in conjunction with data collection. For studies using online participants, sample sizes were determined a priori. No additional data were collected after data analyses began with the exception of the additional conditions for study 2 as requested by a reviewer in the previous round of revisions. The data from study 1 were collected from a local farmer’s market in November 2015 by research assistants under the supervision of the first author. Data for studies 2 and 3 were collected from Amazon’s Mechanical Turk in January 2015 by the first author. Data from replication study 2 were collected in October 2016. The couple data from study 4 were collected during two waves of data collection from April 2014 to November 2014 under the supervision of a lab manager and research assistants. Data for study 5 and the introductory study to study 5 were collected from Amazon’s Mechanical Turk in January 2016 and September 2016. Data from study 6 were collected by research assistants under the supervision of the second author as part of a multiwave study examining romantic relationships. The data for studies 2, 3, and 5 were analyzed by the first author. The data for studies 1, 4, and 6 were analyzed primarily by the first author with support from the second author. The authors are grateful for the constructive comments and helpful suggestions provided by the editor, associate editor, and three anonymous reviewers. In addition, they would like to thank Jim Bettman for his invaluable feedback. APPENDIX Relationship Power Measure Participants are asked to indicate to what extent they agree with the following statements in regard to their relationship. Items are rated on a seven-point Likert scale anchored with “Strongly Disagree” and “Strongly Agree.” Items marked with an * are reverse-scored. I can get my partner to listen to what I say. My wishes do not carry much weight.* I can get my partner to do what I want. Even if I voice them, my views have little sway.* I think I have a great deal of power. My ideas and opinions are often ignored.* Even when I try, I am not able to get my way.* If I want to, I get to make the decisions. Footnotes 1 Some may wonder why we oriented compatibility around brand family (e.g., Coke products) rather than around attribute level similarity (e.g., diet soda). We suggest both a conceptual and an empirical response to this important question. From a conceptual point of view, research has demonstrated that individuals form relationships and attachments with brands themselves—that is, brand families (Aaker, Fournier, and Brasel 2004; Aggarwal 2004; Carroll and Ahuvia 2006; Fournier 1998; Park et al. 2010; Thomson et al. 2005) as opposed to attributes. In addition, many structural factors and environmental cues lead individuals to cognitively organize products around brand families. For example, packaging elements are often consistent across products within brand families (the shape of the bottle for Coca-Cola products; the color scheme across all Colgate toothpastes, etc.); most product categories in grocery stores are organized by brand family as opposed to attribute, even more so in categories where distribution is done by the brand directly (e.g., Lindt’s dark chocolate is next to Lindt’s white chocolate, Pepsi products are grouped together separately from Coke products); and promotional efforts within stores often operate at the level of the brand family rather than at an attribute level (e.g., “buy three for $12”–style offers or “buy two, get one free” offers). From an empirical perspective, note that we employed alternative coding approaches across the studies, one of which gave no specific instructions in terms of how the coders should define brand compatibility. The coders simply rated which brands they thought were compatible with which other brands. This coding approach yielded extremely similar results to our original compatibility coding as well as to the specific coding scheme that gave individual examples within each product category. If people naturally judged brand compatibility, or similarity, according to product attributes, this unstructured approach should have yielded a different pattern of results from the other two approaches. That was not the case—results were consistent across coding schemes. 2 We repeated the analyses using various coding schemes to ensure that the results are not driven by the use of any one specific type of coding scheme. All results are similar. See the web appendix for additional models. 3 To examine the reliability of these effects, we conducted a direct replication of study 2. The findings replicated. The full results are reported in the web appendix. 4 We conducted an additional experimental study to further investigate the role of conflict in current romantic relationships depending upon brand preferences and relationship power. The results of this additional study revealed that, unlike their higher-power counterparts, lower-power partners reported greater conflict in their current relationship over which brands to buy. Please see the web appendix for additional information regarding this study. REFERENCES Aaker David A., Keller Kevin Lane ( 1990), “Consumer Evaluations of Brand Extensions,”  Journal of Marketing , 27– 41. Aaker Jennifer, Fournier Susan, Brasel S. Adam ( 2004), “When Good Brands Do Bad,” Journal of Consumer Research  31 1, 1– 16. Google Scholar CrossRef Search ADS   Acitelli Linda K, Douvan Elizabeth, Veroff Joseph ( 1993), “Perceptions of Conflict in the First Year of Marriage: How Important Are Similarity and Understanding?” Journal of Social and Personal Relationships , 10 1, 5– 19. Google Scholar CrossRef Search ADS   Adams Gary A., King Lynda A., King Daniel W. ( 1996), “Relationships of Job and Family Involvement, Family Social Support, and Work–Family Conflict with Job and Life Satisfaction,” Journal of Applied Psychology , 81 4, 411. Google Scholar CrossRef Search ADS   Aggarwal Pankaj ( 2004), “The Effects of Brand Relationship Norms on Consumer Attitudes and Behavior,” Journal of Consumer Research , 31 1, 87– 101. Google Scholar CrossRef Search ADS   Aiken Leona S., West Stephen G. ( 1991), Multiple Regression: Testing and Interpreting Interactions , Thousand Oaks, CA: Sage. Aknin Lara B., Barrington-Leigh Christopher P., Dunn Elizabeth W., Helliwell John F., Burns Justine, Biswas-Diener Robert, Kemeza Imelda, Nyende Paul, Ashton-James Claire E., Norton Michael I. ( 2013), “Prosocial Spending and Well-Being: Cross-Cultural Evidence for a Psychological Universal,” Journal of Personality and Social Psychology , 104 4, 635. Google Scholar CrossRef Search ADS PubMed  Anderson Cameron, Galinsky Adam D. ( 2006), “Power, Optimism, and Risk‐Taking,” European Journal of Social Psychology , 36 4, 511– 36. Google Scholar CrossRef Search ADS   Anderson Cameron, John Oliver P., Keltner Dacher ( 2012), “The Personal Sense of Power,” Journal of Personality , 80 2, 313– 44. Google Scholar CrossRef Search ADS PubMed  Arrindell W. A., Meeuwesen L., Huyse F. J. ( 1991), “The Satisfaction with Life Scale (Swls): Psychometric Properties in a Non-Psychiatric Medical Outpatients Sample,” Personality and Individual Differences , 12 2, 117– 23. Google Scholar CrossRef Search ADS   Bagchi Rajesh, Cheema Amar ( 2013), “The Effect of Red Background Color on Willingness-to-Pay: The Moderating Role of Selling Mechanism,”  Journal of Consumer Research,   39 5, 947– 60. Google Scholar CrossRef Search ADS   Belk Russell ( 1988), Possessions and Self, Hoboken , NJ: Wiley Online Library. Berdahl Jennifer L., Martorana Paul ( 2006), “Effects of Power on Emotion and Expression During a Controversial Group Discussion,” European Journal of Social Psychology , 36 4, 497. Google Scholar CrossRef Search ADS   Berger Jonah, Heath Chip ( 2007), “Where Consumers Diverge from Others: Identity Signaling and Product Domains,” Journal of Consumer Research , 34 2, 121– 34. Google Scholar CrossRef Search ADS   Berkman Lisa F. ( 1995), “The Role of Social Relations in Health Promotion,” Psychosomatic Medicine , 57 3, 245– 54. Google Scholar CrossRef Search ADS PubMed  Berscheid Ellen, Snyder Mark, Omoto Allen M. ( 1989), “The Relationship Closeness Inventory: Assessing the Closeness of Interpersonal Relationships,” Journal of Personality and Social Psychology , 57 5, 792. Google Scholar CrossRef Search ADS   Blossfeld Hans-Peter ( 2009), “Educational Assortative Marriage in Comparative Perspective,” Annual Review of Sociology , 35, 513– 30. Google Scholar CrossRef Search ADS   Blum Jeffrey S., Mehrabian Albert ( 1999), “Personality and Temperament Correlates of Marital Satisfaction,” Journal of Personality , 67 1, 93– 125. Google Scholar CrossRef Search ADS   Braiker Harriet B., Kelley Harold H. ( 1979), “Conflict in the Development of Close Relationships,” in Social Exchange in Developing Relationships , ed. Burgess Robert L., Huston Ted L., Cambridge, MA: Academic Press, 135– 68. Google Scholar CrossRef Search ADS   Bramlett Matthew D., Mosher William D. ( 2002), “Cohabitation, Marriage, Divorce, and Remarriage in the United States,” Vital Health Statistics , 23 22, 1– 87. Brick Danielle J., Fitzsimons Gavan J. ( 2016), “Oppositional Brand Choice: Using Brands to Respond to Relationship Frustration,” Journal of Consumer Psychology , 27 2, 257– 63. Google Scholar CrossRef Search ADS   Burroughs James E., Rindfleisch Aric ( 2002), “Materialism and Well-Being: A Conflicting Values Perspective,” Journal of Consumer Research , 29 3, 348– 70. Google Scholar CrossRef Search ADS   Byrne Donn Erwin ( 1971), The Attraction Paradigm , Vol. 11, Cambridge, MA: Academic Press. Carroll Barbara A., Ahuvia Aaron C. ( 2006), “Some Antecedents and Outcomes of Brand Love,” Marketing Letters , 17 2, 79– 89. Google Scholar CrossRef Search ADS   Clee Mona A., Wicklund Robert A. ( 1980), “Consumer Behavior and Psychological Reactance,” Journal of Consumer Behavior , 6 4, 389– 405. Google Scholar CrossRef Search ADS   Cohen Sheldon ( 2004), “Social Relationships and Health,” American Psychologist , 59 8, 676. Google Scholar CrossRef Search ADS PubMed  Cohn Michael A., Fredrickson Barbara L., Brown Stephanie L., Mikels Joseph A., Conway Anne M. ( 2009), “Happiness Unpacked: Positive Emotions Increase Life Satisfaction by Building Resilience,” Emotion , 9 3, 361. Google Scholar CrossRef Search ADS PubMed  Corfman Kim P., Lehmann Donald R. ( 1987), “Models of Cooperative Group Decision-Making and Relative Influence: An Experimental Investigation of Family Purchase Decisions,” Journal of Consumer Research , 14 1, 1– 13. Google Scholar CrossRef Search ADS   Cromwell Ronald E., Olsen David H. ( 1975), Power in Families , Thousand Oaks, CA: Sage. Decuyper Mieke, De Bolle Marleen, De Fruyt Filip ( 2012), “Personality Similarity, Perceptual Accuracy, and Relationship Satisfaction in Dating and Married Couples,” Personal Relationships , 19 1, 128– 45. Google Scholar CrossRef Search ADS   Diener Ed ( 2012), “New Findings and Future Directions for Subjective Well-Being Research,” American Psychologist , 67 8, 590. Google Scholar CrossRef Search ADS PubMed  Diener Ed, Biswas-Diener Robert ( 2002), “Will Money Increase Subjective Well-Being?” Social Indicators Research , 57 2, 119– 69. Google Scholar CrossRef Search ADS   Diener Ed, Diener Marissa ( 2009), “Cross-Cultural Correlates of Life Satisfaction and Self-Esteem,” in Culture and Well-Being , ed. Diener Ed, New York: Springer, 71– 91. Google Scholar CrossRef Search ADS   Diener Ed, Emmons Robert A., Larsen Randy J., Griffin Sharon ( 1985), “The Satisfaction with Life Scale,” Journal of Personality Assessment , 49 1, 71– 5. Google Scholar CrossRef Search ADS PubMed  Diener Ed, Ng Weiting, Harter James, Arora Raksha ( 2010), “Wealth and Happiness across the World: Material Prosperity Predicts Life Evaluation, Whereas Psychosocial Prosperity Predicts Positive Feeling,” Journal of Personality and Social Psychology , 99 1, 52. Google Scholar CrossRef Search ADS PubMed  Dyrenforth Portia S., Kashy Deborah A., Brent Donnellan M., Lucas Richard E. ( 2010), “Predicting Relationship and Life Satisfaction from Personality in Nationally Representative Samples from Three Countries: The Relative Importance of Actor, Partner, and Similarity Effects,” Journal of Personality and Social Psychology , 99 4, 690. Google Scholar CrossRef Search ADS PubMed  Dzhogleva Hristina, Lamberton Cait Poynor ( 2014), “Should Birds of a Feather Flock Together? Understanding Self-Control Decisions in Dyads,” Journal of Consumer Research , 41 2, 361– 80. Google Scholar CrossRef Search ADS   Emerson Richard M ( 1962), “Power-Dependence Relations,” American Sociological Review , 31– 41. Emmons Robert A. ( 1986), “Personal Strivings: An Approach to Personality and Subjective Well-Being,” Journal of Personality and Social Psychology , 51 5, 1058. Google Scholar CrossRef Search ADS   Emmons Robert A., King Laura A. ( 1988), “Conflict among Personal Strivings: Immediate and Long-Term Implications for Psychological and Physical Well-Being,” Journal of Personality and Social Psychology , 54 6, 1040. Google Scholar CrossRef Search ADS PubMed  Escalas Jennifer Edson, Bettman James R. ( 2003), “You Are What They Eat: The Influence of Reference Groups on Consumers’ Connections to Brands,” Journal of Consumer Psychology , 13 3, 339– 48. Google Scholar CrossRef Search ADS   Escalas Jennifer Edson, Bettman James R. ( 2005), “Self-Construal, Reference Groups, and Brand Meaning,” Journal of Consumer Research , 32 3, 378– 89. Google Scholar CrossRef Search ADS   Filiatrault Pierre, Ritchie J. R. Brent ( 1980), “Joint Purchasing Decisions: A Comparison of Influence Structure in Family and Couple Decision-Making Units,” Journal of Consumer Research , 7 2, 131– 40. Google Scholar CrossRef Search ADS   Finkel Eli J., Eastwick Paul W., Karney Benjamin R., Reis Harry T., Sprecher Susan ( 2012), “Online Dating a Critical Analysis from the Perspective of Psychological Science,” Psychological Science in the Public Interest , 13 1, 3– 66. Google Scholar CrossRef Search ADS   Fiske Susan T. ( 1993), “Controlling Other People: The Impact of Power on Stereotyping,” American Psychologist , 48 6, 621. Google Scholar CrossRef Search ADS PubMed  Fitzsimons Gavan J. ( 2000), “Consumer Response to Stockouts,” Journal of Consumer Research , 27 2, 249– 66. Google Scholar CrossRef Search ADS   Fitzsimons Gráinne M., Finkel Eli J., vanDellen Michelle R. ( 2015), “Transactive Goal Dynamics,”  Psychological Review,   122 4, 648. Google Scholar CrossRef Search ADS   Fournier Susan ( 1998), “Consumers and Their Brands: Developing Relationship Theory in Consumer Research,”  Journal of Consumer Research,   24 4, 343– 53. Google Scholar CrossRef Search ADS   French John R. P., Raven Bertram ( 1959), “The Bases of Social Power,” in Classics of Organization Theory , ed. Shafritz Jay M., Whitbeck Philip H., Oak Park, IL: Moore Publishing, 311– 20. Fredrickson Barbara L., Joiner Thomas ( 2002), “Positive Emotions Trigger Upward Spirals toward Emotional Well-Being,” Psychological Science , 13 2, 172– 5. Google Scholar CrossRef Search ADS PubMed  Galinsky Adam D., Gruenfeld Deborah H., Magee Joe C. ( 2003), “From Power to Action,” Journal of Personality and Social Psychology , 85 3, 453. Google Scholar CrossRef Search ADS PubMed  Galinsky Adam D., Magee Joe C., Inesi M. Ena, Gruenfeld Deborah H. ( 2006), “Power and Perspectives Not Taken,” Psychological Science , 17 12, 1068– 74. Google Scholar CrossRef Search ADS PubMed  Gaunt Ruth ( 2006), “Couple Similarity and Marital Satisfaction: Are Similar Spouses Happier?” Journal of Personality , 74 5, 1401– 20. Google Scholar CrossRef Search ADS PubMed  Glicksohn Joseph, Golan Hilla ( 2001), “Personality, Cognitive Style, and Assortative Mating,” Personality and Individual Differences , 30 7, 1199– 1209. Google Scholar CrossRef Search ADS   Gonzaga Gian C., Campos Belinda, Bradbury Thomas ( 2007), “Similarity, Convergence, and Relationship Satisfaction in Dating and Married Couples,” Journal of Personality and Social Psychology , 93 1, 34. Google Scholar CrossRef Search ADS PubMed  Gordon Amie M., Chen Serena ( 2016), “Do You Get Where I’m Coming From? Perceived Understanding Buffers against the Negative Impact of Conflict on Relationship Satisfaction,” Journal of Personality and Social Psychology , 110 2, 239. Google Scholar CrossRef Search ADS PubMed  Gray-Little Bernadette, Burks Nancy ( 1983), “Power and Satisfaction in Marriage: A Review and Critique,” Psychological Bulletin , 93 3, 513. Google Scholar CrossRef Search ADS   Grote Nancy K., Clark Margaret S. ( 2001), “Perceiving Unfairness in the Family: Cause or Consequence of Marital Distress?” Journal of Personality and Social Psychology , 80 2, 281– 93. Google Scholar CrossRef Search ADS PubMed  Gustavson Kristin, Røysamb Espen, Borren Ingrid, Torvik Fartein Ask, Karevold Evalill ( 2016), “Life Satisfaction in Close Relationships: Findings from a Longitudinal Study,”  Journal of Happiness Studies   17 3, 1293– 1311. Google Scholar CrossRef Search ADS   Hayes Andrew F., Scharkow Michael ( 2013), “The Relative Trustworthiness of Inferential Tests of the Indirect Effect in Statistical Mediation Analysis: Does Method Really Matter?” Psychological Science  24 10, 1918– 27. Google Scholar CrossRef Search ADS PubMed  Heaton Tim B., Pratt Edith L. ( 1990), “The Effects of Religious Homogamy on Marital Satisfaction and Stability,” Journal of Family Issues , 11 2, 191– 207. Google Scholar CrossRef Search ADS   Heller Daniel, Watson David, Ilies Remus ( 2004), “The Role of Person versus Situation in Life Satisfaction: A Critical Examination,” Psychological Bulletin , 130 4, 574. Google Scholar CrossRef Search ADS PubMed  Hoeffler Steve, Keller Kevin Lane ( 2003), “The Marketing Advantages of Strong Brands,” Journal of Brand Management  10 6, 421– 45. Google Scholar CrossRef Search ADS   Holt Douglas B. ( 2002), “Why Do Brands Cause Trouble? A Dialectical Theory of Consumer Culture and Branding,” Journal of Consumer Research , 29 1, 70– 90. Google Scholar CrossRef Search ADS   Houts Renate M., Robins Elliot, Huston Ted L. ( 1996), “Compatibility and the Development of Premarital Relationships,” Journal of Marriage and the Family , 58 1, 7– 20. Google Scholar CrossRef Search ADS   Jiang Yuwei, Zhan Lingjing, Rucker Derek D. ( 2014), “Power and Action Orientation: Power as a Catalyst for Consumer Switching Behavior,” Journal of Consumer Research , 41 1, 183– 96. Google Scholar CrossRef Search ADS   Kelley Harold H. ( 1979), Personal Relationships: Their Structures and Processes , Hillsdale, NJ: Lawrence Erlbaum Associates. Keller Kevin Lane ( 1993) “Conceptualizing, Measuring, and Managing Customer-based Brand Equity,” Journal of Marketing , 57 1, 1– 22. Google Scholar CrossRef Search ADS   Keltner Dacher, Gruenfeld Deborah H., Anderson Cameron ( 2003), “Power, Approach, and Inhibition,” Psychological Review , 110 2, 265. Google Scholar CrossRef Search ADS PubMed  Kenny David A., Kashy Deborah A., Cook William L. ( 2006), Dyadic Data Analysis , New York: Guilford. Kiecolt-Glaser Janice K., Gouin Jean-Philippe, Hantsoo Liisa ( 2010), “Close Relationships, Inflammation, and Health,” Neuroscience and Biobehavioral Reviews , 35 1, 33– 8. Google Scholar CrossRef Search ADS PubMed  Kiecolt-Glaser Janice K., Loving Timothy J., Stowell Jeffrey R., Malarkey William B., Lemeshow Stanley, Dickinson Stephanie L., Glaser Ronald ( 2005), “Hostile Marital Interactions, Proinflammatory Cytokine Production, and Wound Healing,” Archives of General Psychiatry , 62 12, 1377– 84. Google Scholar CrossRef Search ADS PubMed  Kim Sara, McGill Ann L. ( 2011), “Gaming with Mr. Slot or Gaming the Slot Machine? Power, Anthropomorphism, and Risk Perception,” Journal of Consumer Research , 38 1, 94– 107. Google Scholar CrossRef Search ADS   Kirchler Erich ( 1993), “Spouses’ Joint Purchase Decisions: Determinants of Influence Tactics for Muddling through the Process,” Journal of Economic Psychology , 14 2, 405– 38. Google Scholar CrossRef Search ADS   Kraus Michael W., Chen Serena, Keltner Dacher ( 2011), “The Power to Be Me: Power Elevates Self-Concept Consistency and Authenticity,” Journal of Experimental Social Psychology , 47 5, 974– 80. Google Scholar CrossRef Search ADS   Levinger George, Breedlove James ( 1966), “Interpersonal Attraction and Agreement: A Study of Marriage Partners,”  Journal of Personality and Social Psychology,   3 4, 367. Google Scholar CrossRef Search ADS   Levinger George, Rands Marylyn ( 1985), “Compatibility in Marriage and Other Close Relationships,” in Compatible and Incompatible Relationships , New York: Springer, 309– 31. Google Scholar CrossRef Search ADS   Lim Chaeyoon, Putnam Robert D. ( 2010), “Religion, Social Networks, and Life Satisfaction,” American Sociological Review , 75 6, 914– 33. Google Scholar CrossRef Search ADS   Little Anthony C., Perrett David I. ( 2002), “Putting Beauty Back in the Eye of the Beholder,” The Psychologist  15, 28– 32. Liu Hui, Reczek Corinne ( 2012), “Cohabitation and US Adult Mortality: An Examination by Gender and Race,” Journal of Marriage and Family , 74 4, 794– 811. Google Scholar CrossRef Search ADS   Lowe Michael L., Haws Kelly L. ( 2014), “(Im) Moral Support: The Social Outcomes of Parallel Self-Control Decisions,” Journal of Consumer Research , 41 2, 489– 505. Google Scholar CrossRef Search ADS   Lucas Richard E., Clark Andrew E., Georgellis Yannis, Diener Ed ( 2004a), “Unemployment Alters the Set Point for Life Satisfaction,” Psychological Science , 15 1, 8– 13. Google Scholar CrossRef Search ADS   Lucas Todd W., Wendorf Craig A., Imamoglu E. Olcay, Shen Jiliang, Parkhill Michele R., Weisfeld Carol C., Weisfeld Glenn E. ( 2004b), “Marital Satisfaction in Four Cultures as a Function of Homogamy, Male Dominance, and Female Attractiveness,” Sexualities, Evolution & Gender , 6 ( 2–3), 97– 130. Google Scholar CrossRef Search ADS   Luhmann Maike, Hofmann Wilhelm, Eid Michael, Lucas Richard E. ( 2012), “Subjective Well-Being and Adaptation to Life Events: A Meta-Analysis,” Journal of Personality and Social Psychology , 102 3, 592. Google Scholar CrossRef Search ADS PubMed  Luo Shanhong, Klohnen Eva C ( 2005), “Assortative Mating and Marital Quality in Newlyweds: A Couple-Centered Approach,” Journal of Personality and Social Psychology , 88 2, 304. Google Scholar CrossRef Search ADS PubMed  Luo Xueming ( 2005), “How Does Shopping with Others Influence Impulsive Purchasing?” Journal of Consumer Psychology , 15 4, 288– 94. Google Scholar CrossRef Search ADS   Magee Joe C., Galinsky Adam D. ( 2008), “8 Social Hierarchy: The Self‐Reinforcing Nature of Power and Status,” Academy of Management Annals , 2 1, 351– 98. Google Scholar CrossRef Search ADS   Magee Joe C., Galinsky Adam D., Gruenfeld Deborah H. ( 2007), “Power, Propensity to Negotiate, and Moving First in Competitive Interactions,” Personality and Social Psychology Bulletin  33 2, 200– 12. Google Scholar CrossRef Search ADS PubMed  Malhotra Naresh K. ( 1988), “Self Concept and Product Choice: An Integrated Perspective,” Journal of Economic Psychology , 9 1, 1– 28. Google Scholar CrossRef Search ADS   Mare Robert D. ( 1991), “Five Decades of Educational Assortative Mating,” American Sociological Review , 15– 32. Marmot Michael G. ( 2004), The Status Syndrome: How Social Standing Affects Our Health and Longevity , London: Macmillan. Martin Kelly D., Paul Hill Ronald ( 2012), “Life Satisfaction, Self-Determination, and Consumption Adequacy at the Bottom of the Pyramid,” Journal of Consumer Research , 38 6, 1155– 68. Google Scholar CrossRef Search ADS   Merriam-Webster.com, “Conflict,” http://www.merriam-webster.com/dictionary/conflict. Montoya R. Matthew, Horton Robert S., Kirchner Jeffrey ( 2008), “Is Actual Similarity Necessary for Attraction? A Meta-Analysis of Actual and Perceived Similarity,” Journal of Social and Personal Relationships , 25 6, 889– 922. Google Scholar CrossRef Search ADS   Mourali Mehdi, Yang Zhiyong ( 2013), “The Dual Role of Power in Resisting Social Influence,” Journal of Consumer Research , 40 3, 539– 54. Google Scholar CrossRef Search ADS   Mroczek Daniel K., Spiro AvronIII ( 2005), “Change in Life Satisfaction During Adulthood: Findings from the Veterans Affairs Normative Aging Study,” Journal of Personality and Social Psychology , 88 1, 189. Google Scholar CrossRef Search ADS PubMed  Muniz Albert M.Jr., O’Guinn Thomas C. ( 2001), “Brand Community,” Journal of Consumer Research , 27 4, 412– 32. Google Scholar CrossRef Search ADS   Murray Sandra L., Holmes John G., Bellavia Gina, Griffin Dale W., Dolderman Dan ( 2002), “Kindred Spirits? The Benefits of Egocentrism in Close Relationships,” Journal of Personality and Social Psychology , 82 4, 563. Google Scholar CrossRef Search ADS PubMed  Ng Sharon, Houston Michael J. ( 2006), “Exemplars or Beliefs? The Impact of Self-View on the Nature and Relative Influence of Brand Associations,” Journal of Consumer Research , 32 4, 519– 29. Google Scholar CrossRef Search ADS   Overbeck Jennifer R., Droutman Vitaliya ( 2013), “One for All Social Power Increases Self-Anchoring of Traits, Attitudes, and Emotions,” Psychological Science , 24 8, 1466– 76. Google Scholar CrossRef Search ADS PubMed  Park C. Whan ( 1982), “Joint Decisions in Home Purchasing: A Muddling-through Process,” Journal of Consumer Research , 9 2, 151– 62. Google Scholar CrossRef Search ADS   Park C. Whan, MacInnis Deborah J., Priester Joseph, Eisingerich Andreas B., Iacobucci Dawn ( 2010), “Brand Attachment and Brand Attitude Strength: Conceptual and Empirical Differentiation of Two Critical Brand Equity Drivers,” Journal of Marketing , 74 6, 1– 17. Google Scholar CrossRef Search ADS   Park Ji Kyung, John Deborah Roedder ( 2010), “Got to Get You into My Life: Do Brand Personalities Rub Off on Consumers?,” Journal of Consumer Research , 37 4, 655– 69. Google Scholar CrossRef Search ADS   Pavot William, Diener Ed ( 1993), “Review of the Satisfaction with Life Scale,” Psychological Assessment , 5 2, 164. Google Scholar CrossRef Search ADS   Raghunathan Rajagopal, Corfman Kim ( 2006), “Is Happiness Shared Doubled and Sadness Shared Halved? Social Influence on Enjoyment of Hedonic Experiences,” Journal of Marketing Research , 43 3, 386– 94. Google Scholar CrossRef Search ADS   Rick Scott I., Small Deborah A., Finkel Eli J. ( 2011), “Fatal (Fiscal) Attraction: Spendthrifts and Tightwads in Marriage,”  Journal of Marketing Research,   48 2, 228– 37. Google Scholar CrossRef Search ADS   Robins Richard W., Caspi Avshalom, Moffitt Terrie E. ( 2000), “Two Personalities, One Relationship: Both Partners’ Personality Traits Shape the Quality of Their Relationship,”  Journal of Personality and Social Psychology,   79 2, 251. Google Scholar CrossRef Search ADS   Rucker Derek D., Dubois David, Galinsky Adam D. ( 2011), “Generous Paupers and Stingy Princes: Power Drives Consumer Spending on Self Versus Others,” Journal of Consumer Research , 37 6, 1015– 29. Google Scholar CrossRef Search ADS   Rucker Derek D., Galinsky Adam D. ( 2008), “Desire to Acquire: Powerlessness and Compensatory Consumption,” Journal of Consumer Research , 35 2, 257– 67. Google Scholar CrossRef Search ADS   Rucker Derek D., Galinsky Adam D. ( 2016), “Growing beyond Growth: Why Multiple Mindsets Matter for Consumer Behavior,” Journal of Consumer Psychology , 26 1, 161– 4. Google Scholar CrossRef Search ADS   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. Google Scholar CrossRef Search ADS   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. Google Scholar CrossRef Search ADS   Rusbult Caryl E., Kumashiro Madoka, Kubacka Kaska E., Finkel Eli J. ( 2009), “‘ The Part of Me That You Bring Out’: Ideal Similarity and the Michelangelo Phenomenon,” Journal of Personality and Social Psychology , 96 1, 61. Google Scholar CrossRef Search ADS PubMed  Schwarz Norbert, Bless Herbert, Strack Fritz, Klumpp Gisela, Rittenauer-Schatka Helga, Simons Annette ( 1991), “Ease of Retrieval as Information: Another Look at the Availability Heuristic,” Journal of Personality and Social Psychology , 61 2, 195. Google Scholar CrossRef Search ADS   Selig James P., Preacher Kevin J. ( 2008), “Monte Carlo Method for Assessing Mediation: An Interactive Tool for Creating Confidence Intervals for Indirect Effects [Computer Software].” Shiota Michelle N., Levenson Robert W. ( 2007), “Birds of a Feather Don’t Always Fly Farthest: Similarity in Big Five Personality Predicts More Negative Marital Satisfaction Trajectories in Long-Term Marriages,” Psychology and Aging , 22 4, 666. Google Scholar CrossRef Search ADS PubMed  Simpson Jeffry A., Farrell Allison K., Oriña M. Minda, Rothman Alexander J. ( 2014), “Power and Social Influence in Relationships,” in APA Handbook of Personality and Social Psychology, Volume 3: Interpersonal Relations , ed. Simpson Jeffry A., Dovidio John F., Washington, DC: American Psychological Association, 393– 420. Simpson Jeffry A., Griskevicius Vladas, Rothman Alexander J. ( 2012), “Consumer Decisions in Relationships,” Journal of Consumer Psychology , 22 3, 304– 14. Google Scholar CrossRef Search ADS   Smith Pamela K., Bargh John A. ( 2008) “Nonconscious Effects of Power on Basic Approach and Avoidance Tendencies,” Social Cognition , 26 1, 1. Google Scholar CrossRef Search ADS PubMed  Smith Pamela K., Jostmann Nils B., Galinsky Adam D., van Dijk Wilco W. ( 2008), “Lacking Power Impairs Executive Functions,” Psychological Science , 19 5, 441– 7. Google Scholar CrossRef Search ADS PubMed  Spiller Stephen A., Fitzsimons Gavan J., Lynch John G.Jr., McClelland Gary H. ( 2013), “Spotlights, Floodlights, and the Magic Number Zero: Simple Effects Tests in Moderated Regression,” Journal of Marketing Research , 50 2, 277– 88. Google Scholar CrossRef Search ADS   Steffel Mary, Le Boeuf Robyn A. ( 2014), “Overindividuation in Gift Giving: Shopping for Multiple Recipients Leads Givers to Choose Unique but Less Preferred Gifts,” Journal of Consumer Research , 40 6, 1167– 80. Google Scholar CrossRef Search ADS   Thibaut John W., Kelley Harold H. ( 1959), The Social Psychology of Groups , New York: Wiley. Thomson Matthew, MacInnis Deborah J., Park C. Whan ( 2005), “The Ties that Bind: Measuring the Strength of Consumers’ Emotional Attachments to Brands,” Journal of Consumer Psychology , 15 1, 77– 91. Google Scholar CrossRef Search ADS   Tidwell Natasha D., Eastwick Paul W., Finkel Eli J. ( 2013), “Perceived, Not Actual, Similarity Predicts Initial Attraction in a Live Romantic Context: Evidence from the Speed‐Dating Paradigm,” Personal Relationships , 20 2, 199– 215. Google Scholar CrossRef Search ADS   Waite Linda, Gallagher Maggie ( 2002), The Case for Marriage: Why Married People Are Happier, Healthier and Better Off Financially , New York: Broadway Books. Watson David, Klohnen Eva C., Casillas Alex, Simms Ericka Nus, Haig Jeffrey, Berry Diane S. ( 2004), “Match Makers and Deal Breakers: Analyses of Assortative Mating in Newlywed Couples,” Journal of Personality , 72 5, 1029– 68. Google Scholar CrossRef Search ADS PubMed  Weisfeld Glenn E., Russell R. J. H., Weisfeld C. C., Wells Pamela A. ( 1992), “Correlates of Satisfaction in British Marriages,” Ethology and Sociobiology , 13 2, 125– 45. Google Scholar CrossRef Search ADS   Zhao Xinshu, Lynch John G.Jr., Chen Qimei ( 2010), “Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis,” Journal of Consumer Research , 37 2, 197– 206. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Consumer Research Oxford University Press

Coke vs. Pepsi: Brand Compatibility, Relationship Power, and Life Satisfaction

Loading next page...
 
/lp/ou_press/coke-vs-pepsi-brand-compatibility-relationship-power-and-life-DcoKZ1CgnK
Publisher
University of Chicago Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0093-5301
eISSN
1537-5277
D.O.I.
10.1093/jcr/ucx079
Publisher site
See Article on Publisher Site

Abstract

Individuals often evaluate, purchase, and consume brands in the presence of others, including close others. Yet relatively little is known about the role brand preferences play in relationships. In the present research, the authors explore how the novel concept of brand compatibility, defined as the extent to which individuals have similar brand preferences (e.g., both partners prefer the same brand of soda), influences life satisfaction. The authors propose that when brand compatibility is high, life satisfaction will also be high. Conversely, because low brand compatibility may be a source of conflict for the relationship, the authors propose that it will be associated with reduced life satisfaction. Importantly, the authors predict that the effects of brand compatibility on conflict and life satisfaction will depend upon relationship power. Across multiple studies and methodologies, including experimental designs (studies 2, 3, 5) and dyadic data from real-life couples (studies 1, 4, 6), the authors test and find support for their hypotheses. By exploring how a potentially unique form of compatibility influences life satisfaction, including identifying a key moderator and an underlying mechanism, the current research contributes to the literatures on branding, close relationships, consumer well-being, and relationship power. brand compatibility, close relationships, relationship power, consumer well-being, multilevel modeling, actor-partner interdependence model Consider for a moment some of your favorite brands. For example, perhaps Coke, Starbucks, and Crest are your favorite brands of soda, coffee, and toothpaste, respectively. Now think about your partner’s favorite brands in the same product categories. Perhaps your partner likes the same brands, or, perhaps your partner prefers Pepsi, Dunkin’ Donuts, and Colgate. Does this compatibility, or lack of compatibility, matter? In the present research, we propose that brand preferences within close relationships do matter, and that they can have significant consequences, including influencing life satisfaction. To investigate this proposition, we introduce the term brand compatibility, which we define as the extent to which individuals in close relationships share similar brand preferences, and we examine whether brand compatibility is related to life satisfaction. Importantly, we hypothesize that the link between brand compatibility and life satisfaction will depend upon power in the relationship. Power is typically defined as the ability to control valued resources and the capacity to influence the behavior of others (Anderson and Galinsky 2006; Emerson 1962; French and Raven 1959; Keltner, Gruenfeld, and Anderson 2003). Imagine that you and your partner have different brand preferences. Now imagine that you perceive you have a lot of power in the relationship. Most likely, your partner’s preferences would not affect your day-to-day consumer decisions. You would probably still drink Coke and go to Starbucks. Further, research has shown that high-power individuals are less likely to notice the attitudes and opinions of their lower-power counterparts (Berdahl and Martorana 2006; Fiske 1993), so you might not even be aware that your partner prefers Pepsi and Dunkin’ Donuts. Thus, brand (in)compatibility most likely would not be an issue for you. On the other hand, imagine that you do not have a lot of power in the relationship. In this situation, you are more likely to be aware of your partner’s different brand preferences as you find yourself drinking more Pepsi and stopping at Dunkin’ Donuts more often than you would prefer. Given that these differences are more pronounced for low-power partners, you may view the relationship as having more disagreements or greater conflict, which in turn may lead to lower happiness in your day-to-day life. Thus, we suggest that the link between brand compatibility, conflict, and life satisfaction depends upon whether one is relatively high or low in power within one’s romantic relationship. When one is high in power, then one should be able to control relationship outcomes and get the brands that one prefers. Thus, we predict that for those who are high in relationship power, brand compatibility will not affect life satisfaction presumably because they are able to acquire their preferred brands regardless of their partner’s preferences. Conversely, when one is low in relationship power, one is less likely to get the brands that one prefers. As a result, lower-power partners may be more likely to perceive greater conflict in the relationship. Therefore, we predict that for lower-power individuals, brand compatibility will be related to life satisfaction. Specifically, when brand compatibility is high, life satisfaction will also be high because there is no conflict—lower-power individuals are able to enjoy their preferred brands (which are the same as their partner’s preferred brands). But for low-power individuals, when brand compatibility is low, perceived conflict should be increased and life satisfaction will be reduced. We test these predictions across six studies, using a mix of dyadic studies of real relationship partners and experimental studies that allow us to manipulate perceptions of brand compatibility, and thus provide causal evidence for its importance above and beyond other types of compatibility. In addition, we provide evidence that perception of conflict is the mechanism through which brand compatibility and power interact to influence life satisfaction. The present research makes several contributions. First, we explore a new construct, brand compatibility, and demonstrate that it can have important downstream consequences for life satisfaction. By connecting this consumer construct to a well-established marker of psychological well-being, our findings support the importance of consumer behavioral constructs to the health and well-being literature (Diener and Biswas-Diener 2002; Fredrickson and Joiner 2002). Second, we explore how brand preferences intersect with relationship contexts, contributing to burgeoning efforts to understand how consumer behavior both shapes and is shaped by social relationships (Corfman and Lehmann 1987; Dzhogleva and Lamberton 2014; Lowe and Haws 2014; Luo 2005; Park 1982; Raghunathan and Corfman 2006; Steffel and Le Boeuf 2014). Third, our findings illustrate the importance of brand preferences in relationships. Relationship scholars have long been interested in compatibility (Blossfeld 2009; Bramlett and Mosher 2002; Byrne 1971; Gonzaga, Campos, and Bradbury 2007; Little and Perrett 2002; Mare 1991; Rusbult et al. 2009), but most theorists have argued that similarity of important values such as religion and ethical beliefs is what matters. In the current studies, we show that even mundane, everyday forms of compatibility, such as brand compatibility, can have significant consequences both for the relationship, in terms of perceived conflict, and for the individual, in terms of life satisfaction. Our findings support classic theoretical arguments that social similarities (e.g., race, religion) are likely to be only modestly related to how well couples get along, whereas areas of similarity that affect the day-to-day decisions partners make are likely to be more important (Levinger and Breedlove 1966; Levinger and Rands 1985). Fourth, our findings point to the importance of studying power in close relationships. While most research on power explores organizational and broader social contexts, recently theorists have highlighted the need to reintroduce the power concept to the study of close relationships (Simpson et al. 2014; Simpson, Griskevicius, and Rothman 2012). In particular, the current research challenges the long-held assumption in the relationship literature that compatibility is inherently good (Acitelli, Douvan, and Veroff 1993; Byrne 1971; Montoya, Horton, and Kirchner 2008; Murray et al. 2002); instead, our findings suggest that compatibility may matter primarily for those who perceive having low power in the relationship. THEORETICAL BACKGROUND Brand Compatibility Brands play many roles in consumers’ day-to-day lives. As brands can represent one’s identity, personality, beliefs, social connections, culture, and heritage (Escalas and Bettman 2003; Holt 2002; Muniz and O’Guinn 2001; Ng and Houston 2006; Park and John 2010), individuals may use brands to communicate to others who they are or to represent their self (Belk 1988; Berger and Heath 2007; Escalas and Bettman 2005; Malhotra 1988). Research has shown that individuals may use brands as relationship partners (Aaker, Fournier, and Brasel 2004; Aggarwal 2004; Carroll and Ahuvia 2006; Fournier 1998; Park et al. 2010; Thomson, MacInnis, and Park 2005), or individuals may merely use brands as a heuristic to help guide choice in a world increasingly overrun with options (Aaker and Keller 1990; Hoeffler and Keller 2003; Keller 1993). Regardless of the reason, individuals commonly evaluate, purchase, and consume their preferred brands. Although individuals typically consume the brands that they prefer, there are situations in which they may not be able to do so. For example, workplace lunches provide certain brands of soda; certain shopping centers have only one coffee shop. When brand preferences are constrained by situational factors, consumers tend to feel less satisfied with the experience (Clee and Wicklund 1980; Fitzsimons 2000). We suggest that close relationships, in addition to stockouts and other consumer variables (e.g., time, money), can also constrain the expression of brand preferences. Indeed, close relationships are one of the strongest, most controlling, and long-term types of social situations, so their opportunity for influence is extremely high (Thibaut and Kelley 1959). As close relationships involve repeated, frequent interactions in diverse settings, and strong mutual interdependence whereby the outcomes of one’s decisions affect both the other person and the relationship itself (Berscheid, Snyder, and Omoto 1989; Kelley 2013; Thibaut and Kelley 1959), individuals are often constrained by their partner’s preferences and the demands of the relationship. This constraint should be especially evident and influential in everyday life when the partners have different consumer preferences, including brand preferences, and be most likely to influence life satisfaction. To illustrate why brand preferences may matter, consider the following two scenarios. Imagine you and your partner are looking for a new car. You have always wanted an Audi. In the first scenario, imagine your partner also loves Audi. You buy the Audi and are both happy. In this first scenario, you and your partner have high brand compatibility, defined as the degree to which individuals have similar brand preferences. However, imagine a second scenario in which your partner does not like Audi. Instead, your partner prefers Lexus cars, and in this scenario, you get the Lexus. This is obviously not a terrible situation (it is still a luxury car brand), but it is not the car brand you had always wanted. Now imagine you are driving with your partner to get coffee. First, imagine you and your partner both like Starbucks. There is no conflict. You stop at Starbucks, you both get to enjoy the coffee you prefer, and you are on your way. However, in the second scenario, you like Starbucks, but your partner prefers Dunkin’ Donuts. Where do you stop? Ideally, you stop at both places, but perhaps due to time constraints you can stop at only one. Further, imagine that instead of just one time, this is your daily commute to work. This simple brand decision, which brand of coffee to buy, is part of your everyday life as a couple. Daily, you find yourself driving in the Lexus (the car you did not want), drinking Dunkin’ Donuts (coffee that is not your favorite). Extending beyond the car and coffee example, you, as a couple, may also have to choose between two different brands of soda, beer, or even toilet paper to purchase and consume at home. In the moment, and over time, these small, seemingly trivial brand choices may have consequences. High brand compatibility means that both partners’ desire to satisfy their brand preferences can be easily accomplished, in the moment and on a regular, daily basis, and we conjecture that it may be influential in the couple’s overall well-being, as is fulfillment of other goals and needs in everyday life (Emmons 1986; Emmons and King 1988). On the other hand, over the lifetime of a relationship, low brand compatibility means that a couple is likely to make hundreds or thousands of small decisions in which the two partners cannot both satisfy their preferred outcome. Having dissimilar brand preferences, or low brand compatibility, thus creates multiple opportunities for conflict and presents a challenge to the couple. If you really like Diet Coke, Starbucks, and Audis but find yourself drinking your partner’s favorites, Diet Pepsi and Dunkin’ Donuts, and driving a Lexus, you might perceive greater conflict in the relationship and be less happy. This is what we propose in the current research: that low brand compatibility will be associated with greater perceived conflict and reduced life satisfaction for those partners who fall on the “losing” side of incompatible preferences (i.e., those low in power; see the following section). We propose this will happen both in the moment—that people will temporarily feel less satisfied—and over time, as these experiences accumulate in memory. The potential costs and benefits of similarity in relationships has been a topic of immense interest in the relationships literature for decades (Acitelli et al. 1993; Bramlett and Mosher 2002; Byrne 1971; Dyrenforth et al. 2010; Houts, Robins, and Huston 1996; Montoya et al. 2008; Watson et al. 2004), but compatibility researchers have more typically focused on deeply held values like religious beliefs and political ideologies (Heaton and Pratt 1990) and on personality traits (Blum and Mehrabian 1999; Glicksohn and Golan 2001). Here, we extend the logic of compatibility to study a far more mundane and everyday type of similarity—namely, brand preferences. Importantly (and this is where our research turns away from the standard assumption that compatibility is straightforwardly beneficial), we suggest that these two situations—of high and low compatibility—are of differential importance to the members of the couple, depending on how much power each partner holds in the relationship. Power in Relationships Power is a common and pervasive component of social interactions and relationships (Anderson, John, and Keltner 2012; Galinsky, Gruenfeld, and Magee 2003). Power is commonly defined as having relative control over valued resources and capacity to influence the behavior of others (Anderson and Galinsky 2006; Emerson 1962; French and Raven 1959; Keltner et al. 2003; Magee and Galinsky 2008), while resisting the influence of others over oneself (Cromwell and Olson 1975). Power in close or romantic relationships can be thought of similarly—as the capacity to influence outcomes in the relationship. In the current research, we focus directly on perceived relationship power. We assume that perceived relationship power is related to decisions and experiences within the relationship, including the outcomes of brand decisions. That is, we assume that partners who perceive themselves as high power more reliably obtain and consume their desired brands than do partners who perceive themselves as low power. Although this has not been experimentally demonstrated, it appears to be a safe assumption: by definition, partners who are relatively higher in power are more likely to report obtaining their desired outcomes. Indeed, providing some support for the notion, historical research on gender and family decision making suggested that men reported greater power and reported controlling more of the important purchasing decisions in families (Filiatraut and Ritchie 1980; Kirchler 1993). If this logic extends to minor brand decisions in everyday life, partners who perceive higher power should more frequently consume the brands they want to consume. This tendency for high-power partners to “get what they want” may occur via several interrelated mechanisms. First, and most simply, they may just control the outcomes directly, by demands or pressure on their partners (Simpson et al. 2014). The Starbucks-loving high-power partner may simply scoff at the suggestion of stopping at Dunkin’ Donuts. However, the process may also emerge from subtler and more indirect psychological dynamics in the couple. High-power individuals are more likely to express their opinions to others (Berdahl and Martorana 2006); thus, perhaps high-power partners will simply make their brand preferences known to their partners more often or more clearly. High-power individuals have also been shown to be more impervious to others’ attitudes. For example, they have a reduced tendency for perspective taking and comprehending how others think and feel (Galinsky et al. 2006) and are less motivated to attend to others’ thoughts and behaviors (Fiske 1993; Keltner et al. 2003). Thus, high-power partners are less likely to notice the discrepant preferences of their relationship partners, which likely reduces the chance they would consider those preferences in their brand choices. In the realm of decision making, consumers high in power tended to discount others’ opinions (Mourali and Yang 2013). Third, high-power individuals may be less likely to value their partners’ preferences. For example, consumers led to feel high in power spent more on purchases for themselves, and less on purchases for others, while consumers led to feel low in power showed the reverse pattern, spending more on purchases for others (Rucker, Dubois, and Galinsky 2011). As a result of these dynamics, in everyday life, individuals who perceive greater power in the relationship are more likely to obtain their desired brands regardless of their partner’s preferences. Therefore, we predict that for high-power individuals, brand compatibility is not a particularly important construct in determining their quality of life or sense of well-being. If they have high brand compatibility with their romantic partner, that is great, but if not, that is still good—their own outcomes are relatively unaffected by their partner’s preferences. In other words, differences in brand preferences should not be perceived as a source of conflict for high-power partners. Thus, we predict that for individuals who are high in power, brand compatibility will not affect life satisfaction. Quite in contrast, low-power partners, by definition, are less likely to have control over outcomes within the relationship, including brand choices. Low-power partners are much less likely to attempt to control the outcome directly through pressure or demands (Simpson et al. 2014). In addition, they have been found to express their opinions to others less (Berdahl and Martorana 2006), and they are more likely to pay attention to the preferences, attitudes, and feelings of their partners (Fiske 1993; Keltner et al. 2003). Furthermore, research has shown that they conform to other people’s opinions (Mourali and Yang 2013) and value their outcomes (Rucker et al. 2011). Consequently, they may be more likely simply to know their partner’s preferred brands and to conform to them when making brand choices, even if those preferences do not reflect their own. For these reasons, low-power individuals are relatively less likely to “win” in conflicts over brand preferences in close relationships. As a result, brand compatibility may play a much larger role for low-power partners. Given this reasoning, we hypothesize that the construct of brand compatibility should be of particular importance in predicting the well-being, in terms of relationship conflict and life satisfaction, of low-power individuals. CURRENT RESEARCH In the current research, we investigate how power dynamics and brand preferences in close romantic relationships affect life satisfaction. As high-power partners are less likely to be aware of their partners’ preferences (Berdahl and Martorana 2006; Fiske 1993) and more likely to perceive control over outcomes within the relationship (Galinsky et al. 2003; Simpson et al. 2012), including brand outcomes, we predict that brand compatibility will not affect life satisfaction for high-power individuals. On the other hand, we predict that for low-power individuals, as brand compatibility decreases, so too will life satisfaction. As low-power individuals are more likely to be aware of differences in brand preferences and less likely to perceive control over outcomes, they will have less control over consumption decisions relative to their partner, making the similarity of the two partners’ preferences very important to their outcomes. We further predict that low-power partners will perceive greater conflict in the relationship when brand compatibility is low. This greater conflict will in turn contribute to their reduced life satisfaction. We tested these hypotheses across six studies (and one direct replication). As this is a new construct, in our first study we examined whether brand compatibility between romantic partners was related to life satisfaction and other forms of compatibility. In our second study, we experimentally manipulated perceived brand compatibility and measured relationship power and life satisfaction. In our third study, we experimentally manipulated perceived power in the relationship and measured brand compatibility and life satisfaction. In our fourth study, we simultaneously measured brand compatibility, power, and life satisfaction in a laboratory setting with both members of romantic couples. In dyadic analyses, we used the Actor-Partner Interdependence Model (Kenny, Kashy, and Cook 2006) to further investigate how power influences the effect of brand compatibility on life satisfaction. We explored the possibility of an actor-by-partner interaction, examining whether the strongest effect of brand compatibility on life satisfaction would emerge for low-power individuals with high-power partners. In our fifth study, we examined potential mechanisms of this effect. Using a scenario study in which we were able to manipulate both brand compatibility and relationship power, we explored the role of perceived conflict. Finally, in our sixth study, we again examined both members of romantic couples to examine the dyadic effects, and test the notion that brand compatibility and power’s interactive effect on life satisfaction is mediated by perceptions of conflict within the relationship. If high-power individuals are not aware of or affected by incompatibility, they should be less likely to perceive conflict. If low-power individuals are more aware of their partner’s preferences and more affected by incompatibility, they should perceive greater relationship conflict, and thus, feel less satisfied with their lives. Across all of the studies, we report all data exclusions (if any), all manipulations, and all measures in the study (with the exceptions noted below). First, for each study involving both members of the couple, sample sizes were based on subject availability as well as the needs of unrelated research projects that were run in conjunction with data collection. For studies using online participants, sample sizes were determined a priori. No additional data were collected after data analyses began (with the exception of study 2, in which additional conditions were added in response to reviewer request). In some experiments, as indicated in the text, we included additional irrelevant measures, such as filler items or measures for other research projects. As discussed in the methods section of study 4, data from this study were collected in two waves during multistudy events with other researchers in the lab; we do not have access to all those measures or data. Finally, in study 6, a multiwave, longitudinal dyadic study, not all measures in the study are reported due to the sheer number of variables and measures. All measures related to the hypotheses are described, and only the described measures were analyzed for the current hypothesis. STUDY 1 As this is the first research to examine the role of brand compatibility on life satisfaction, we first sought to explore whether brand compatibility is associated with life satisfaction and other forms of compatibility. In study 1, we survey both partners within a couple. We ask them about various measures that are commonly examined in similarity and compatibility research (race, religion, personality, values, etc.). In addition, we ask each partner to report his/her own brand preferences and measure life satisfaction, or general happiness with how one’s life is going, as the dependent variable. By comparing the participant and the partner on each domain, we are thus able to construct objective measures of these various forms of compatibility using outside raters. Method Participants We recruited participants from the local farmer’s market in a southeastern city. In order to participate, participants had to be in a relationship with their partner for at least six months and living with their partner for at least three months. Both partners had to be present and willing to complete the survey in order to participate. Both partners of 63 romantic relationships (52% female, 1% trans, 1% both, 1% neither) completed the survey. Due to comprehension/incomplete data issues, two participants and their partners were excluded from all analyses. Partners ranged in age from 25 to 70 years with an average age of 40.67 years (SD = 12.94) and had been in a relationship for 13.61 years on average (SD = 12.96). Couples received financial compensation in exchange for their participation. Measures and Procedure After indicating consent, each partner was given a clipboard with a survey and told that there were several parts to the survey. Each partner was instructed to complete the survey separately and to not discuss their responses with their partner. For each of the compatibility factors we created a compatibility score for each couple. (Note: We repeated the analyses using various coding schemes, to ensure that the results are not driven by the use of any specific type of coding scheme.) Brand Compatibility Each partner was asked to list his/her favorite brand in each of the following five categories: coffee, chocolate, car, beer, and soda. We selected these categories because they are common categories wherein most individuals in our sample would have experience with different brands, and in which individuals generally have specific brand preferences. In addition, they are categories in which individuals regularly make choices and wherein a partner is repeatedly exposed to these choices (e.g., every time they drive the household car they are reminded of their partner’s preference, or every time they open the fridge they may see their partner’s brand of soda, or every time they pass a Starbucks they are reminded of their partner’s favorite brand of coffee). Furthermore, these categories vary in terms of price and type (e.g., durable vs. nondurable). Essentially, we simply sought to sample a wide variety of product types. Four undergraduate research assistants blind to the hypothesis of the study rated each of the brand pairs (i.e., how the brand that partner 1 within couple 1 listed as his/her favorite compared with the brand that partner 2 within couple 1 listed as his/her favorite) within a product category on a 1 (completely incompatible) to 5 (completely compatible) scale. Coders were instructed to evaluate the brands on how compatible, or similar, one partner’s brand was with the other partner’s brand. Each coder was given some examples of brand compatibility coding using the category of soda. They were told that the exact same brand response within a couple for a category, such as partner 1 responding with Diet Coke and partner 2 also responding with Diet Coke, would be a 5. Brands that are close but not exact, such as partner 1 responding Coke and partner 2 responding Diet Coke, would be a 4. A 2 or 3 would be given for brand pairs that are not identical, but not competitors. For example, if partner 1 said Coke and Partner 2 said Sprite, they would be given a 3. These brands are not completely compatible, but have an overarching relationship—in this case, that they are owned by the same parent company. Competitor or opposite brands would be coded as a 1. For example, if partner 1 said Coke and partner 2 said Diet Pepsi, they would be given a 1. This coding scheme has the advantage of allowing us to differentiate between responses such as “No preference” from “I don’t like soda,” which would have different implications for the couple. For example, if a participant said he liked Coke but his partner said she had no preference, this would be coded as more compatible than if the participant said he liked Coke, but his partner said she does not like soda. In the former case, purchasing a 12-pack of Coke could satisfy both partners, whereas in the latter case no brand of soda will satisfy both partners.1 In addition, coders were told that this is a subjective rating—that there are no right or wrong answers, but that they should be consistent in their ratings. Coders were told that if they had any questions about the brands, they could look them up online. The four coders’ ratings were averaged to create a brand compatibility score for each brand category (Mbeer = 2.41, SD = 1.22; Mcar = 2.69, SD = 1.20; Mchoc = 2.81, SD = 1.38; Mcoffee = 3.22, SD = 1.34; Msoda = 3.00, SD = 1.47). Coders were highly reliable within each brand category (interrater reliability: beer α = .89; car α = .91; chocolate α = .93; coffee α = .94; soda α = .94). Brand categories were then averaged to create one mean brand compatibility score for each couple, which served as the independent variable. Other Measures of Compatibility The following were included as additional measures of compatibility: age, education, race, religiousness, religion, political orientation, personality, and values. We selected these items and calculated similarity within couples based on previous research in similarity and compatibility (Houts et al. 1996; Watson et al. 2004; see Finkel et al. 2012 for a review). (Please see the web appendix for detailed descriptions regarding the coding of these items.)2 Each partner also completed the Satisfaction with Life Scale (Diener et al. 1985; α = .85). (Note: This measure was completed third, after participants indicated their favorite brands and completed the values measure.) This is a well-established measure of life satisfaction, commonly used in research on well-being (Aknin et al. 2013; Burroughs and Rindfleisch 2002; Cohn et al. 2009; Diener and Biswas-Diener 2002; Diener et al. 2010; Luhmann et al. 2012; Martin and Hill 2012) and has been shown to correlate highly with other important measures of well-being and quality of life, such as health, happiness, and self-determination (Arrindell, Meeuwesen, and Huyse 1991; Diener 2012; Pavot and Diener 1993). The measure asks participants to evaluate their life overall, and provide a general sense of how well things are going. For example, items include “In most ways my life is close to my ideal” and “I am satisfied with my life.” Participants completed additional items, including filler items and demographic items. Results We investigated whether brand compatibility predicted life satisfaction while controlling for other variables. We used a multilevel modeling approach (Kenny et al. 2006), with individuals nested within couples to account for violations of statistical independence. Life satisfaction served as our outcome variable, and brand compatibility, race compatibility, political orientation compatibility, religiousness compatibility, education compatibility, personality compatibility, and value compatibility served as our predictor variables. All of the predictor variables were grand mean-centered (Aiken and West 1991; Kenny et al. 2006). When we controlled for various other forms of compatibility, brand compatibility was significantly and positively associated with life satisfaction (B = .40, t (51.74) = 2.66, p = .010). See table 1 for details about all of the predictor variables, and the web appendix for information and a table regarding correlations of all variables. We note that these results were robust across a wide array of models including various coding schemes of compatibility, such as absolute difference versus difference score approaches. We report the model that offers the best fit (e.g., lowest AIC of all the models, most parsimonious). (For information regarding the other models and correlations, please see the web appendix.) Table 1 Regression Coefficients Predicting Life Satisfaction in Study 1 Variable  Estimate  Standard Error  t-test  p value  Intercept  5.46  0.09  59.28  .000  Brand compatibility  0.40  0.15  2.66  .01  Age  0.02  0.09  0.18  .86  Race  –0.23  0.15  –1.56  .12  Political orientation  0.09  0.09  0.98  .33  Religiousness  0.09  0.08  1.22  .23  Education  –0.06  0.09  –0.66  .52  Personality (Big 5)  –0.01  0.18  –0.07  .95  Values (LOV)  0.15  0.30  0.51  .61  Variable  Estimate  Standard Error  t-test  p value  Intercept  5.46  0.09  59.28  .000  Brand compatibility  0.40  0.15  2.66  .01  Age  0.02  0.09  0.18  .86  Race  –0.23  0.15  –1.56  .12  Political orientation  0.09  0.09  0.98  .33  Religiousness  0.09  0.08  1.22  .23  Education  –0.06  0.09  –0.66  .52  Personality (Big 5)  –0.01  0.18  –0.07  .95  Values (LOV)  0.15  0.30  0.51  .61  Dependent variable: life satisfaction; AIC 312.323. Discussion In the present research, we find that brand compatibility within couples is related to other forms of compatibility, namely education, race, and values. We also find that, when all forms of compatibility are simultaneously entered as predictors, brand compatibility remains an important, and significant, predictor of life satisfaction. One might wonder how similarity in soda (i.e., brand) preferences could be more influential than similarity in age or religious preferences. We have several thoughts as to why this pattern of associations emerged. First, it is important to acknowledge that mixed results on the big compatibility dimensions (e.g., age, race, religion, personality) are the norm in this field (Tidwell, Eastwick, and Finkel 2013). For example, in the study of sociodemographic similarity, although some research has shown that marriages between individuals of the same race, religious denomination, parental wealth, and earned income are longer-lasting and more satisfying (Bramlett and Mosher 2002; Heaton and Pratt 1990; Weisfeld et al. 1992), other research has failed to find similar benefits (Houts et al. 1996; Watson et al. 2004), while still other research has found positive associations for some variables, but weak or inconsistent associations for others (Gaunt 2006). In the study of personality, some researchers have found positive effects for personality similarity (Luo and Klohnen 2005; Robins, Caspi, and Moffitt 2000), while others have found that after they controlled for the main effects of each partner’s personality, similarity between personalities had a weak relation to romantic outcomes (Blum and Mehrabian 2001; Dyrenforth et al. 2010; Glicksohn and Golan 2001; Tidwell et al. 2013). Indeed, some research has even found that personality similarity has a negative predictive effect on marital satisfaction (Shiota and Levenson 2007). Second, we suggest that brand compatibility produced a stronger correlation precisely because it is a relatively mundane, everyday type of compatibility. As early as 1966, Levinger and Breedlove “emphasized the importance of identifying areas of similarity that bear upon the day-to-day decisions partners make” (Houts et al. 1996). It may be the case that small, everyday decisions are the ones that more directly affect couples’ sense of well-being, as compared with broader and overarching variables like personality and religion. Because individuals commonly evaluate, purchase, and consume brands in the presence of close others, over time the significance of brand choices within a relationship accumulates. Due to their symbolic nature plus the consumer tendency to repeatedly purchase brands, brands represent a salient source of similarity (or potential dissimilarity) in close relationships. Thus, it is possible that brand compatibility is a more specific or idiosyncratic measure of compatibility, and therefore, likelier to be more predictive than general measures. The main aim of the study was to investigate the extent of the overlap among brand compatibility and previously studied forms of compatibility. In line with our theorizing, there is evidence to suggest that the various forms of compatibility are relatively distinct, and that brand compatibility does seem to have some unique predictive power. On the question of whether brand compatibility is a better or weaker predictor than other forms of compatibility, we are certainly not suggesting brands are the only important form of compatibility, but merely one such form, due to their ubiquity in everyday life. Importantly, we predict that the effect of brand compatibility on different outcomes will be related to relationship power. We test this prediction in the rest of the studies. STUDY 2 Study 2 tests our prediction that the effects of brand compatibility on life satisfaction will depend upon power in the relationship. Because high-power individuals have been shown to project their attitudes and feelings on to others (Keltner et al. 2003; Overbeck and Droutman 2013), measuring both brand compatibility and power from one partner could result in biased responses. Therefore, in our second study, we manipulate perceptions of brand compatibility. We hypothesize that for high-power partners, there will be no effect of brand compatibility condition on life satisfaction; however, for low-power partners, we predict that low brand compatibility will be associated with decreased life satisfaction, compared with high brand compatibility. Method Participants Three hundred twenty-five participants (46% men) from Amazon’s Mechanical Turk completed the study in exchange for financial compensation. In order to participate, individuals had to be in a romantic relationship for at least six months and in the United States. Participants ranged in age from 19 to 73 years with an average age of 35.41 years (SD = 11.62) and had been in a relationship for 6.76 years on average (SD = 7.99). Measures and Procedure After individuals indicated consent, they were told that there were several parts to the study and that the researchers were interested in different topics. Participants first completed the Relationship Power Measure (α = .91), which we created by adapting items from the Personal Sense of Power scale (Anderson et al. 2012). The scale ranged from 1 (strongly disagree) to 7 (strongly agree). Instructions indicated that participants should consider their current romantic relationship when answering items. Example items include, “I can get my partner to listen to what I say,” “My wishes do not carry much weight,” and “I think I have a great deal of power.” (See the appendix for items.) We manipulated perceived brand compatibility using an ease-of-retrieval manipulation (Schwarz et al. 1991). Specifically, participants in the high-brand-compatibility condition saw two lines in the survey and were asked to list two favorite brands that they had in common with their partner. Participants in the low-brand-compatibility condition saw eight lines in the survey and were asked to list up to eight brands. Participants had to list a response in the space provided, either a brand name or “NA,” in order to move on to the next question in the survey, to maximize the experience of ease versus difficulty in the two conditions. In order to be consistent with wording across the two conditions all participants were told the following: Some people in relationships are very compatible in their brand preferences. In other words, their favorite brand of soda is their partner’s favorite brand of soda. Other couples are less compatible in their brand preferences and their favorite brand in a product category is different from their partner’s. In the spaces below, please list brands that both you and your partner consider to be your favorite in that product category. For example, if you and your partner have the same favorite brand of soda, chocolate, coffee, car, beer, etc., you would enter the brand name in the space below. Please enter as many shared favorite brands that you and your partner have as you can. If you run out of shared favorite brands, then please enter “NA.” Next, participants completed the same Satisfaction with Life Scale (Diener et al. 1985; α = .80) used in the previous study. Finally, participants completed demographic items—age, gender, relationship type, and relationship length. Results Post-Test Manipulation Check We failed to include a manipulation check in the original study, and thus conducted a post-hoc manipulation check. Four hundred fifteen participants (49% men) from Amazon’s Mechanical Turk were recruited. Individuals had to be in a relationship for at least six months and in the United States in order to participate. Participants were randomized to either the same high- or low-brand-compatibility condition as used in the main study. As our measure of compatibility, participants were asked to indicate the extent to which they agreed or disagreed with the following statements on a seven-point scale: “My partner and I like the same brands,” “My partner and I are very compatible in our brand preferences,” and “My partner and I have similar brand preferences.” We combined these items to form one measure of perceived compatibility (α = .95). Two participants indicated generic categories (e.g., car, restaurants) and were excluded from the analyses. Because the majority of participants viewed themselves as compatible or highly compatible with their partner (85% of the average responses were in the upper half of the scale), we log-transformed the data (Bagchi and Cheema 2013; Dzhogleva and Lamberton 2014). As predicted, individuals in the high-brand-compatibility condition reported significantly greater compatibility (M = 5.23, SD = 1.16) than did individuals in the low-brand-compatibility condition (M = 5.01, SD = 1.36; t(411) = 2.08, p < .04). Main Analyses Two participants indicated that they were single, and were therefore excluded from the analyses. In addition, six participants’ mean power scores were more than three SDs below the mean and were excluded (Smith et al. 2008), leaving 317 participants. To investigate our hypothesis that low brand compatibility would be associated with decreased life satisfaction for low (but not high) power partners, we conducted a linear regression, with life satisfaction as the outcome variable, and brand compatibility and mean-centered power as the predictor variables. Results revealed a main effect for power (β = .37, t(313) = 7.09, p < .0001), such that greater power was associated with greater life satisfaction. There was no main effect of brand compatibility condition (β = .049, t(313) = 0.94, p = .35) on life satisfaction. Importantly, in line with our predictions, and as illustrated in figure 1, results revealed a significant interaction (β = .11, t(313) = 2.08 p < .04). Because we were interested in whether the effects of brand compatibility on life satisfaction differ depending upon power in the relationship, we first examined whether life satisfaction differed across the brand compatibility conditions for high-power individuals. In line with our predictions, we found that for high-power participants (+1 SD above the mean), life satisfaction did not differ across the two brand compatibility conditions (t(313) = .81; p = .42; 95% CI: –.19 to .47). However, for low-power participants (–1 SD), the effect of brand compatibility condition on life satisfaction was significant (t(313) = 2.14; p = .03; 95% CI: –.69 to –.03). Specifically, lower-power participants in the low-brand-compatibility condition reported significantly lower life satisfaction than did individuals in the high-brand-compatibility condition (figure 1). In addition, we conducted a floodlight analysis (Spiller et al. 2013). The results revealed that the difference between the brand compatibility conditions on life satisfaction occurred for participants whose relationship power levels were at or below power levels of 4.82 (approximately half a SD below the mean). FIGURE 1 View largeDownload slide RESULTS FROM STUDY 2: MANIPULATED BRAND COMPATIBILITY AND RELATIONSHIP POWER PREDICTING LIFE SATISFACTION NOTE.—Brand compatibility condition did not affect life satisfaction for high-power individuals. For low-power individuals, the low-brand-compatibility condition was associated with significantly decreased life satisfaction. FIGURE 1 View largeDownload slide RESULTS FROM STUDY 2: MANIPULATED BRAND COMPATIBILITY AND RELATIONSHIP POWER PREDICTING LIFE SATISFACTION NOTE.—Brand compatibility condition did not affect life satisfaction for high-power individuals. For low-power individuals, the low-brand-compatibility condition was associated with significantly decreased life satisfaction. Discussion3 In this study, we measured participants’ chronic sense of power in their relationship using a standard measure, and experimentally manipulated participants’ perception of brand compatibility. In line with our predictions, we found no effect of brand compatibility condition on life satisfaction for high-power individuals. In contrast, as predicted, we found that for low-power individuals, low brand compatibility led to reduced life satisfaction. Because we manipulated brand compatibility, these results highlight that brand compatibility is not just a marker of existing conflict in the relationship, but that brand compatibility, as a unique construct, interacts with power to predict feelings of life satisfaction. STUDY 3 In study 2, we measured power and manipulated brand compatibility. In study 3, we aimed to conduct a complementary study, increasing the generalizability of our findings. First, we experimentally manipulated individuals’ sense of power in the relationship in order to increase our ability to draw causal conclusions about the role of power in these effects. In line with the literature on social power, we theorize that power is both a chronic, long-term variable reflecting structural realities, and a situational variable, in flux in many relationships and social settings (Galinsky et al. 2003). As described in Jiang, Zhan, and Rucker (2014), because of its prevalence in everyday life, power is not only a long-term structural construct that can be measured, but also a more variable mindset that can be triggered or activated through common experimental techniques. Research has shown that power can be evoked through episodic recall, role playing, and even semantic priming (Galinsky et al. 2003; Magee, Galinsky, and Gruenfeld 2007). In other words, because it is a mindset, all people can draw on experiences of feeling both high and low power—even within the same relationship. One can think of parent-child or professor-student relationships as an example. Although the parent and professor chronically hold more power, one can probably recall a few times when it felt as though the child and/or the student had more power, which in turn can elicit responses associated with feeling low in relationship power. Research has also shown that chronic and manipulated power affect consumer outcomes, including thought and behavior, in similar ways (Rucker and Galinsky 2008, 2016; Rucker, Galinsky, and Dubois 2012). These previous results suggest that manipulating perceptions of relationship power by asking participants to recall an instance in their current relationship when they had low power should produce similar effects as chronic perceptions of power in the relationship. Therefore, as in study 2, we predict that for individuals in the high-power condition, there will be no effect of brand compatibility on life satisfaction. On the other hand, we predict that individuals in the low-power condition will report lower life satisfaction as brand compatibility decreases. Method Participants One hundred eighty-one individuals (51% men) from Amazon’s Mechanical Turk completed the study. As in the previous study, individuals had to be in a relationship for at least six months and in the United States in order to participate. The participants ranged in age from 18 to 68 years with an average age of 33.93 years (SD = 11.76) and had been in their relationship for an average of 6.30 years (SD = 7.78). Participants were paid in exchange for participation in the study. Measures and Procedure After providing consent, participants were asked to list their partner’s favorite brand in the same five categories as used in prior studies: coffee, chocolate, car, beer, and soda. Participants were then asked to list their own favorite brands in each of the same brand categories. Again, we selected these categories because we wanted a range of categories in terms of price and type, and these are brand categories in which individuals often have a preference. Next, participants were randomized to either a high- or low-power condition using a standard power manipulation, which we modified slightly to reflect romantic relationships (Galinsky et al. 2003). Specifically, participants in the high- (low-) power condition were told the following: Please recall a particular incident in which you had power over your partner (your partner had power over you). By power, we mean a situation in which you (your partner) controlled the ability of your partner (you) to get something they (you) wanted, or were in a position to evaluate your partner (you). Please describe this situation in which you had power—what happened, how you felt, etc. Participants then completed the Satisfaction with Life scale (Diener et al. 1985; α = .81) as used in study 2. They also completed a series of demographic items—gender, age, relationship type, and relationship length. Similar to study 1, we created a measure of perceived brand compatibility from participants’ brand responses. Two undergraduate research assistants blind to the hypothesis of the study rated each of the brand pairs (i.e., how compatible the brand that the participant listed as his/her favorite compared with the brand that the participant listed as his/her partner’s favorite) within a product category on the same 1 (completely incompatible) to 5 (completely compatible) scale used in study 1. Raters were highly reliable within each brand category (interrater reliability: beer α = .88; car α = .87; chocolate α = .90; coffee α=.91; soda α = .91). Brand categories were then averaged to create one mean brand compatibility score (Mbeer = 2.62, SD = 1.59; Mcar = 2.88, SD = 1.45; Mchoc = 2.85, SD = 1.58; Mcoffee = 3.10, SD = 1.68; Msoda = 2.88, SD = 1.67). Results One participant indicated that he was single and was therefore excluded from the subsequent analyses, leaving 180 participants. Manipulation Check Following prior research (Galinsky et al. 2003; Mourali and Yang 2013; Smith and Bargh 2008), an undergraduate research assistant blind to both condition and hypotheses rated participants’ responses for how much power the participant reported having over his/her partner using a seven-point Likert scale. Participants in the high-power condition described themselves having significantly more power over their partner (M = 5.27, SD = 1.00) than did individuals in the low-power condition (M = 2.48, SD = 1.24; t(178) = 16.54, p < .0001). Main Analyses A linear regression, with life satisfaction as the outcome variable and power condition and mean-centered brand compatibility as the predictor variables, revealed a marginally significant main effect for brand compatibility (β = .20, t(176) = 1.71; p < .09), such that greater brand compatibility was associated with greater life satisfaction. There was no main effect of power condition on life satisfaction (β = .06, t(176) = 0.69; p = .49). Results also revealed a marginally significant interaction (β = –.20, t(176) = –1.75; p = .08). Examining the interaction (see figure 2), we found that the pattern of results replicated the findings of study 2. Specifically, for individuals in the high-power condition, there was no effect of brand compatibility on life satisfaction, (t(176) = –0.02; p, NS; 95% CI = –.35 to .34). However, for individuals in the low-power condition, lower brand compatibility was associated with reduced life satisfaction (t(176) = 2.62; p < .01; 95% CI = .10 to .71) (figure 2). FIGURE 2 View largeDownload slide RESULTS FROM STUDY 3: MANIPULATED RELATIONSHIP POWER AND MEASURED BRAND COMPATIBILITY PREDICTING LIFE SATISFACTION NOTE.—For individuals who were temporarily made to feel high in relationship power, there is no effect of brand compatibility on life satisfaction. However, for individuals who were temporarily made to feel low in power, as brand compatibility decreases, so does life satisfaction. FIGURE 2 View largeDownload slide RESULTS FROM STUDY 3: MANIPULATED RELATIONSHIP POWER AND MEASURED BRAND COMPATIBILITY PREDICTING LIFE SATISFACTION NOTE.—For individuals who were temporarily made to feel high in relationship power, there is no effect of brand compatibility on life satisfaction. However, for individuals who were temporarily made to feel low in power, as brand compatibility decreases, so does life satisfaction. Discussion In this study, we expanded upon the results from the previous study by manipulating, instead of measuring, power in order to provide evidence of the causality of power in moderating this association between compatibility and life satisfaction. Our interaction effect was marginal in this study. However, as predicted, we found that for individuals temporarily led to feel high in power in their romantic relationship, perceived brand compatibility did not affect life satisfaction. On the other hand, for individuals who are temporarily led to feel low in power, we found that as brand compatibility decreases, so too did life satisfaction. These results thus conceptually replicated our findings from the previous study and demonstrated that the effects of brand compatibility on life satisfaction depend upon power in the relationship. STUDY 4 In our fourth study, we brought both members of existing romantic couples to the laboratory to examine our hypothesis from a dyadic perspective (Simpson et al. 2014). By studying both members of the couple, we were able to measure power from both partners’ perspectives, meaning that we could explore the possibility for actor-partner interaction effects (Kenny et al. 2006). Furthermore, we sought to measure brand compatibility in a more objective manner, by separately asking both members of the couple about their brand preferences and then having raters code the brands’ compatibility (as done in study 1). We sought to replicate the effects from the experimental studies in this dyadic sample—namely, that for high-power individuals, brand compatibility would not affect life satisfaction, while for low-power individuals, as brand compatibility decreases, so too would life satisfaction. Furthermore, we explored the possibility that the effect of brand compatibility on life satisfaction would be affected by an actor-partner interaction effect (Kenny et al. 2006). We speculated that the effect of brand compatibility would be strongest for low-power participants with high-power partners. Method Participants Because of the difficulty of recruiting both members of romantic couples to the laboratory, we recruited participants in two waves. The first wave (N = 54 couples) completed this study as part of a multistudy event; the second wave (N = 50 couples) completed this study as part of a different multistudy event. All participants received financial compensation in exchange for their participation. Thus, participants included both members of 104 romantic relationships, a total of 208 participants. They ranged in age from 19 to 62 years with an average age of 27.33 years (SD = 7.36) and had been in a relationship for 4.08 years on average (SD = 5.07). Measures and Procedure After indicating consent, partners were told that there were several parts to the study and were instructed to complete the measures individually. They completed the Satisfaction with Life Scale (Diener et al. 1985; α = .83), as used in the previous studies, before the other measures reported here. They also completed the same relationship power measure used in study 2 (α = .83). Each partner was asked to list his/her favorite brand in each of the same five categories used in the previous studies 1 and 3: coffee, chocolate, car, beer, and soda. As in study 1, we created an index of brand compatibility from each partner’s brand responses. Four undergraduate research assistants rated each couples’ responses within each product category (i.e., couple 1 partner 1’s favorite brand of soda compared with couple 1 partner 2’s favorite brand of soda) on the same 1 (not at all compatible) to 5 (completely compatible). As mentioned previously, using this coding scheme allows us to differentiate between responses such as “no preference” from “I don’t drink soda.” The coders’ ratings were again averaged to create a brand compatibility score for each brand category (Mbeer = 3.14, SD = 1.40; Mcar = 2.89, SD = 1.20; Mchoc = 2.71, SD = 1.27; Mcoffee = 3.40, SD = 1.52; Msoda = 3.10, SD = 1.28). Coders were highly reliable within each brand category (interrater reliability: beer α = .94; car α = .92; chocolate α = .93; coffee α = .95; soda α = .93). Brand categories were averaged to create one mean brand compatibility score for each couple, which served as the independent variable. Results Primary Analyses We hypothesized that the effects of brand compatibility on life satisfaction would depend upon power in the relationship. We first conducted analyses using a multilevel modeling approach (Kenny et al. 2006), with individuals nested within couples to account for violations of statistical independence. Life satisfaction served as our outcome variable, and power, brand compatibility, and the interaction between them served as our predictor variables. All of the predictor variables were grand mean-centered (Aiken and West 1991; Kenny et al. 2006). In these analyses, we initially included a factor for wave of data collection. The effect of wave was not significant and did not interact with our predictors; therefore, wave will not be discussed further. All analyses were conducted using the Mixed Models procedure in SPSS, with all predictors as fixed effects, and power as a level-one (participant-level) predictor and brand compatibility as a level-two (dyad-level) predictor. There was a significant main effect of power on life satisfaction, such that greater relationship power was associated with greater life satisfaction (B = .58, t(187.26) = 7.93, p < .0001), and a nonsignificant positive association between brand compatibility and life satisfaction (B = .22, t(98.22) = 1.55, p < .13). In line with our predictions, and replicating the findings in the two experimental studies, results revealed a significant interaction between power and brand compatibility on life satisfaction (B = –.31, t(180.80) = –2.40, p < .02). We found that for high-power individuals (+1SD), there was no effect of brand compatibility on life satisfaction (B = –.06, t(155.02) = –0.31; p = .76). On the other hand, for low-power individuals (–1 SD), the association of brand compatibility with life satisfaction was significant (B = .48, t(178.29) = 2.60; p = .01). Specifically, for low-power individuals, reduced brand compatibility was associated with significantly lower life satisfaction (figure 3). FIGURE 3 View largeDownload slide RESULTS FROM STUDY 4: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING LIFE SATISFACTION NOTE.—For high-power partners (1 SD above the mean), there is no effect of brand compatibility on life satisfaction. For low-power partners (1 SD below the mean), as brand compatibility decreases, so does life satisfaction. FIGURE 3 View largeDownload slide RESULTS FROM STUDY 4: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING LIFE SATISFACTION NOTE.—For high-power partners (1 SD above the mean), there is no effect of brand compatibility on life satisfaction. For low-power partners (1 SD below the mean), as brand compatibility decreases, so does life satisfaction. Actor-Partner Interdependence Model Analyses In order to explore the possibility that the effect of brand compatibility on life satisfaction would be particularly strong for individuals who saw themselves as low in power and who had partners who saw themselves as high in power, we conducted additional analyses using the actor-partner interdependence model (APIM; Kenny et al. 2006). These analyses separately estimate actor and partner effects within a multilevel modeling framework, and allow us to test the effects of brand compatibility on life satisfaction within couples of varying power combinations (i.e., high-high, high-low, low-high, and low-low). Actor and partner effects are estimated controlling for each other’s influence. As in our previous analyses, and following the norms in the field (Aiken and West 1991; Kenny et al. 2006), all predictor variables were grand mean-centered. APIM analyses regressing life satisfaction on brand compatibility, actor’s power, partner’s power, and all interaction terms revealed significant main effects for both actor power (B = .67, SE =  .07, t(185.84) = 9.11, p < .001) and partner power (B = .21, SE = .07, t(185.84) = 2.88, p = .004), and a nonsignificant trend for brand compatibility (B = .19, SE = .13, t(98.00) = 1.44, p = .15). A significant three-way interaction emerged (B = –.55, SE = .21, t(98.00) = –2.68, p = .009) (figure 4). In line with our theory, there was no effect of brand compatibility on life satisfaction (i.e., the slopes were not significantly different from zero) for high-power actors (1 SD above the mean), regardless of whether their partner was high power (1 SD above the mean) or low power (1 SD below the mean) (slopes 1 and 2 in figure 4; slope 1: B = .11, t(98.00) = 0.52, p = .61; slope 2: B = .19, t(165.12) = .74, p = .46). Further, there was not a significant difference between the slopes for high-power actors (i.e., slopes 1 and 2 were not significantly different from each other: B = –.04, t(133.07) = –0.22, p = .83). There was a negative trend for the effect of brand compatibility on life satisfaction for low-power actors (1 SD below the mean) with low-power partners (slope 4: B = –.57, t(98.00) = –1.80, p = .08); however, given the extremely small number of couples (N = 4 couples) in this cell, it is hard to interpret this result. Interestingly, and in line with our theorizing about the possible actor-partner interaction effect, there was a significant effect of brand compatibility on life satisfaction for low-power actors with high-power partners (slope 3: B = 1.03, t(165.12) = 4.05, p < .0001). Furthermore, the slope for low-power actors with high-power partners was significantly different from the slope for low-power actors with low-power partners (i.e., slope 3 vs. 4: B = .92, t(118.45) = –2.68, p < .0001). FIGURE 4 View largeDownload slide ACTOR-PARTNER INTERDEPENDENCE MODEL (APIM) RESULTS FROM STUDY 4 NOTE.—APIM analyses revealed that for low-power individuals (1 SD above the mean) with high-power partners, as brand compatibility decreases, so too does life satisfaction (slope 3). FIGURE 4 View largeDownload slide ACTOR-PARTNER INTERDEPENDENCE MODEL (APIM) RESULTS FROM STUDY 4 NOTE.—APIM analyses revealed that for low-power individuals (1 SD above the mean) with high-power partners, as brand compatibility decreases, so too does life satisfaction (slope 3). Discussion Study 4 examined in a dyadic laboratory study the association between brand compatibility, which we measured objectively by asking both members of romantic couples for their preferences, and life satisfaction in romantic couples. In line with our predictions, and replicating the findings of the two experimental studies, we found that the effects of brand compatibility on life satisfaction depended upon power in the relationship. We found that for high-power individuals, brand compatibility had no effect on life satisfaction. In other words, people with high relationship power reported similar levels of life satisfaction whether or not the couple had similar brand preferences among common consumption categories. In contrast, for low-power individuals, as brand compatibility decreased, so too did life satisfaction. In general, for these individuals, the compatibility of brand preferences was meaningful, such that those with low brand compatibility reported lower life satisfaction. Thus, these dyadic analyses replicated the findings of our prior experimental studies. In APIM analyses, it appeared that these effects of decreased brand compatibility on reduced life satisfaction were specific to low-power individuals with high-power partners. For low-power actors with high-power partners, the lower-power actor’s life satisfaction was strongly dependent on brand compatibility. On the other hand, for high-power participants, regardless of partner reports of power, life satisfaction was independent of brand compatibility. Unexpectedly, when both partners reported that they had low power, there was a trend for a negative association between brand compatibility and life satisfaction. Although interesting, we hesitate to overinterpret these patterns, as there were very few couples who were low-low power. Speculatively, when each partner is low power, it may suggest an overall level of dysfunction, as neither partner feels able to control the outcomes in the relationship. Such partnerships may be rare, and when they do occur, may be unstable. Either way, this finding highlights the lack of research on these low-low-power pairs (Simpson et al. 2014). Overall, the findings from study 4 contribute to the current research by replicating the experimental findings of studies 2 and 3 and demonstrating the interaction of brand compatibility and power on life satisfaction in both members of real-world romantic couples. STUDY 5 We propose that brand compatibility and power may lead to differences in perceptions of conflict in the relationship. Merriam-Webster defines conflict as an “opposing action of incompatibles” or a “mental struggle resulting from incompatible or opposing needs, drives, wishes, or external or internal demands” (Merriam-Webster.com). Previous research has shown that high-power partners are less likely to notice, appreciate, or consider others’ preferences, including the preferences of their low-power partners (Fiske 1993; Galinsky et al. 2006; Keltner et al. 2003). Furthermore, by definition high-power partners are more likely to control outcomes in the relationships, suggesting that they most likely do not perceive “an opposing action of incompatibles.” Consequently, they may be unaffected by differences in brand preferences within the relationship. For high-power partners, brand compatibility should not be a source of conflict. On the other hand, previous research has found that low-power partners are not only less likely to express their opinions to others (Berdahl and Martorana 2006), but also more likely to pay attention to the preferences, attitudes, and feelings of their partners (Fiske 1993; Keltner et al. 2003). When brand compatibility is low, low-power partners may therefore be more likely to perceive a “struggle resulting from incompatible or opposing needs, drives, [and] wishes.” They are also more likely to “lose out” to their higher-power partners, which may lead to viewing these decisions as unfair. In order to test whether low-power partners view decisions as unfair when brand compatibility is low, we conducted an online study in which we manipulated perceptions of power (high vs. low) and brand compatibility (high vs. low) and measured perceived fairness as the outcome variable. In line with predictions, individuals in the low-power/low-brand-compatibility condition reported significantly less (at the p < .05 level) perceived fairness with how decisions would be made in the relationship compared with the other three conditions (see the web appendix for additional details regarding this study). As perceived unfairness has been shown to be an underlying driver of relationship conflict (Grote and Clark 2001), we propose that brand compatibility and power interact to influence perceptions of conflict in the relationship. Specifically, we predict that perceptions of conflict will be greatest when brand compatibility and power are both low. We test this hypothesis—that low brand compatibility is more likely to be a source of perceived conflict for low-power partners—in the current study. In keeping with much prior research on power, we utilized a scenario paradigm. Research has shown that manipulating perceptions of power can influence perceptions of other dynamics in social contexts (e.g., high-power individuals have been shown to project their attitudes and feelings on to others; Keltner et al. 2003; Overbeck and Droutman 2013). A scenario method allows us to more cleanly manipulate both perceptions of power and of brand compatibility. In addition, we included a measure of autonomy threat to rule out an alternative process explanation. For parsimony, we present the findings related to conflict here. (Findings related to autonomy threat are provided in the web appendix.) Method Participants Six hundred nine participants (40% men) from Amazon’s Mechanical Turk completed the study in exchange for financial compensation. Participants ranged in age from 18 to 74 years with an average age of 35.43 years (SDage = 11.46). Measures and Procedure After indicating consent, participants were asked to read a scenario and imagine that they were in it. They were told they would be asked questions about the scenario. Participants were randomized to one of four conditions, in which we varied brand compatibility and relationship power, and were presented with a scenario that corresponded to that condition. We manipulated brand compatibility by telling participants they either liked the same or opposing brands as their partner. We manipulated relationship power by including statements from the modified power measure used in the previous studies (Anderson et al. 2012), which we presented as if the participant reported feeling that way about the described relationship. (See the web appendix for the conditions.) While imagining themselves in the scenario, participants were next asked to indicate to what extent they agreed or disagreed with several statements. All participants rated their agreement on a seven-point Likert scale (anchored with Strongly Disagree and Strongly Agree). As our measure of perceived relationship conflict, participants answered to what extent they agreed with the following two items: “My partner and I probably argue a lot” and “My partner and I probably have a lot of conflict” (α = .95). Participants completed other items, including autonomy threat (see the web appendix), filler items, and demographic variables. Results Eight participants indicated they had taken the survey multiple times and were excluded from the following analyses. We hypothesized that participants in the low-brand-compatibility/ low-power condition would report the greatest levels of perceived conflict. In order to test this hypothesis, we conducted an ANOVA with brand compatibility condition, power condition, and their interaction as the predictor variables and conflict as the outcome variable. Results revealed a main effect for brand compatibility condition (F(1, 597) = 30.52, p < .0001), such that the low-brand-compatibility condition reported greater conflict, and a main effect for power (F(1, 597) = 82.37, p < .0001), such that the low-power condition reported greater perceived conflict. Importantly, results also revealed a significant interaction (F(1, 597) = 3.87, p < .05). Planned contrasts indicated that each of the groups was significantly different from the others, with the low-power/low-brand-compatibility condition reporting significantly greater perceived conflict (M = 4.10) than not only the high-power conditions (vs. high-power/low-brand-compatibility condition M = 3.26, t(597) = 4.96, p < .0001; vs. high-power/high-brand-compatibility condition M = 2.47, t(597) = 10.37, p < .0001), but also the low-power/high-brand-compatibility condition (M = 3.68, t(597) = 2.52, p = .01) (figure 5). (Note: Because all of the other studies involve individuals currently in a relationship, while here we asked even those outside of a relationship to imagine themselves in the scenario, we repeated these analyses controlling for relationship status, and the results remain the same.) FIGURE 5 View largeDownload slide RESULTS FROM STUDY 5: PERCEIVED RELATIONSHIP CONFLICT AS A FUNCTION OF MANIPULATED BRAND COMPATIBILITY AND MANIPULATED RELATIONSHIP POWER NOTE.—As predicted, the low-power/low-brand-compatibility condition reported significantly greater perceptions of relationship conflict than the low-power/high-brand-compatibility condition and both of the high-power conditions. FIGURE 5 View largeDownload slide RESULTS FROM STUDY 5: PERCEIVED RELATIONSHIP CONFLICT AS A FUNCTION OF MANIPULATED BRAND COMPATIBILITY AND MANIPULATED RELATIONSHIP POWER NOTE.—As predicted, the low-power/low-brand-compatibility condition reported significantly greater perceptions of relationship conflict than the low-power/high-brand-compatibility condition and both of the high-power conditions. Discussion In a recent article, conflict was predicted by spouses’ differing financial preferences (Rick, Small, and Finkel 2011); extending that work here, we find evidence to support the notion that incompatibility in brand preferences also predicts perceived conflict. These findings suggest that the consistent pattern of evidence across studies demonstrating that low brand compatibility decreases life satisfaction for low-power partners may be related to relationship conflict. In other words, perceived conflict may be the mechanism through which brand compatibility and power interact to influence life satisfaction.4 STUDY 6 In our sixth study, a dyadic study, we again examined brand compatibility, power, and life satisfaction in both members of romantic couples, and we further explore the role of perceived conflict. In line with study 5’s findings, we hypothesized that for low-power partners, low brand compatibility would be associated with increased reports of perceived conflict, given that these individuals are unable to control outcomes, and thus, brand choices likely create some notable problems for their everyday consumption experiences. Previous research has demonstrated that perception of relationship conflict has been shown to predict reductions in general satisfaction (Gordon and Chen 2016; Gustavson et al. 2016). Therefore, we further hypothesized that for low-power individuals, increased perceived conflict would, in turn, drive reduced life satisfaction. Method Participants Study 6 used data from a multiwave study examining romantic relationships conducted at a large public university in Canada. Participants were recruited through advertisements placed in the university newspaper, flyers posted around campus, and presentations in large undergraduate classes. Couples were eligible to participate in the study if both partners indicated they were in an exclusive heterosexual romantic relationship, had been involved for at least six months, and were 18 years of age or older. All participants received financial compensation in exchange for their participation; for completion of the entire multipart longitudinal study, individuals received $100. Participants were both members of romantic relationships (N = 139 couples; Mage = 20.80, SD = 2.16). Reflecting the ethnic diversity of the university campus, participants were majority Caucasian (56.5%) and Asian (33.5%), with 10% reporting mixed or other races. Measures and Procedure Partners took part in a multiwave study with three main components: an initial online battery of questionnaires, an in-lab session with videotaped conversations that occurred one to two weeks later, and then three online follow-up questionnaires, each approximately one month apart, beginning one month after the lab session. Most of the materials and procedures of this study are irrelevant to the current hypothesis, as the study was designed to explore goal dynamics over time and how goals are related to relationship processes (Fitzsimons, Finkel, and vanDellen 2015). All the relevant measures to test our hypothesis about brand compatibility are described here (see the web appendix for any related measures that we did not use in our analyses). During the first online follow-up questionnaire (taken two months after the lab session), we included measures to test the current hypotheses—that is, measures of brand compatibility, relationship power, perceived conflict, and life satisfaction. Because retention is challenging in longitudinal studies, especially ones that require both members of a dyad, we used shortened versions of the relationship power, conflict, and life satisfaction scales. (Note: The measures were completed in that order and the brand compatibility measure was collected after the other variables.) Relationship power was measured by three items taken from the same relationship power measure used in prior studies. The items included: “I can get my partner to do what I want,” “I think I have a great deal of power,” and “My ideas and opinions are often ignored” (reverse-scored) (α = .66). Perceived conflict was measured by five items that measured the degree of conflict partners perceived in their relationship over the course of the past month (Braiker and Kelley 1979). The first two items—“How often did you and your partner argue with each other this month?” and “How often did you feel angry or resentful toward your partner this month?”—were measured on a 1 (never) to 6 (constantly) scale. The other three items—“When you and your partner argued, how serious were the problems or arguments?” “To what extent did you wish you could change things about your partner this month?” and “To what extent did you communicate negative feelings toward your partner (e.g., anger, dissatisfaction, frustration) this month?”—were measured on a 1 (not at all) to 7 (extremely) scale. Items were standardized before being combined into one measure of conflict (α = .83). Life satisfaction was measured with three items taken from the same Diener et al. (1985) measure used in prior studies, “In the last month, in most ways, my life has been close to ideal,” “The conditions of my life were excellent this month,” and “I am satisfied with my life right now,” that were rated on a 1 (strongly disagree) to 7 (strongly agree) scale (Diener et al. 1985; α = .86). As in studies 1 and 4, each partner within a couple was asked to list his/her favorite brand in the same brand categories: coffee, chocolate, car, beer, soda. (In addition, in this study, partners were also asked about their favorite brand of toothpaste. To be consistent, we report the analyses using the same five brand categories as in the other studies. Results remain the same when toothpaste is included.) As in the previous studies, we created a measure of brand compatibility from each pairing of partners’ brand responses. Four undergraduate research assistants rated the brands on the same 1 (not at all compatible) to 5 (completely compatible) scale as used in the previous studies. The coders’ scores were again averaged to create a brand compatibility rating for each brand category (Mbeer = 2.63, SD = 1.46; Mcar = 2.48, SD = 1.16; Mchoc = 2.63, SD = 1.02; Mcoffee = 3.56, SD = 1.52; Msoda = 2.59, SD = 1.26). Raters were highly reliable within each brand category (interrater reliability: beer α = .92; car α = .89; chocolate α = .84; coffee α = .95; soda α = .92). As in the previous studies, brand categories were then averaged to create one mean brand compatibility score. Results In this study, we investigate the role that brand compatibility and power have on conflict, which we predict will in turn affect life satisfaction. Specifically, we hypothesize that for low-power partners, as brand compatibility increases, conflict will decrease. Additionally, we hypothesize that conflict will in turn predict life satisfaction, suggesting that for low-power partners, as brand compatibility increases, conflict decreases and the reduction in conflict is associated with greater life satisfaction. We test our predictions in several steps. First, we examine whether power affects the link between brand compatibility and conflict. Because this is dyadic data, and thus violates assumptions of independence required for standard regression analyses, we again conducted a multilevel analysis (Kenny et al. 2006), with individuals nested within couples. Conflict served as our outcome variable, and power, brand compatibility, and the interaction between them served as our predictor variables. As in the previous study, all of the predictor variables were grand mean-centered. All analyses were conducted using the Mixed Models procedure in SPSS, with all predictors as fixed effects, and power as a level-one (participant-level) predictor and brand compatibility as a level-two (dyad-level) predictor. The results revealed a significant negative association between partners’ reports of power and conflict (B = –.12, t(186.51) = –2.70, p = .008), indicating that as partners felt less power in their relationship, they perceived more conflict. Importantly, as predicted, results revealed a significant interaction between power and brand compatibility on perceived conflict (B = .16, t(181.29) = 2.14, p = .03). Examining the interaction, we find that for high-power partners (+1 SD), brand compatibility is nonsignificantly positively related to conflict (B = .15, t(217.90) = 1.33, p = .19). On the other hand, for low-power partners (–1 SD), brand compatibility is negatively related to perceived conflict (B = –.17, t(218.90) = –1.50, p = .14) (figure 6). Further examining the interaction, we find that low brand compatibility is significantly associated with greater perceived conflict for partners whose power scores are less than 3. FIGURE 6 View largeDownload slide RESULTS FROM STUDY 6: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING PERCEIVED CONFLICT NOTE.—Results from study 6: For high-power (+1 SD) partners, as brand compatibility increases, conflict increases. For low-power (–1 SD) partners, as brand compatibility increases, conflict decreases. FIGURE 6 View largeDownload slide RESULTS FROM STUDY 6: BRAND COMPATIBILITY AND RELATIONSHIP POWER IN COUPLES PREDICTING PERCEIVED CONFLICT NOTE.—Results from study 6: For high-power (+1 SD) partners, as brand compatibility increases, conflict increases. For low-power (–1 SD) partners, as brand compatibility increases, conflict decreases. Next, we examined whether conflict was related to life satisfaction. A multilevel analysis (with individuals nested within couples) revealed a significant, negative relationship between conflict and life satisfaction (B = –.41, SE = .11, t(243.43) = –3.73, p < .0001), indicating that as conflict in the relationship increased, life satisfaction decreased. In order to test our prediction that an indirect pathway would exist between brand compatibility and power, conflict, and life satisfaction, we conducted an analysis using the Monte Carlo Method for Assessing Mediation (Selig and Preacher 2008). Previous research has indicated that the Monte Carlo method is the better method for assessing indirect effects over the Sobel test and is about as good as nonparametric bootstrapping (Hayes and Scharkow 2013). We used the coefficients and the standard errors from the multilevel model parameters for the interaction term (B = .16, SE = .07) and the mediator term (B = –.40, SE = .11). The 95% confidence interval for the distribution of the indirect effect did not contain zero [–.1396, –.007558], thus supporting the existence of an indirect pathway from brand compatibility and power to life satisfaction through conflict. Following Zhao, Lynch, and Chen (2010), and because this is multilevel data that violates statistical assumptions of independence, we investigated the effect of the predictor variables to the mediator, and the effect of the mediator to the outcome variable. In line with predictions, the reported indirect pathway does support our overall theorizing that the repeated, day-to-day aspect of brand compatibility is driving the increased conflict for low-power partners and leading to the reduction in life satisfaction. However, we also tested the direct pathway on life satisfaction. Although brand compatibility (B = .43, t(118.73) = 3.10, p = .002) and relationship power (B = .27, t(203.53) = 3.27, p = .001) had significant positive effects on life satisfaction, the interaction between brand compatibility and power on life satisfaction was not significant (B = .08, t(200.30) = 0.61, p > .25). To investigate why the direct effect might not have been as strong as in previous studies, we conducted an internal analysis where we examined whether there were differences in relationship type in this study compared to the other studies. In all other studies, participants were older and the couples were likely to be living with their partner (e.g., 100% were living together in study 1, 73% in study 2, 66% in study 3, 63% in study 4). This particular study was run at a university using an undergraduate sample that was much younger (median age 20.0 years) than the other samples. Interestingly, 26% of the couples were in long-distance dating relationships. Given that our proposed theory suggests that for brand compatibility to matter, there should be repeated interactions at the day-to-day level, this may play a role. In an analysis that includes only the couples who are not long distance, the interaction of brand compatibility and power on life satisfaction is marginally significant (B = .29, t(141.03) = –1.89, p = .06); and the effect of the interaction of brand compatibility and power on conflict is significant (B = .29, t(126.99) = 3.72, p < .0001). Of course, this post-hoc speculation needs to be further explored in future research, but it does provide additional insight to our theory. Discussion In our final study, using both members of the couple, we tested our predictions that brand compatibility and power are related to conflict within the relationship. It is important to note that conflict and brand compatibility were not directly correlated (B = –.01, t(122) = –.17, p = .87), suggesting that brand compatibility is not just another measure of conflict in the relationship. Importantly, and in line with our hypotheses, we found that brand compatibility and power significantly interacted to predict conflict. Results from the mediation analysis further suggested the presence of an indirect pathway from the interaction of compatibility and power through conflict to life satisfaction. In other words, high-power partners do not perceive an increase in conflict as a result of low brand compatibility; however, for low-power partners, when brand compatibility is low, conflict is increased, and this increase in perceived conflict is associated with reduced life satisfaction. In addition, in this study, we found evidence in our internal analysis to suggest that it is the repeated day-to-day interactions of couples in terms of brand compatibility and power that is influencing relationship conflict and life satisfaction. GENERAL DISCUSSION This set of studies presents a new construct connecting consumer experiences with close relationships—brand compatibility—and explores its significance for psychological well-being. Across several studies, examining both individuals and couples, we find that brand compatibility within a close relationship predicts life satisfaction, but, importantly, only for individuals who experience low power in their relationship. In other words, the effect of brand compatibility on life satisfaction depends upon power in the relationship. For high-power individuals, brand compatibility is not related to life satisfaction, while for low-power individuals, brand compatibility positively predicts life satisfaction: when brand compatibility is higher, life satisfaction is higher; however, when brand compatibility is low, life satisfaction is reduced. A strength of the studies is that they use different methods to manipulate and measure both brand compatibility and power, allowing us to draw causal conclusions about their roles in predicting life satisfaction, and to provide more assurance of the generalizability of the effects. Brand compatibility was measured as a perception (study 3), was manipulated (studies 2 and 5), and was measured dyadically to obtain a clear and objective indicator of compatibility (studies 1, 4, and 6). Similarly, power was manipulated (studies 3 and 5) and measured (studies 2, 4, and 6). Contributions and Implications Our findings examine how consumer behavior within the context of close relationships affects overall life satisfaction. As close relationships are an integral part of life that predict many meaningful outcomes including mortality rates, financial well-being, health, and happiness (Berkman 1995; Cohen 2004; Keltner et al. 2003; Kiecolt-Glaser, Gouin, and Hantsoo 2010; Kiecolt-Glaser et al. 2005; Liu and Reczek 2012; Marmot 2004; Waite and Gallagher 2002), it is beneficial for consumer researchers to explore links between close relationships and consumer variables. This research makes several contributions to the study of relationships. It is the first to look at how compatibility—of any kind—interacts with power in close relationships. Surprisingly, no research on compatibility in close relationships has examined how it may be more important for some partners than others, instead highlighting the general benefits of compatibility (Bramlett and Mosher 2002; Decuyper, De Bolle, and De Fruyt 2012; Gaunt 2006; Heaton and Pratt 1990; Lucas et al. 2004b; Murray et al. 2002) or questioning the value of compatibility for all partners (Dyrenforth et al. 2010; Houts et al. 1996; Montoya et al. 2008; Tidwell et al. 2013). In this work, we suggest that compatibility is crucial for partners who are low in power, but far less so for those who are high in power. Using the context of brand preferences, we found that compatibility has a predictive influence for individuals who chronically feel low in power and for those who were temporarily led, as a result of an experimental manipulation, to feel low in power. This work thus contributes to prior work on compatibility in close relationships by highlighting the importance of considering other relationship variables, like power, when studying the role of compatibility. Furthermore, the findings suggest that even seemingly trivial forms of compatibility—whether partners report liking the same soda and coffee brands—can affect important psychological phenomena. Life satisfaction is a complex evaluation, but most research studying it has looked at certainly more profound predictors, such as marriage quality, religion, health, self-esteem, job satisfaction, and unemployment (Adams, King, and King 1996; Diener and Diener 2009; Diener et al. 2010; Heller, Watson, and Ilies 2004; Lim and Putnam 2010; Lucas et al. 2004a; Mroczek and Spiro 2005). Here we show that for some people—those low in power—even something as seemingly mundane as brand compatibility within a close relationship influences people’s satisfaction with their lives. Although we studied these ideas only within the brand context, we suggest that they may also be worth studying with other low-level forms of day-to-day compatibility, such as activity, spending, and parenting preferences. If couples are incompatible on dimensions that matter in day-to-day life, our findings here suggest that low-power partners will suffer as a result. In taking a dyadic perspective on consumers’ preference for brands, the studies begin to answer calls for a more social and interpersonal perspective on consumer behavior (Simpson et al. 2012), and for the study of power in consumer behavior more specifically (Brick and Fitzsimons 2016; Rucker et al. 2012). Although research on social power has long been a large and influential topic in psychology, organizational behavior, sociology, and other fields in social science (see Magee and Galinsky 2008 for a review), power in the consumer context has historically received less attention. As Rucker and colleagues noted, “Despite the long-recognized value and experimental study of power across the social sciences, the construct of power has been largely absent from efforts to understand consumer behavior. This is somewhat surprising given that different degrees of power exist and arise in consumers’ everyday activities” (Rucker et al. 2012). Indeed, recent research supports the relevance of experiences of social and interpersonal power for consumer behavior (Jiang et al. 2014; Kim and McGill 2011; Rucker and Galinsky 2008). Our findings thus contribute to the literature on power, and more specifically relationship power, within consumer behavior. This research also has implications in terms of brand loyalty. Individuals’ brand preferences may shift as a function of their close relationships and close relationship status. One could imagine being loyal to particular brands when one is single, and shifting or changing those brand preferences once one enters into a relationship. In addition, as individuals change their relationship status, moving from single to engaged to married, brand preferences and brand loyalties may change. It would be interesting to examine how an individual’s brand preferences shift over the course of his or her lifetime as a function of relationship status. Furthermore, what happens when an individual ends a close, romantic relationship if he/she has shifted brand preferences? Does the individual go back to using the brands he/she preferred from before the relationship, or continue using the brands that he/she used while in a relationship? On a practical note, the current findings have implications for matchmaking, including online dating services. Online dating is now a major multibillion-dollar industry. For example, IAC/InterActiveCorp., the company that owns Match.com, reported over $1.2 billion in revenue last year for its dating websites alone (JMP Securities Initiation Report, http://ir.iac.com/results.cfm; see Supplemental Financial Information and Metrics). Recent research has highlighted the ways in which online dating differs from conventional offline dating (Finkel et al. 2012), and taking consumer preference information into account may be another way. Very few, if any, of these online dating websites ask about everyday consumer preferences and use this information in their matchmaking algorithms. However, our research suggests that one aspect of consumer preferences, namely brand preferences, can affect relationship conflict and well-being. We are not suggesting that brand preferences are the only form of consumer behavior that matchmakers should take into account, but the present research suggests that when designing future questions and surveys, matchmakers, including online services, should consider including items to assess consumer behavior at the day-to-day level, such as brand preferences, to better meet their customers’ needs. For those who will ultimately be high in power, consumer preferences like brand compatibility may be irrelevant, but for those who will ultimately be low in power, it would be beneficial for them to consider. Limitations and Future Research There are several limitations to the current work. One important limitation is that we only ask about brand preferences and do not specifically examine brand use. A potential problem with this could be that couples trade off or engage in multibrand use within the household, and thus their preferences really do not matter for day-to-day life. To explore this, we conducted a simple online study in which we asked individuals who live with their partner to count the total number of brands that are currently in their household across various product categories. Specifically, 406 participants (45% men) were given the same definition of a brand as used throughout the current research. They then counted and listed each of the brands currently in or used by their household within the following product categories: soda, coffee, beer, chocolate, car, toothpaste, toilet paper, laundry detergent, grocery store, orange juice, ketchup, and paper towels. The mean number of brands within each product category was 1.06 (SD = 0.27). The product categories with the highest mean were grocery store (M = 1.83, SD = 0.71) and car (M = 1.54, SD = 0.73), and for all categories except grocery store, the modal response was 0 or 1. Overall, this survey suggests that, on average, at least for the product categories studied, couples tend to purchase at most one brand for the household. Because couples are generally choosing just one brand for the household, brand preferences do matter. Over time, incompatible brand preferences may indeed be a source of conflict for couples, particularly if one partner is continually getting his or her way at the expense of the other partner. Another limitation of the current work may be the robustness of the results. In a couple of studies, the p values were relatively weak. Thus, readers may be concerned about the strength of the evidence for the key interaction. Readers can be reassured that all reporting was fully transparent, as explained in the introduction. All measures and conditions were reported (with exceptions noted), and all attempts were made to follow best-practice guidelines for transparency in research collection and reporting. Because of the relatively weak p values, we also sought additional evidence in the form of a direct replication (study 2B). Despite these efforts, it is reasonable to question the strength of the evidence provided here. We acknowledge this as a potential limitation of the current work. That said, we do believe that the consistent pattern of findings across studies, using multiple methods, including measuring and manipulating brand compatibility and power using various coding schemes and analyses, supports our hypothesis. Further, we hope that our novel findings encourage future research on this important topic. Finally, because it is difficult to recruit both partners of a couple to come to the lab together, we used data in studies 4 and 6 from multistudy events examining romantic relationships. Using data from multistudy events limits control over the order of administration of focal measures. In addition, it also increases the number of other measures to which participants are exposed. One direction for future research could be to further study couples who are both low-low or high-high in relationship power. In our research, we focused on perceptions of relationship power. Accordingly, and as discussed in study 4, this means that within a couple there can be two members who are high-high in power and two members who are low-low in power. What does it mean for both partners to be high in perceptions of relationship power? On the one hand, it could mean that one partner is oblivious. On the other hand, if power is defined as the ability to influence others (Anderson and Galinsky 2006; Emerson 1962; Keltner et al. 2003; Magee and Galinsky 2008), it suggests that both partners believe that they have control in the relationship. If both partners perceive control, or power, it suggests the relationship is an equitable one, and close relationships research has highlighted that the most satisfying relationships are indeed those viewed as equitable (Gray-Little and Burks 1983). Perhaps, then, couples who are high-high in relationship power are happy because they believe they have equal control in the relationship, regardless of the actual breakdown and perhaps regardless of the specific domain. Future research could also investigate whether couples who are high-high in relationship power pursue complementary goal strategies. In other words, do they alternate control across domains that are of differing importance or do they alternate within a domain? In terms of low-low relationships, there were very few couples who were low-low power, and we hesitate to overinterpret this pattern. However, this finding does suggest another area for future research. Overall, our results indicate that perceived power in relationships is not always zero-sum, which is also an interesting area for future research. Future research could also further investigate the role of relationship power in consumer behavior. Recent research has demonstrated that power influences purchasing decisions (Rucker et al. 2012; Rucker, Hu, and Galinsky 2014). For example, Jiang et al. (2014) found that increased perceptions of power result in greater switching behavior. This research, like much of the previous research, has focused on the level of the individual, whereas the present research focuses on partners within romantic relationships. Thus, it would be interesting to examine the role of power in relationships and switching behavior. Although the previous research has found that high-power individuals are more likely to engage in switching behavior (Jiang et al. 2014), a close relationships perspective suggests that low-power partners may be more likely to switch and adopt their high-power partners’ preferences. Longitudinal research could examine how brand preferences shift over the course of a relationship. A related area for future research could examine how brand preferences change over the course of one’s lifetime as a result of relationship status and power. Previous research has indicated that high-power individuals tend to have more stable self-concepts over time (Kraus, Chen, and Keltner 2011), and so it might be that individuals who tend to have power in relationships do not shift their brand preferences. However, for those who generally yield power to their partners, perhaps their brand preferences shift back to the brands they preferred before they entered into the relationship. Alternatively, maybe they do not change brand preferences until they enter into a new relationship. Future research could examine this topic and how it relates to brand loyalty over a consumer’s lifetime. The manner in which consumers make decisions with other people can be different from the manner in which consumers make decisions by themselves (Park 1982). The current research supports this idea and suggests that the brands people consume when they are with their romantic partner may be different from ones that they select when they are alone, especially for low-power partners. One question that arises is whether evaluations of brands change when a partner, especially a high-power partner, is present. Would partners who are low power, and who generally prefer a different brand, change their evaluations to be in line with their partner’s views when their partner is present? In other words, would lower-power partners “go along to get along”? Previous research suggests that they would (Fiske 1993; Keltner et al. 2003; Simpson et al. 2014), and our research suggests that they may subsequently be less happy. Another question is how do brand compatibility and relationship power interact to influence consumers’ brand evaluations depending upon whether only one or both partners like the brand? Future research could tease apart differences in brand preferences, evaluations, and selections when low-power individuals are with versus without their partners, and whether the decision itself is shared or made by one member of the couple. Conclusion Individuals often use and consume brands in the presence of their close partners. Yet very little is known about the role that brand preferences play in close relationships. In the present research, we seek to provide a preliminary examination of this topic by examining how brand compatibility, or the extent to which individuals have similar brand preferences as their partners, influences life satisfaction. For low-power partners, it seems that brand compatibility can reliably predict how satisfied they feel with their lives. For those who tend to yield power to their romantic partners, the brand preferences of their partners take on a rather surprising significance, suggesting the importance of brand preferences in everyday life. If you tend to yield power to your romantic partners, perhaps you should first ensure that they share your love of Coke, or you may find yourself drinking Pepsi yet again. DATA COLLECTION INFORMATION For each study involving both members of the couple, sample sizes were based on subject availability as well as the needs of unrelated research projects that were run in conjunction with data collection. For studies using online participants, sample sizes were determined a priori. No additional data were collected after data analyses began with the exception of the additional conditions for study 2 as requested by a reviewer in the previous round of revisions. The data from study 1 were collected from a local farmer’s market in November 2015 by research assistants under the supervision of the first author. Data for studies 2 and 3 were collected from Amazon’s Mechanical Turk in January 2015 by the first author. Data from replication study 2 were collected in October 2016. The couple data from study 4 were collected during two waves of data collection from April 2014 to November 2014 under the supervision of a lab manager and research assistants. Data for study 5 and the introductory study to study 5 were collected from Amazon’s Mechanical Turk in January 2016 and September 2016. Data from study 6 were collected by research assistants under the supervision of the second author as part of a multiwave study examining romantic relationships. The data for studies 2, 3, and 5 were analyzed by the first author. The data for studies 1, 4, and 6 were analyzed primarily by the first author with support from the second author. The authors are grateful for the constructive comments and helpful suggestions provided by the editor, associate editor, and three anonymous reviewers. In addition, they would like to thank Jim Bettman for his invaluable feedback. APPENDIX Relationship Power Measure Participants are asked to indicate to what extent they agree with the following statements in regard to their relationship. Items are rated on a seven-point Likert scale anchored with “Strongly Disagree” and “Strongly Agree.” Items marked with an * are reverse-scored. I can get my partner to listen to what I say. My wishes do not carry much weight.* I can get my partner to do what I want. Even if I voice them, my views have little sway.* I think I have a great deal of power. My ideas and opinions are often ignored.* Even when I try, I am not able to get my way.* If I want to, I get to make the decisions. Footnotes 1 Some may wonder why we oriented compatibility around brand family (e.g., Coke products) rather than around attribute level similarity (e.g., diet soda). We suggest both a conceptual and an empirical response to this important question. From a conceptual point of view, research has demonstrated that individuals form relationships and attachments with brands themselves—that is, brand families (Aaker, Fournier, and Brasel 2004; Aggarwal 2004; Carroll and Ahuvia 2006; Fournier 1998; Park et al. 2010; Thomson et al. 2005) as opposed to attributes. In addition, many structural factors and environmental cues lead individuals to cognitively organize products around brand families. For example, packaging elements are often consistent across products within brand families (the shape of the bottle for Coca-Cola products; the color scheme across all Colgate toothpastes, etc.); most product categories in grocery stores are organized by brand family as opposed to attribute, even more so in categories where distribution is done by the brand directly (e.g., Lindt’s dark chocolate is next to Lindt’s white chocolate, Pepsi products are grouped together separately from Coke products); and promotional efforts within stores often operate at the level of the brand family rather than at an attribute level (e.g., “buy three for $12”–style offers or “buy two, get one free” offers). From an empirical perspective, note that we employed alternative coding approaches across the studies, one of which gave no specific instructions in terms of how the coders should define brand compatibility. The coders simply rated which brands they thought were compatible with which other brands. This coding approach yielded extremely similar results to our original compatibility coding as well as to the specific coding scheme that gave individual examples within each product category. If people naturally judged brand compatibility, or similarity, according to product attributes, this unstructured approach should have yielded a different pattern of results from the other two approaches. That was not the case—results were consistent across coding schemes. 2 We repeated the analyses using various coding schemes to ensure that the results are not driven by the use of any one specific type of coding scheme. All results are similar. See the web appendix for additional models. 3 To examine the reliability of these effects, we conducted a direct replication of study 2. The findings replicated. The full results are reported in the web appendix. 4 We conducted an additional experimental study to further investigate the role of conflict in current romantic relationships depending upon brand preferences and relationship power. The results of this additional study revealed that, unlike their higher-power counterparts, lower-power partners reported greater conflict in their current relationship over which brands to buy. Please see the web appendix for additional information regarding this study. REFERENCES Aaker David A., Keller Kevin Lane ( 1990), “Consumer Evaluations of Brand Extensions,”  Journal of Marketing , 27– 41. Aaker Jennifer, Fournier Susan, Brasel S. Adam ( 2004), “When Good Brands Do Bad,” Journal of Consumer Research  31 1, 1– 16. Google Scholar CrossRef Search ADS   Acitelli Linda K, Douvan Elizabeth, Veroff Joseph ( 1993), “Perceptions of Conflict in the First Year of Marriage: How Important Are Similarity and Understanding?” Journal of Social and Personal Relationships , 10 1, 5– 19. Google Scholar CrossRef Search ADS   Adams Gary A., King Lynda A., King Daniel W. ( 1996), “Relationships of Job and Family Involvement, Family Social Support, and Work–Family Conflict with Job and Life Satisfaction,” Journal of Applied Psychology , 81 4, 411. Google Scholar CrossRef Search ADS   Aggarwal Pankaj ( 2004), “The Effects of Brand Relationship Norms on Consumer Attitudes and Behavior,” Journal of Consumer Research , 31 1, 87– 101. Google Scholar CrossRef Search ADS   Aiken Leona S., West Stephen G. ( 1991), Multiple Regression: Testing and Interpreting Interactions , Thousand Oaks, CA: Sage. Aknin Lara B., Barrington-Leigh Christopher P., Dunn Elizabeth W., Helliwell John F., Burns Justine, Biswas-Diener Robert, Kemeza Imelda, Nyende Paul, Ashton-James Claire E., Norton Michael I. ( 2013), “Prosocial Spending and Well-Being: Cross-Cultural Evidence for a Psychological Universal,” Journal of Personality and Social Psychology , 104 4, 635. Google Scholar CrossRef Search ADS PubMed  Anderson Cameron, Galinsky Adam D. ( 2006), “Power, Optimism, and Risk‐Taking,” European Journal of Social Psychology , 36 4, 511– 36. Google Scholar CrossRef Search ADS   Anderson Cameron, John Oliver P., Keltner Dacher ( 2012), “The Personal Sense of Power,” Journal of Personality , 80 2, 313– 44. Google Scholar CrossRef Search ADS PubMed  Arrindell W. A., Meeuwesen L., Huyse F. J. ( 1991), “The Satisfaction with Life Scale (Swls): Psychometric Properties in a Non-Psychiatric Medical Outpatients Sample,” Personality and Individual Differences , 12 2, 117– 23. Google Scholar CrossRef Search ADS   Bagchi Rajesh, Cheema Amar ( 2013), “The Effect of Red Background Color on Willingness-to-Pay: The Moderating Role of Selling Mechanism,”  Journal of Consumer Research,   39 5, 947– 60. Google Scholar CrossRef Search ADS   Belk Russell ( 1988), Possessions and Self, Hoboken , NJ: Wiley Online Library. Berdahl Jennifer L., Martorana Paul ( 2006), “Effects of Power on Emotion and Expression During a Controversial Group Discussion,” European Journal of Social Psychology , 36 4, 497. Google Scholar CrossRef Search ADS   Berger Jonah, Heath Chip ( 2007), “Where Consumers Diverge from Others: Identity Signaling and Product Domains,” Journal of Consumer Research , 34 2, 121– 34. Google Scholar CrossRef Search ADS   Berkman Lisa F. ( 1995), “The Role of Social Relations in Health Promotion,” Psychosomatic Medicine , 57 3, 245– 54. Google Scholar CrossRef Search ADS PubMed  Berscheid Ellen, Snyder Mark, Omoto Allen M. ( 1989), “The Relationship Closeness Inventory: Assessing the Closeness of Interpersonal Relationships,” Journal of Personality and Social Psychology , 57 5, 792. Google Scholar CrossRef Search ADS   Blossfeld Hans-Peter ( 2009), “Educational Assortative Marriage in Comparative Perspective,” Annual Review of Sociology , 35, 513– 30. Google Scholar CrossRef Search ADS   Blum Jeffrey S., Mehrabian Albert ( 1999), “Personality and Temperament Correlates of Marital Satisfaction,” Journal of Personality , 67 1, 93– 125. Google Scholar CrossRef Search ADS   Braiker Harriet B., Kelley Harold H. ( 1979), “Conflict in the Development of Close Relationships,” in Social Exchange in Developing Relationships , ed. Burgess Robert L., Huston Ted L., Cambridge, MA: Academic Press, 135– 68. Google Scholar CrossRef Search ADS   Bramlett Matthew D., Mosher William D. ( 2002), “Cohabitation, Marriage, Divorce, and Remarriage in the United States,” Vital Health Statistics , 23 22, 1– 87. Brick Danielle J., Fitzsimons Gavan J. ( 2016), “Oppositional Brand Choice: Using Brands to Respond to Relationship Frustration,” Journal of Consumer Psychology , 27 2, 257– 63. Google Scholar CrossRef Search ADS   Burroughs James E., Rindfleisch Aric ( 2002), “Materialism and Well-Being: A Conflicting Values Perspective,” Journal of Consumer Research , 29 3, 348– 70. Google Scholar CrossRef Search ADS   Byrne Donn Erwin ( 1971), The Attraction Paradigm , Vol. 11, Cambridge, MA: Academic Press. Carroll Barbara A., Ahuvia Aaron C. ( 2006), “Some Antecedents and Outcomes of Brand Love,” Marketing Letters , 17 2, 79– 89. Google Scholar CrossRef Search ADS   Clee Mona A., Wicklund Robert A. ( 1980), “Consumer Behavior and Psychological Reactance,” Journal of Consumer Behavior , 6 4, 389– 405. Google Scholar CrossRef Search ADS   Cohen Sheldon ( 2004), “Social Relationships and Health,” American Psychologist , 59 8, 676. Google Scholar CrossRef Search ADS PubMed  Cohn Michael A., Fredrickson Barbara L., Brown Stephanie L., Mikels Joseph A., Conway Anne M. ( 2009), “Happiness Unpacked: Positive Emotions Increase Life Satisfaction by Building Resilience,” Emotion , 9 3, 361. Google Scholar CrossRef Search ADS PubMed  Corfman Kim P., Lehmann Donald R. ( 1987), “Models of Cooperative Group Decision-Making and Relative Influence: An Experimental Investigation of Family Purchase Decisions,” Journal of Consumer Research , 14 1, 1– 13. Google Scholar CrossRef Search ADS   Cromwell Ronald E., Olsen David H. ( 1975), Power in Families , Thousand Oaks, CA: Sage. Decuyper Mieke, De Bolle Marleen, De Fruyt Filip ( 2012), “Personality Similarity, Perceptual Accuracy, and Relationship Satisfaction in Dating and Married Couples,” Personal Relationships , 19 1, 128– 45. Google Scholar CrossRef Search ADS   Diener Ed ( 2012), “New Findings and Future Directions for Subjective Well-Being Research,” American Psychologist , 67 8, 590. Google Scholar CrossRef Search ADS PubMed  Diener Ed, Biswas-Diener Robert ( 2002), “Will Money Increase Subjective Well-Being?” Social Indicators Research , 57 2, 119– 69. Google Scholar CrossRef Search ADS   Diener Ed, Diener Marissa ( 2009), “Cross-Cultural Correlates of Life Satisfaction and Self-Esteem,” in Culture and Well-Being , ed. Diener Ed, New York: Springer, 71– 91. Google Scholar CrossRef Search ADS   Diener Ed, Emmons Robert A., Larsen Randy J., Griffin Sharon ( 1985), “The Satisfaction with Life Scale,” Journal of Personality Assessment , 49 1, 71– 5. Google Scholar CrossRef Search ADS PubMed  Diener Ed, Ng Weiting, Harter James, Arora Raksha ( 2010), “Wealth and Happiness across the World: Material Prosperity Predicts Life Evaluation, Whereas Psychosocial Prosperity Predicts Positive Feeling,” Journal of Personality and Social Psychology , 99 1, 52. Google Scholar CrossRef Search ADS PubMed  Dyrenforth Portia S., Kashy Deborah A., Brent Donnellan M., Lucas Richard E. ( 2010), “Predicting Relationship and Life Satisfaction from Personality in Nationally Representative Samples from Three Countries: The Relative Importance of Actor, Partner, and Similarity Effects,” Journal of Personality and Social Psychology , 99 4, 690. Google Scholar CrossRef Search ADS PubMed  Dzhogleva Hristina, Lamberton Cait Poynor ( 2014), “Should Birds of a Feather Flock Together? Understanding Self-Control Decisions in Dyads,” Journal of Consumer Research , 41 2, 361– 80. Google Scholar CrossRef Search ADS   Emerson Richard M ( 1962), “Power-Dependence Relations,” American Sociological Review , 31– 41. Emmons Robert A. ( 1986), “Personal Strivings: An Approach to Personality and Subjective Well-Being,” Journal of Personality and Social Psychology , 51 5, 1058. Google Scholar CrossRef Search ADS   Emmons Robert A., King Laura A. ( 1988), “Conflict among Personal Strivings: Immediate and Long-Term Implications for Psychological and Physical Well-Being,” Journal of Personality and Social Psychology , 54 6, 1040. Google Scholar CrossRef Search ADS PubMed  Escalas Jennifer Edson, Bettman James R. ( 2003), “You Are What They Eat: The Influence of Reference Groups on Consumers’ Connections to Brands,” Journal of Consumer Psychology , 13 3, 339– 48. Google Scholar CrossRef Search ADS   Escalas Jennifer Edson, Bettman James R. ( 2005), “Self-Construal, Reference Groups, and Brand Meaning,” Journal of Consumer Research , 32 3, 378– 89. Google Scholar CrossRef Search ADS   Filiatrault Pierre, Ritchie J. R. Brent ( 1980), “Joint Purchasing Decisions: A Comparison of Influence Structure in Family and Couple Decision-Making Units,” Journal of Consumer Research , 7 2, 131– 40. Google Scholar CrossRef Search ADS   Finkel Eli J., Eastwick Paul W., Karney Benjamin R., Reis Harry T., Sprecher Susan ( 2012), “Online Dating a Critical Analysis from the Perspective of Psychological Science,” Psychological Science in the Public Interest , 13 1, 3– 66. Google Scholar CrossRef Search ADS   Fiske Susan T. ( 1993), “Controlling Other People: The Impact of Power on Stereotyping,” American Psychologist , 48 6, 621. Google Scholar CrossRef Search ADS PubMed  Fitzsimons Gavan J. ( 2000), “Consumer Response to Stockouts,” Journal of Consumer Research , 27 2, 249– 66. Google Scholar CrossRef Search ADS   Fitzsimons Gráinne M., Finkel Eli J., vanDellen Michelle R. ( 2015), “Transactive Goal Dynamics,”  Psychological Review,   122 4, 648. Google Scholar CrossRef Search ADS   Fournier Susan ( 1998), “Consumers and Their Brands: Developing Relationship Theory in Consumer Research,”  Journal of Consumer Research,   24 4, 343– 53. Google Scholar CrossRef Search ADS   French John R. P., Raven Bertram ( 1959), “The Bases of Social Power,” in Classics of Organization Theory , ed. Shafritz Jay M., Whitbeck Philip H., Oak Park, IL: Moore Publishing, 311– 20. Fredrickson Barbara L., Joiner Thomas ( 2002), “Positive Emotions Trigger Upward Spirals toward Emotional Well-Being,” Psychological Science , 13 2, 172– 5. Google Scholar CrossRef Search ADS PubMed  Galinsky Adam D., Gruenfeld Deborah H., Magee Joe C. ( 2003), “From Power to Action,” Journal of Personality and Social Psychology , 85 3, 453. Google Scholar CrossRef Search ADS PubMed  Galinsky Adam D., Magee Joe C., Inesi M. Ena, Gruenfeld Deborah H. ( 2006), “Power and Perspectives Not Taken,” Psychological Science , 17 12, 1068– 74. Google Scholar CrossRef Search ADS PubMed  Gaunt Ruth ( 2006), “Couple Similarity and Marital Satisfaction: Are Similar Spouses Happier?” Journal of Personality , 74 5, 1401– 20. Google Scholar CrossRef Search ADS PubMed  Glicksohn Joseph, Golan Hilla ( 2001), “Personality, Cognitive Style, and Assortative Mating,” Personality and Individual Differences , 30 7, 1199– 1209. Google Scholar CrossRef Search ADS   Gonzaga Gian C., Campos Belinda, Bradbury Thomas ( 2007), “Similarity, Convergence, and Relationship Satisfaction in Dating and Married Couples,” Journal of Personality and Social Psychology , 93 1, 34. Google Scholar CrossRef Search ADS PubMed  Gordon Amie M., Chen Serena ( 2016), “Do You Get Where I’m Coming From? Perceived Understanding Buffers against the Negative Impact of Conflict on Relationship Satisfaction,” Journal of Personality and Social Psychology , 110 2, 239. Google Scholar CrossRef Search ADS PubMed  Gray-Little Bernadette, Burks Nancy ( 1983), “Power and Satisfaction in Marriage: A Review and Critique,” Psychological Bulletin , 93 3, 513. Google Scholar CrossRef Search ADS   Grote Nancy K., Clark Margaret S. ( 2001), “Perceiving Unfairness in the Family: Cause or Consequence of Marital Distress?” Journal of Personality and Social Psychology , 80 2, 281– 93. Google Scholar CrossRef Search ADS PubMed  Gustavson Kristin, Røysamb Espen, Borren Ingrid, Torvik Fartein Ask, Karevold Evalill ( 2016), “Life Satisfaction in Close Relationships: Findings from a Longitudinal Study,”  Journal of Happiness Studies   17 3, 1293– 1311. Google Scholar CrossRef Search ADS   Hayes Andrew F., Scharkow Michael ( 2013), “The Relative Trustworthiness of Inferential Tests of the Indirect Effect in Statistical Mediation Analysis: Does Method Really Matter?” Psychological Science  24 10, 1918– 27. Google Scholar CrossRef Search ADS PubMed  Heaton Tim B., Pratt Edith L. ( 1990), “The Effects of Religious Homogamy on Marital Satisfaction and Stability,” Journal of Family Issues , 11 2, 191– 207. Google Scholar CrossRef Search ADS   Heller Daniel, Watson David, Ilies Remus ( 2004), “The Role of Person versus Situation in Life Satisfaction: A Critical Examination,” Psychological Bulletin , 130 4, 574. Google Scholar CrossRef Search ADS PubMed  Hoeffler Steve, Keller Kevin Lane ( 2003), “The Marketing Advantages of Strong Brands,” Journal of Brand Management  10 6, 421– 45. Google Scholar CrossRef Search ADS   Holt Douglas B. ( 2002), “Why Do Brands Cause Trouble? A Dialectical Theory of Consumer Culture and Branding,” Journal of Consumer Research , 29 1, 70– 90. Google Scholar CrossRef Search ADS   Houts Renate M., Robins Elliot, Huston Ted L. ( 1996), “Compatibility and the Development of Premarital Relationships,” Journal of Marriage and the Family , 58 1, 7– 20. Google Scholar CrossRef Search ADS   Jiang Yuwei, Zhan Lingjing, Rucker Derek D. ( 2014), “Power and Action Orientation: Power as a Catalyst for Consumer Switching Behavior,” Journal of Consumer Research , 41 1, 183– 96. Google Scholar CrossRef Search ADS   Kelley Harold H. ( 1979), Personal Relationships: Their Structures and Processes , Hillsdale, NJ: Lawrence Erlbaum Associates. Keller Kevin Lane ( 1993) “Conceptualizing, Measuring, and Managing Customer-based Brand Equity,” Journal of Marketing , 57 1, 1– 22. Google Scholar CrossRef Search ADS   Keltner Dacher, Gruenfeld Deborah H., Anderson Cameron ( 2003), “Power, Approach, and Inhibition,” Psychological Review , 110 2, 265. Google Scholar CrossRef Search ADS PubMed  Kenny David A., Kashy Deborah A., Cook William L. ( 2006), Dyadic Data Analysis , New York: Guilford. Kiecolt-Glaser Janice K., Gouin Jean-Philippe, Hantsoo Liisa ( 2010), “Close Relationships, Inflammation, and Health,” Neuroscience and Biobehavioral Reviews , 35 1, 33– 8. Google Scholar CrossRef Search ADS PubMed  Kiecolt-Glaser Janice K., Loving Timothy J., Stowell Jeffrey R., Malarkey William B., Lemeshow Stanley, Dickinson Stephanie L., Glaser Ronald ( 2005), “Hostile Marital Interactions, Proinflammatory Cytokine Production, and Wound Healing,” Archives of General Psychiatry , 62 12, 1377– 84. Google Scholar CrossRef Search ADS PubMed  Kim Sara, McGill Ann L. ( 2011), “Gaming with Mr. Slot or Gaming the Slot Machine? Power, Anthropomorphism, and Risk Perception,” Journal of Consumer Research , 38 1, 94– 107. Google Scholar CrossRef Search ADS   Kirchler Erich ( 1993), “Spouses’ Joint Purchase Decisions: Determinants of Influence Tactics for Muddling through the Process,” Journal of Economic Psychology , 14 2, 405– 38. Google Scholar CrossRef Search ADS   Kraus Michael W., Chen Serena, Keltner Dacher ( 2011), “The Power to Be Me: Power Elevates Self-Concept Consistency and Authenticity,” Journal of Experimental Social Psychology , 47 5, 974– 80. Google Scholar CrossRef Search ADS   Levinger George, Breedlove James ( 1966), “Interpersonal Attraction and Agreement: A Study of Marriage Partners,”  Journal of Personality and Social Psychology,   3 4, 367. Google Scholar CrossRef Search ADS   Levinger George, Rands Marylyn ( 1985), “Compatibility in Marriage and Other Close Relationships,” in Compatible and Incompatible Relationships , New York: Springer, 309– 31. Google Scholar CrossRef Search ADS   Lim Chaeyoon, Putnam Robert D. ( 2010), “Religion, Social Networks, and Life Satisfaction,” American Sociological Review , 75 6, 914– 33. Google Scholar CrossRef Search ADS   Little Anthony C., Perrett David I. ( 2002), “Putting Beauty Back in the Eye of the Beholder,” The Psychologist  15, 28– 32. Liu Hui, Reczek Corinne ( 2012), “Cohabitation and US Adult Mortality: An Examination by Gender and Race,” Journal of Marriage and Family , 74 4, 794– 811. Google Scholar CrossRef Search ADS   Lowe Michael L., Haws Kelly L. ( 2014), “(Im) Moral Support: The Social Outcomes of Parallel Self-Control Decisions,” Journal of Consumer Research , 41 2, 489– 505. Google Scholar CrossRef Search ADS   Lucas Richard E., Clark Andrew E., Georgellis Yannis, Diener Ed ( 2004a), “Unemployment Alters the Set Point for Life Satisfaction,” Psychological Science , 15 1, 8– 13. Google Scholar CrossRef Search ADS   Lucas Todd W., Wendorf Craig A., Imamoglu E. Olcay, Shen Jiliang, Parkhill Michele R., Weisfeld Carol C., Weisfeld Glenn E. ( 2004b), “Marital Satisfaction in Four Cultures as a Function of Homogamy, Male Dominance, and Female Attractiveness,” Sexualities, Evolution & Gender , 6 ( 2–3), 97– 130. Google Scholar CrossRef Search ADS   Luhmann Maike, Hofmann Wilhelm, Eid Michael, Lucas Richard E. ( 2012), “Subjective Well-Being and Adaptation to Life Events: A Meta-Analysis,” Journal of Personality and Social Psychology , 102 3, 592. Google Scholar CrossRef Search ADS PubMed  Luo Shanhong, Klohnen Eva C ( 2005), “Assortative Mating and Marital Quality in Newlyweds: A Couple-Centered Approach,” Journal of Personality and Social Psychology , 88 2, 304. Google Scholar CrossRef Search ADS PubMed  Luo Xueming ( 2005), “How Does Shopping with Others Influence Impulsive Purchasing?” Journal of Consumer Psychology , 15 4, 288– 94. Google Scholar CrossRef Search ADS   Magee Joe C., Galinsky Adam D. ( 2008), “8 Social Hierarchy: The Self‐Reinforcing Nature of Power and Status,” Academy of Management Annals , 2 1, 351– 98. Google Scholar CrossRef Search ADS   Magee Joe C., Galinsky Adam D., Gruenfeld Deborah H. ( 2007), “Power, Propensity to Negotiate, and Moving First in Competitive Interactions,” Personality and Social Psychology Bulletin  33 2, 200– 12. Google Scholar CrossRef Search ADS PubMed  Malhotra Naresh K. ( 1988), “Self Concept and Product Choice: An Integrated Perspective,” Journal of Economic Psychology , 9 1, 1– 28. Google Scholar CrossRef Search ADS   Mare Robert D. ( 1991), “Five Decades of Educational Assortative Mating,” American Sociological Review , 15– 32. Marmot Michael G. ( 2004), The Status Syndrome: How Social Standing Affects Our Health and Longevity , London: Macmillan. Martin Kelly D., Paul Hill Ronald ( 2012), “Life Satisfaction, Self-Determination, and Consumption Adequacy at the Bottom of the Pyramid,” Journal of Consumer Research , 38 6, 1155– 68. Google Scholar CrossRef Search ADS   Merriam-Webster.com, “Conflict,” http://www.merriam-webster.com/dictionary/conflict. Montoya R. Matthew, Horton Robert S., Kirchner Jeffrey ( 2008), “Is Actual Similarity Necessary for Attraction? A Meta-Analysis of Actual and Perceived Similarity,” Journal of Social and Personal Relationships , 25 6, 889– 922. Google Scholar CrossRef Search ADS   Mourali Mehdi, Yang Zhiyong ( 2013), “The Dual Role of Power in Resisting Social Influence,” Journal of Consumer Research , 40 3, 539– 54. Google Scholar CrossRef Search ADS   Mroczek Daniel K., Spiro AvronIII ( 2005), “Change in Life Satisfaction During Adulthood: Findings from the Veterans Affairs Normative Aging Study,” Journal of Personality and Social Psychology , 88 1, 189. Google Scholar CrossRef Search ADS PubMed  Muniz Albert M.Jr., O’Guinn Thomas C. ( 2001), “Brand Community,” Journal of Consumer Research , 27 4, 412– 32. Google Scholar CrossRef Search ADS   Murray Sandra L., Holmes John G., Bellavia Gina, Griffin Dale W., Dolderman Dan ( 2002), “Kindred Spirits? The Benefits of Egocentrism in Close Relationships,” Journal of Personality and Social Psychology , 82 4, 563. Google Scholar CrossRef Search ADS PubMed  Ng Sharon, Houston Michael J. ( 2006), “Exemplars or Beliefs? The Impact of Self-View on the Nature and Relative Influence of Brand Associations,” Journal of Consumer Research , 32 4, 519– 29. Google Scholar CrossRef Search ADS   Overbeck Jennifer R., Droutman Vitaliya ( 2013), “One for All Social Power Increases Self-Anchoring of Traits, Attitudes, and Emotions,” Psychological Science , 24 8, 1466– 76. Google Scholar CrossRef Search ADS PubMed  Park C. Whan ( 1982), “Joint Decisions in Home Purchasing: A Muddling-through Process,” Journal of Consumer Research , 9 2, 151– 62. Google Scholar CrossRef Search ADS   Park C. Whan, MacInnis Deborah J., Priester Joseph, Eisingerich Andreas B., Iacobucci Dawn ( 2010), “Brand Attachment and Brand Attitude Strength: Conceptual and Empirical Differentiation of Two Critical Brand Equity Drivers,” Journal of Marketing , 74 6, 1– 17. Google Scholar CrossRef Search ADS   Park Ji Kyung, John Deborah Roedder ( 2010), “Got to Get You into My Life: Do Brand Personalities Rub Off on Consumers?,” Journal of Consumer Research , 37 4, 655– 69. Google Scholar CrossRef Search ADS   Pavot William, Diener Ed ( 1993), “Review of the Satisfaction with Life Scale,” Psychological Assessment , 5 2, 164. Google Scholar CrossRef Search ADS   Raghunathan Rajagopal, Corfman Kim ( 2006), “Is Happiness Shared Doubled and Sadness Shared Halved? Social Influence on Enjoyment of Hedonic Experiences,” Journal of Marketing Research , 43 3, 386– 94. Google Scholar CrossRef Search ADS   Rick Scott I., Small Deborah A., Finkel Eli J. ( 2011), “Fatal (Fiscal) Attraction: Spendthrifts and Tightwads in Marriage,”  Journal of Marketing Research,   48 2, 228– 37. Google Scholar CrossRef Search ADS   Robins Richard W., Caspi Avshalom, Moffitt Terrie E. ( 2000), “Two Personalities, One Relationship: Both Partners’ Personality Traits Shape the Quality of Their Relationship,”  Journal of Personality and Social Psychology,   79 2, 251. Google Scholar CrossRef Search ADS   Rucker Derek D., Dubois David, Galinsky Adam D. ( 2011), “Generous Paupers and Stingy Princes: Power Drives Consumer Spending on Self Versus Others,” Journal of Consumer Research , 37 6, 1015– 29. Google Scholar CrossRef Search ADS   Rucker Derek D., Galinsky Adam D. ( 2008), “Desire to Acquire: Powerlessness and Compensatory Consumption,” Journal of Consumer Research , 35 2, 257– 67. Google Scholar CrossRef Search ADS   Rucker Derek D., Galinsky Adam D. ( 2016), “Growing beyond Growth: Why Multiple Mindsets Matter for Consumer Behavior,” Journal of Consumer Psychology , 26 1, 161– 4. Google Scholar CrossRef Search ADS   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. Google Scholar CrossRef Search ADS   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. Google Scholar CrossRef Search ADS   Rusbult Caryl E., Kumashiro Madoka, Kubacka Kaska E., Finkel Eli J. ( 2009), “‘ The Part of Me That You Bring Out’: Ideal Similarity and the Michelangelo Phenomenon,” Journal of Personality and Social Psychology , 96 1, 61. Google Scholar CrossRef Search ADS PubMed  Schwarz Norbert, Bless Herbert, Strack Fritz, Klumpp Gisela, Rittenauer-Schatka Helga, Simons Annette ( 1991), “Ease of Retrieval as Information: Another Look at the Availability Heuristic,” Journal of Personality and Social Psychology , 61 2, 195. Google Scholar CrossRef Search ADS   Selig James P., Preacher Kevin J. ( 2008), “Monte Carlo Method for Assessing Mediation: An Interactive Tool for Creating Confidence Intervals for Indirect Effects [Computer Software].” Shiota Michelle N., Levenson Robert W. ( 2007), “Birds of a Feather Don’t Always Fly Farthest: Similarity in Big Five Personality Predicts More Negative Marital Satisfaction Trajectories in Long-Term Marriages,” Psychology and Aging , 22 4, 666. Google Scholar CrossRef Search ADS PubMed  Simpson Jeffry A., Farrell Allison K., Oriña M. Minda, Rothman Alexander J. ( 2014), “Power and Social Influence in Relationships,” in APA Handbook of Personality and Social Psychology, Volume 3: Interpersonal Relations , ed. Simpson Jeffry A., Dovidio John F., Washington, DC: American Psychological Association, 393– 420. Simpson Jeffry A., Griskevicius Vladas, Rothman Alexander J. ( 2012), “Consumer Decisions in Relationships,” Journal of Consumer Psychology , 22 3, 304– 14. Google Scholar CrossRef Search ADS   Smith Pamela K., Bargh John A. ( 2008) “Nonconscious Effects of Power on Basic Approach and Avoidance Tendencies,” Social Cognition , 26 1, 1. Google Scholar CrossRef Search ADS PubMed  Smith Pamela K., Jostmann Nils B., Galinsky Adam D., van Dijk Wilco W. ( 2008), “Lacking Power Impairs Executive Functions,” Psychological Science , 19 5, 441– 7. Google Scholar CrossRef Search ADS PubMed  Spiller Stephen A., Fitzsimons Gavan J., Lynch John G.Jr., McClelland Gary H. ( 2013), “Spotlights, Floodlights, and the Magic Number Zero: Simple Effects Tests in Moderated Regression,” Journal of Marketing Research , 50 2, 277– 88. Google Scholar CrossRef Search ADS   Steffel Mary, Le Boeuf Robyn A. ( 2014), “Overindividuation in Gift Giving: Shopping for Multiple Recipients Leads Givers to Choose Unique but Less Preferred Gifts,” Journal of Consumer Research , 40 6, 1167– 80. Google Scholar CrossRef Search ADS   Thibaut John W., Kelley Harold H. ( 1959), The Social Psychology of Groups , New York: Wiley. Thomson Matthew, MacInnis Deborah J., Park C. Whan ( 2005), “The Ties that Bind: Measuring the Strength of Consumers’ Emotional Attachments to Brands,” Journal of Consumer Psychology , 15 1, 77– 91. Google Scholar CrossRef Search ADS   Tidwell Natasha D., Eastwick Paul W., Finkel Eli J. ( 2013), “Perceived, Not Actual, Similarity Predicts Initial Attraction in a Live Romantic Context: Evidence from the Speed‐Dating Paradigm,” Personal Relationships , 20 2, 199– 215. Google Scholar CrossRef Search ADS   Waite Linda, Gallagher Maggie ( 2002), The Case for Marriage: Why Married People Are Happier, Healthier and Better Off Financially , New York: Broadway Books. Watson David, Klohnen Eva C., Casillas Alex, Simms Ericka Nus, Haig Jeffrey, Berry Diane S. ( 2004), “Match Makers and Deal Breakers: Analyses of Assortative Mating in Newlywed Couples,” Journal of Personality , 72 5, 1029– 68. Google Scholar CrossRef Search ADS PubMed  Weisfeld Glenn E., Russell R. J. H., Weisfeld C. C., Wells Pamela A. ( 1992), “Correlates of Satisfaction in British Marriages,” Ethology and Sociobiology , 13 2, 125– 45. Google Scholar CrossRef Search ADS   Zhao Xinshu, Lynch John G.Jr., Chen Qimei ( 2010), “Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis,” Journal of Consumer Research , 37 2, 197– 206. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of Journal of Consumer Research, Inc. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

Journal of Consumer ResearchOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

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.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from Google Scholar, PubMed
Create lists to organize your research
Export lists, citations
Access to DeepDyve database
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off