1 Co-branding Strategy in Cause-related Advertising : The Fit between Brand and Cause

The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-pro t purposes provided that: • a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult the full DRO policy for further details.


INTRODUCTION
Social cause messages appear frequently in goods and services advertising to make brands stand out among an overload of advertisements in media. Firms commonly use advertising to communicate practices to stakeholders of donating a portion of sales revenue to charities (Robinson, Irmak and Jayachandran, 2012), the purpose of which is to encourage consumers to buy advertised products, and contribute to social causes (Andrews, Luo, Fang and Aspara, 2014).
To publicize the involvement of companies in social causes is important, because numerous authors associate this support with better brand attitudes, preferences, and a greater willingness to purchase products (Aguinis and Glavas, 2012;Du et al., 2010;Torres et al., 2012), but growing interest in these topics in the literature is contrasted with a decline in corporate contributions to social causes. Although the number of socially responsible companies has been growing, and Jong and Van der Meer (2015) report that nearly all contemporary companies are involved in some type of social responsibility, their contributions in this area are declining. For example, in the United States, corporate donations to charities since the mid-1980s increased from $3.67 billion to $ 18 billion between 1986 and 2012, but relative contributions fell from 2.1% in before-tax profit in 1986 to 0.8% in 2012 (Stern, 2013). Business donations represent only about 6% in the private sector, and just over 1% of the $ 1.5 trillion charity economy in 2012 (Stern, 2013).
Both businesses and consumers provide several reasons to explain this situation. First, firms that implement cause-related marketing usually follow financial and marketing goals (Fine, 1990), but in some studies in which financial performance is used as a criterion to evaluate willingness to pay, etc.), which are extrinsic reasons, by means of which companies expect to increase financial benefits. As Andreoni (1989) suggests, this is a case of impure altruism because profit derives from such actions. Third, companies react to pressures from stakeholders, and society, generally (Groza et al., 2011). A large number of studies suggest that egoism, not altruism, bases social cause support, and they address these extrinsic motivations by analysing the influence that support for CSR has on an organization's results (Arora and Henderson, 2007;Becker-Olsen et al., 2006;Robinson, Irmak, and Jayachandran, 2012). Findings indicate positive results in most cases (Graafland and Mazereeuw-Van der Duijn Schouten, 2012).
Researchers who have worked on messages' contents, from the linguistic perspective, have pointed out that what is implicit in the message is often much more persuasive than the explicit content. That is, excessively apparent and obvious intentions in messages reduce their persuasiveness, and it is gaining much support the idea that advertising acts as publicity in the sense that it builds brand equity in a long term strategy (Cook, 2002). In the same way, companies that support social causes should consider if doing so as a tactical tool to increase sales, or should have a strategic approach to improve brand image (Ross et al., 1992). Since tactical tools offer limited capacities, as Müller et al. (2014) suggest, and without considering selfishness or altruistic motivations, the best alternative is to consider supporting a cause from a strategic perspective. Research suggests that long-term-focused institutional approaches to supporting social causes induces more favourable attitude and loyalty toward the firm from consumers, and decreases consumers' scepticism, in comparison to short-term approaches (Kim et al., 2012).
From a strategic viewpoint, one major issue with which practitioners must deal is finding causes that benefit the firm (Kim et al., 2012). This problem is common in literature on brand alliances and extensions, and it suggests the importance of perceived fit-two brands that can be marketed together (Aaker, 1996). Brands have good fit if they generate and transmit synergies to consumers, effects that can also be generated in brand-cause partnerships (Lichtenstein et al., 2004). A brand-cause alliance stimulates image transfer from social and ecological purposes of a cause to a brand (Moosmayer and Fuljhn, 2013), and therefore advertising an alliance is essential to increasing public awareness of a cause and brand, willingness to buy products, and improvements to brand image (Aguinis and Glavas, 2012;Du et al., 2010;Smith and Langford, 2009;Torres et al., 2012).
This study focuses on consumer perceptions when announcing agreements between brands and causes, and explores how strategic fit operates, including its influence on consumers (Al-Dmour et al., 2016). From a managerial viewpoint, it is important to determine whether strategic adjustment represents an opportunity, and if so, how its effects can be enhanced.
Examining methods of persuading consumers to pay more attention to brand-cause joint advertising is more important than ever due to the information saturation in which contemporary consumers live. This study contributes to the literature in two ways. First, it demonstrates that a spontaneous alliance, created artificially among a sample of consumers in a laboratory, is able to transmit the existence of strategic fit elements to receivers. Second, the synergies that a brandcause alliance produces and that advertising transmits are emotional in nature, contributing to improving a message's persuasiveness. Qualitative and quantitative designs are used to discover strategic fit in two factors that Zdravkovic et al. (2010) propose. This study also develops a simple procedure to build strategic brand-cause alliances that offer strategic fit.

CONCEPTUAL BACKGROUND
Cause-related marketing is the practice of donating a portion of revenue from product sales to a charitable cause (Varadarajan and Menon, 1988;He et al., 2016). Therefore publicize this practice is a type of promotional campaign during which instead of offering product discounts or other commercial promotions to consumers, a company agrees to donate a portion of sales to a charitable cause (Winterich and Barone, 2011). Although, as it was pointed out above, this instrument has a limited capacity, however a company declaring commitment to devoting part of its income to support a charitable cause arouses feelings of appreciation in consumers, who consequently are more willing to reward the company with subsequent purchases (Gneezy and List, 2013), a behaviour that is more likely if consumers know that the company supports a social cause.
Incorporating messages of support a social cause aim to transfer the positive feelings generated by this support to the brand image (Grohmann and Bodur, 2015;Rivera et al 2016).
But if most consumers do not know what causes are supported by the products they buy (Bhattacharya and Sen, 2004;Fatma and Rahman, 2015;Hartmann et al., 2013) it will be problematic to generate such affective transfer. Therefore, it is difficult to build co-branding image unless the long-term, brand-cause commitment is advertised (Eckert et al., 2012).
Organizations announce participation in CSR through broadly diffused, quality communication to consumers, obtaining a reward for good behaviour, and this is clearly an emerging and relevant research field in the CSR domain. Some studies on cause-related marketing focus on advertising, and therefore have already laid a foundation for the current study on the influence of advertising on affect (Arvidsson, 2010;Chaudhri, 2014;Du et al., 2010;Skard and Thorbjornsen, 2014;Van Rekom et al., 2014). However, although consumers know about a firm's support of causes, they often regard it with skepticism and distrust because they believe that such support reflects a cynical stance of brands (Cronin et al., 2011;Fassin and Buelens, 2011;Luo and Bhattacharya, 2006;Skarmeas et al., 2014;Wagner et al., 2009).
Contemporary organizations operate in an environment in which some companies resort to greenwashing, while other approaches that improve image are becoming more frequent (Chen et al., 2014;Nyilasy et al., 2014). This might at least lead consumers and other stakeholders to adopt a vigilant attitude when they receive information about companies' CSR (De Jong and der Meer, 2015). Taking social actions to improve the environment, health, or safety is insufficient; it is essential for a target audience to be aware of such actions, through messages on labels and advertising, or by transmitting a message that such actions are part of company policy and not a marketing ploy.
The use of a tactical approach in cause-marketing communication, that is considering the support a social cause as adding an attribute to a product (e.g., donating a portion of profit to a social cause), contributes to consumer scepticism (Barone et al., 2007;Lafferty et al., 2004). It also makes them suspicious of the real motives of social cause support, thereby undermining emotional inferences between a cause and brand during evaluation (Fein et al., 1990;Wagner et al., 2009). It has been indicated that for joint brand-cause advertising contribute to improve brand image (Ross et al., 1992) and brand equity must act as publicity, i.e., strategically and long-term (Cook, 2002). Strategic fit and joint advertisements are concepts from brand extension and alliance literature, which suggests that synergy and transferability of intangible elements exist such as image and values between partners (Aaker, 1996). A cause-brand agreement is similar to brand alliance, which contributes to greater confidence, strengthening of brand notoriety, and joint credibility (Hoeffler and Keller, 2002;Lafferty et al., 2004;Polonsky and Speed, 2001), and raises consumer awareness of brands (Barone et al., 2000).
An essential variable during communication is the fit between an organization and its CSR (De Jong and der Meer, 2015). Some studies suggest that CSR should have high fit because stakeholders are more likely to appreciate support for causes that form a natural part of an organization's core business (Bigné-Alcañiz et al., 2010;Gupta and Pirsch, 2006). However, lower fit means that the combination of an organization and its CSR occurs randomly (De Jong and der Meer, 2015). Some studies even support low to moderate fit (Kim, 2011;Simmons and Becker-Olsen, 2006), but others find no effect (Chernew and Blair, 2015). Although extant literature on fit rarely distinguishes a company and its brands (Lim et al. 2012), our focus is on brands and one of the most common ways to practice CSR-cause-related marketing.
Brand-cause fit generated much debate in the literature during the last decade, defined as an overall assessment of the similarity between characteristics of both concepts (Du et al., 2010).
However, the nature and type of such fit, and how to operationalize it, has undergone some change. In the beginning, it was a general concept, and the usual way of measuring it was degree of fit such as a differential semantic scale ranging from "very good fit" to "very bad fit" (Hamlin and Wilson, 2004;Simmons and Becker-Olsen, 2006). Later, it expanded to a multidimensional scale, in which organizational features were considered during measurement (Kuo and Rice, 2015). Lafferty et al. (2004) consider fit of brand name and product category, and carried out an adjustment using a three-item scale (e.g., consistent/not consistent, complementary/not complementary, and makes sense/does not make sense). The proliferation of instruments for measuring the same phenomenon meant it was necessary to understand its nature. Bigné-Alcañiz et al. (2010) note a lack of specification regarding whether cause-brand fit is a matter of degree (i.e., magnitude) or type (i.e., category), making it difficult for the literature to reach consensus. Yuan et al. (2011) consider three areas of matching internal consistency when fit occurred at the organizational level, external consistency when goods or services met stakeholder demand, and coherence when the activities of both organizations were compatible. Another proposal was offered by Bigné et al. (2012), who distinguish functional fit related to type of good or service, and image fit related to brand and cause. Becker-Olsen and Simmons (2006) distinguish natural and artificial fit, the latter of which is created through communication campaigns. Kuo and Rice (2015) distinguish conceptual and perceptual fit. Conceptual fit refers to image and positioning coherence between brand and cause, and perceptual fit relates to similar appearance and the same colour. Another method was to distinguish strategic and tactical adjustments (i.e., duration of agreement). Zdravkovic et al. (2010) propose a strategic concept of fit, suggesting five strategic types of adjustment-slogan, mission, target, promotion, and geographic type. The current study measures strategic fit that generates brand-cause alliances constructed in a laboratory, considering the five dimensions of strategic fit that Zdravkovic et al. (2010) propose.

RESEARCH HYPOTHESES
Many consumers are unaware of which causes are supported by branded goods that they usually buy (Bhattacharya and Sen, 2004), and the only way to make consumers aware of this is using communication tools. Lafferty and Edmondson (2009) report that print advertising is a common method of communicating social and environmental actions, though social media are increasingly important to spreading messages of participation in social actions (Ashley and Tuten, 2015). However, companies continue to use magazines (Luo and Bhattacharya, 2006), which require greater subject involvement and effort to process a message, in comparison to other media such as television (Heath and Feldwick, 2008). Some precepts of strategic alliances are assumed where strategic fit relates to knowledge or skill transfer, and synergies generated when developing joint activities. In the case of brand image, the concept is perceived fit, and similar to strategic alliances. A transfer of intangible associations between brand and cause is to be expected that are capable of achieving market value, enhancing brand image, and creating greater willingness to purchase (Aguinis and Glavas, 2012;Du et al., 2010;Smith and Langford, 2009;Torres et al., 2012). In advertising, fit associates with consumer perceptions of similarity, relatedness, relevance, and congruence in a message .
The meanings transfer model that McCracken (1989) proposes suggests that meaning associated with an object, as, for example, a celebrity, can be transferred to another object such as a brand, and explains fit effects on consumer attributions. Thus, a shared, positive association published in an advertisement can be generated from a relationship between two objects. The model also suggests that strong fit offers positive evaluations. Another theory that explains the persuasiveness of advertisements is based on the valence affective hypothesis, which distinguishes arousal that generates positive and negative feelings (Schwarz, 1997). If an advertising campaign announces that a brand is supporting a social cause, the message generates positive feelings in consumers. Consumers are attracted to the opportunity to contribute to the improvement of society, and it provides them with a feeling of self-satisfaction, generating emotional wellbeing (Andrews, Luo, Fang and Aspara, 2014). The "warm-glow giving" concept was proposed by Andreoni (1989), who explains that when people give to charities, they do so prompted by impure altruism because donating makes them feel that they are doing something useful. This feeling of usefulness comes in the form of a "warm glow", a positive emotional feeling experienced by people when helping others, as, for example, from charity-related purchases (Strahilevitz and Myers, 1998).

Brands and social causes generate disparate emotional responses and attitudes in
consumers. For products, familiarity is more important than both brand awareness and advertising knowledge regarding brand attitudes (Ehrenberg, 2000), while in a social cause case, consumers are much more familiar with names, rather than through any personal experience.
Nevertheless, consumers might develop positive attitudes toward a cause based on feelings generated simply by hearing their names (Lafferty et al., 2004). According to Pham and Avnet (2004), people evaluate advertising and issue a verdict from two types of inputs: information related to the essence of the object (i.e., message strength) and affective responses (i.e., feelings generated while viewing an advertisement). Regarding social initiatives, with the belief that a sponsoring corporation is socially responsible (Ross et al., 1992), consumers show positive attitudes, or affinity, toward such initiatives (Webb & Mohr 1998), and thus demonstrate greater willingness to purchase sponsoring products (Gneezy and List, 2013;Smith and Alcorn, 1991).
However, the theory of affective valence suggests that if an advertisement arouses positive feelings, such as a "warm glow," consumers use mental shortcuts and process information simply (Batra and Stayman, 1990). Arora and Henderson (2007) suggest that perceptions and assessments of cause-related marketing campaigns differ among individuals, and choosing a cause to which people show affinity improves its effectiveness. Drumwright (1996), who argues that the success of a communication campaign depends on people's affinity to a cause, supports this argument. Affinity toward a social cause influences affective response (Barone et al., 2007), and therefore: H1: If consumers express positive affinity toward a social cause, a positive increase in affective response to an advertisement develops.
If, as Ehrenberg (2000) argues, the role of advertising is simply to enhance brand attitudes, it is a prerequisite for determining prior attitudes not only toward the brand, but also the cause. However, since the relationship between brand and advertisement attitudes has been demonstrated empirically (Mitchell and Olson 1981;MacKenzie, Lutz and Belch, 1986), and in more recent advertising research (Halkias and Kokkinaki, 2014;Scheinin, Varki, and Ashley, 2011), the relationship in our model is used, but do not present it as a hypothesis. Regarding the synergistic effect of brand-cause strategic alliances, when consumers perceive both as an overt connection, the meaning transfer model proposes a transmission of emotional resources from cause to brand (Hoeffler and Keller, 2002;Ellen et al., 2006;Bigné-Alcañiz et al., 2009;Samu and Wymer, 2009); favorable attitudes toward a sponsored cause lead to favourable attitudes toward a sponsoring brand, triggered by creation of new emotional associations with the brand (Simmons and Becker-Olsen, 2006). This is consistent with research that demonstrates that consumer attitudes toward a brand relate closely with attitudes toward brand alliances (Simonin and Ruth, 1998) and brand extensions (Sullivan, 1990), and working together, they achieve superior results in comparison to when each operates alone. Concerning brand alliances, Park et al. (1996) found that partners achieve greater success when complementary attributes fit. The degree of brand-cause fit affects the credibility of advertising campaigns and consumers' emotional perceptions (Buil et al., 2012). When a firm incorporates information about socialcause sponsoring in advertisements, and consumers perceive that this support has a high degree of fit, such coherence reinforces the company's image (Menon and Kanh, 2003). Co-branding strategic fit might contribute to greater confidence, thereby strengthening brand notoriety and joint credibility (Hoeffler and Keller, 2002;Lafferty et al., 2004;Polonsky and Speed, 2001).
Since this study constructs an artificial, strategic, brand-cause alliance, exploratory analyses are used to determine which strategic fit variables from Zdravkovic et al. (2010)  Two more relationships are added: emotional responses and strength of a message that influence purchase intentions. Neither relationship is proposed as a hypothesis, because they have both been demonstrated broadly in the literature (Pham and Avnet, 2004). Two hypotheses and a relationship summary are shown in the model in Figure 1, which suggests that an advertisement's evaluations of both affective and message strength influence purchase intentions. The affective dimension is influenced by a consumer's cause affinity, attitudes toward a brand, and strategic fit between a brand and cause for any of the five dimensions. ************************ Place Figure 1 about here ************************

14
To examine how strategic fit between brand and cause operates in consumers' minds, a twostage study is used: a qualitative part, in which brand-cause relationships are built artificially, and a quantitative part, in which one brand-cause relationship is chosen and evaluate its adjustment components to determine how fit influences purchase intentions. Literature that analyses fit does not often distinguish a company and brand, or non-profit organizations and causes (Lim et al. 2012). In this study, such demarcations matter because a brand represents a unique service, and a non-profit foundation supports only a specific cause. When people perceive an event, multiple representations of the same event are encoded in their memories, and if a researcher wants to assess the effectiveness with which facts were stored in memory, the most common tests include free recall, memory with stimuli, and recognition (Krishnan and Chakravarti, 1999). During the qualitative stage, memory without a stimuli test was used to build a brand-cause partnership. When decisions are based on information stored in memory, selected brands are part of the choice set (Hauser and Wernerfelt, 1990). According to Lee (2002), purchase decisions and brand choices are based on stimuli, information available in the physical environment, or alternatively memory (i.e., information retrieved from memory). Free recall with a sample of eighty-four undergraduates from a large university in Barcelona is used, who, using an open-ended questionnaire, recalled service brands they used routinely and social causes they knew about or with which they collaborated. With this type of test, the frequency with which a brand is repeated is an indicator that the brand belongs to the choice set in a memorybased process decision (Nedungadi 1990). It is also an effective measure if the objective of a study is to assess degree of awareness, positive effects, or purchase intentions (Stewart et al., 1985). The purpose was to choose brands and causes that were within the choice sets, and that were familiar to the target. It also reduced effects that generate varying degrees of familiarity.
Brand-cause familiarity involves prior knowledge resulting from direct or indirect experiences with brands (Alba and Hutchinson, 1987), and greater ability to assess brand attributes than when brands are unknown (Hoek et al., 2000;Krishnan and Chakravarti, 1999). Some researchers recommend using familiar, non-profit causes from which to choose a partner (Pringle and Thompson, 1999), but if consumers are unfamiliar with a non-profit brand, nothing can be transferred to the commercial brand.
Two researchers analyzed responses, and according to Lee (2002), they discarded the three most commonly cited brands, and those cited by less than 15% of participants, to avoid ceiling and floor effects. Floor and ceiling effects refer to the notion that when causal analysis of data of an independent variable reaches extreme positions in the range of variance (i.e., when data cannot assume higher or lower values), it has no effect on a dependent variable (Everitt and Skrondal, 2010). This selection criterion ensured that subsequent brand-cause links were not restricted by strong brand preferences. If a strong brand is highly accessible by memory, new stimuli do not improve accessibility (Negundagi, 1990). Results finalized four brand services (i.e., VIENA Restaurants, Gyms DIR, VUELING Airline, and ZARA stores) and four causes A second exercise consisted of presenting respondents with two columns-one containing the set of brands, and the other the list of causes. Sixty-three undergraduate students who expressed familiarity with the four brands and four causes linked each brand and cause with an arrow if they perceived fit between them so they could be advertised together. In accordance with Simmons and Becker-Olsen (2006), who distinguish natural and artificially created fit through communication, the survey assessed natural fit from consumers' perceptions. It was necessary to establish the links without incorporating attributes or slogans from brands and causes, and therefore the choice had to be made only with information retrieved from memory (Alba et al., 1992). Establishing new associations of concepts through development of integrated processing of perceptions from advertisements is common in advertising studies to detect unconscious traces from purchasing behaviours (Krishnan and Chakravarti, 1999). Although the students were encouraged to match all four brands and causes, some stated that there were causes with which they did not perceive a link, so some were left unconnected, demonstrating that in some circumstances, fit is not an easy issue for consumers. Generally, causes are much more difficult to classify than products, and this is especially true for humanitarian causes such as those used during this study (Lafferty et al., 2004). However, as Krishnan and Chakravarti (1999) argue, a combination of direct and indirect memory tests provides a much more complete picture of the effects of advertisements than recognition and recall tests.
Analysis of results suggest that one of the most popular brands among the target audience (i.e., the chain of gyms DIR) and one of the most popular social causes in Barcelona (i.e., Josep Carreras Foundation against Leukaemia) were chosen. This pair was selected among those cited most frequently because the cause is led by Josep Carreras, an opera singer who sang at the Olympic Games ceremony in Barcelona 1992. He is well-known in the city, so the pair provides a good context for the study (Hoeffler and Keller, 2002). Moreover, social message effects are more pronounced when they occur in a local area versus those used in national advertisements (Ross et al., 1992). However, as pointed out above, both the brand, DIR, and cause, Josep Carreras Foundation, only perform an activity. In the latter case, the fight against leukemia and the celebrity only support the foundation.

During the quantitative stage, the relationship between DIR and Josep Carreras
Foundation was evaluated to determine fit components, and analyze their influences on purchase intentions. Consumer attitudes toward the sponsored cause and brand were examined, and both of their influences on affective response. According to Ehrenberg (2000), the role of advertising is simply to improve attitudes toward an advertised brand, but it does not change attitudes. It is therefore necessary to analyze its effects to determine brand attitudes prior to advertising.
Participants in the second survey were undergraduates from the two largest universities in Barcelona (Spain). Subjects completed a structured questionnaire after viewing three print advertisements (Appendix 1). An advertisement with the DIR brand logo and slogan "DIR: el moviment Barcelona" (DIR: The Barcelona movement) was displayed on a screen, and participants completed a questionnaire with items related to the brand and their brand attitudes.
After completing the first part of the questionnaire, respondents were exposed to another advertisement containing a portrait of Josep Carreras and the slogan "Fundació Josep Carreras contra la leucemia" (Josep Carreras Foundation: Against Leukemia), and again completed a questionnaire with items related to the cause. Before moving on to the third advertisement, and as Lee (2002) recommends, participants were given a distracting task during which they answered questions that had no connection to the topic under investigation, which took approximately fifteen minutes. Respondents then viewed a picture of an interview with Josep Carreras, and under the photograph were the two brands, DIR and Josep Carreras Foundation, with their slogans. Respondents then completed another questionnaire regarding the fit between brand and cause. Other questions covered purchase intentions, advertising strength, affective responses, and classification data. Participants' knowledge concerning the brand and cause were measured before they completed the questionnaires. Use of print advertisements, and particularly magazine advertisements, was justified by broad use of this medium to broadcast information about companies' social activities (Luo and Bhattacharya, 2006;Lafferty and Edmondson, 2009). The context of this study is particularly relevant to college students because they are regular users of gyms and sports services, and are concerned about the actions of CSR (Wagner et al., 2009). This was also justified given the theoretical focus of the study (Calder et al., 1981).
248 questionnaires were collected from participants, but excluded incomplete questionnaires and those from respondents who did not know the brand or social cause. The final sample included 229 participants. Respondents were selected randomly and participated voluntarily. The sample consisted of 109 women (47.9%) and 120 males (52.1%), aged 19 to 40 years (mean=22.5, SD=2.5). Nearly all subjects were Spanish (96.1%). Although all stated that they were familiar with both the brand and cause, 21.7% reported they were users of DIR gyms, in comparison to 3.14% who collaborated with the social cause.
To measure the model, scales were used proposed by several authors in marketing literature. English scales were employed as a base, and translated and adapted them into Spanish. The translation and content validity of the items were assessed by ten bilingual Master's students, who proposed improvements regarding whether items were representative of the underlying constructs. Then a pilot test with twelve doctoral students was conducted to refine the questionnaire (Appendix 2). The primary variable in the model was purchase intentions, a measurement of intentions to buy a product. A three-item scale from Putrevu and Lord (1994) was used. For attitudes toward the brand, we used a three-item, reduced scale from Lafferty and Goldsmith (2005) and Simmons and Becker-Olsen (2006). Perceived affinity toward a social cause was measured using a three-item scale from Grau and Folse (2007). For brand-cause fit, the model of five strategic scales proposed by Zdravkovic et al (2010) followed, which include affinity of slogan, mission, target market, promotion, and geographic compatibility. All scales used a seven-point, Likert-type gradation.
When individuals evaluate an advertisement, they consider two types of inputs: information related to the essence of an object and information that generates affective responses. In this study, the essential input was affective responses because they represent the type expected to transmit fit between brand and cause. However, the two types correlate strongly (Eagly and Chaiken, 1995;Pham and Avnet, 2004), and it is therefore prudent to consider both during analysis. Two three-item scales for these inputs were used, both proposed by Pham and Avnet (2004). However, in the second scale, the third item, which used a reverse-rating scale, did not achieve a sufficient correlation, and so was removed. To test the hypotheses, a structural equation model using maximum-likelihood estimation was specified. The analysis was divided into two parts. First the psychometric properties of the scales was examined using exploratory and confirmatory analyses, and then tested the hypotheses with causal model validation.

Reliability and validity
Content validity was established through a literature review and using the qualitative portion of the study. Based on these procedures, the measures met conditions of content validity.
Discriminant, convergent, and scale reliabilities were assessed through confirmatory factor analysis (CFA), according with procedures recommended by Gerbing and Anderson (1993) (Tables 1 and 2). The chi-square for the model was 267.233 (p<0.001), with 188 degrees of freedom. Four other measures of fit were examined: comparative fit index (CFI=0.972), Tucker-Lewis fit index (TLI=0.966), incremental fit index (IFI=0.973), and the root mean square error of approximation (RMSEA=0.043). Results suggest that the measures were internally consistent, discriminated the constructs well, and provided good fit between model and data. Inspection of these results suggest that the items measuring the constructs were both valid (i.e., convergent and discriminant validity) and reliable (i.e., composite reliability, variance extracted, and internal reliability). Convergent validity was evidenced by large, standardized loadings (t>1.96, p<0.05) for items measuring respective constructs. Discriminant validity was assessed by observing construct inter-correlations. All were different from 1, and shared variance between any two constructs (i.e., the square of their correlation) was less than the average variance extracted (AVE) by each item for its respective construct (Fornell and Larcker, 1981). The correlation matrix for the constructs is shown in Table 2. Adequate discriminant validity was evident for all constructs since diagonal elements were greater than off-diagonal elements in corresponding rows and columns in the upper triangle. Regarding construct reliability, Table 1 presents the results of composite reliability, variance extracted, and internal consistency (i.e., Cronbach's alpha). Values for composite reliability exceeded the cut-off of 0.60 that Bagozzi and Yi (1988) recommend. The minimum composite reliability was calculated for the constructaffinity toward the social cause (0.78), and the maximum (0.93) for both geographical fit and attitudes toward the brand. In terms of variance extracted, all constructs were above the 0.50 guideline. All Cronbach's alpha coefficients were above the cut-off point of 0.7 recommended in the literature (Hair et al., 2010). Therefore, we conclude that for all constructs, the indicators were sufficient in terms of how the measurement model was specified.

Testing of hypotheses
Due to the complexity of the model and the need to test relationships among constructs simultaneously, structural equations were employed by applying maximum-likelihood method.

Theoretical implications
This paper shows how consumers' affinity with the social cause is an important precedent in consumers' affective response to a brand-cause message and, moreover, if brand-cause is perceived that fit strategically in some of the strategic fit dimensions proposed by Zdravkovic et al. (2010). This also helps to strengthen the affective response. This first result is in line with those of Sheikh and Beise-Zee (2011) who investigate how consumers respond to campaigns cause-related marketing, depending on their higher or lower affinity to the cause. Moreover, this study contributes to the literature in several ways. First, it addresses an issue of considerable practical importance-the perception of cause-brand strategic fit and how this type of fit creates synergies in advertising. Second, if consumers believe that the purpose of a brand-cause agreement is not tactical but strategic, the decision is the most suitable option for overcoming consumer reluctance to this type of action (Wagner et al., 2009), and benefits the transfer of the image effects from cause to brand, as the transfer model suggests. This strategic vision accords with studies from Barone et al. (2000), Ellen et al. (2006), and Lafferty et al. (2004), some of which use concepts from brand management literature, while others suggest that these partnerships generate synergies (Aaker, 1996). However, for these alliances to generate synergies, consumers must perceive fit between brand and cause.
A review of literature on fit suggests that this concept remains confusing. Various definitions and instruments have appeared since no consensus exists regarding whether fit is a matter of degree or type (Bigné-Alcañiz et al., 2010). There is no agreement about whether it relates to internal consistency at the organizational level, external consistency at the product level, or coherence relative to the compatibility of both organizations (Yuan et al., 2011), or whether it associates with conceptual or perceptual fit (Kuo and Rice, 2015). This study suggests that fit is a matter of the duration of an agreement. It can be tactical or strategic, and it is offered as a measure of the five dimensions that Zdravkovic et al. (2010) propose. This is the first time that these dimensions have been used in an empirical study, and therefore it contributes to validation.
This study also demonstrates that it is possible to build a laboratory, brand-cause alliance, and that the alliance reflects the fit of two of the five strategic dimensions. This was sufficient to show that the synergistic effect generated by brand-cause alliance operates because they have strategic fit. Findings suggest that in the case of brand, DIR, and cause, Josep Carreras foundation, one of the least significant effects on formation of affective response to the advertisement was brand attitudes, in comparison to cause affinity. Strategic fit produced synergy from the slogan, and a little less was produced by geographic fit. In accordance with the theory of affective valence, the advertisement led to enhancement of positive feelings such as a "warm glow" (Andrews et al., 2014;Strahilevitz and Myers, 1998). Therefore, the brand derived a benefit from its alliance with the cause because following the meaning transfer model, emotional resources transferred to the brand (Bigné-Alcañiz et al., 2009;Hoeffler and Keller, 2002;Samu and Wymer, 2009), including a synergistic effect generated by brand-cause strategic fit (Aaker, 1996, Park et al., 1996. An exploratory analysis was conducted to determine which type of fit was present in the alliance since it is difficult to predict the type of fit that consumers perceive. For example, results of this study demonstrate fit between brand slogan, "Barcelona DIR movement," and the cause slogan, "Josep Carreras Foundation against Leukaemia," which would have been difficult to hypothesize. However, as Kuo and Rice (2015) point out, respondents might have considered perceptual fit between the two, and they focused on appearance (i.e., they both use blue). Had we hypothesised about the other significant fit-geographical-it would have been more plausible to fit the answer given the extensive knowledge that people in Barcelona have about both brands.
Although most studies support strong brand-cause fit, some favour low to moderate adjustment (Kim, 2011) and others find no effect (Chernew and Blair, 2015). Therefore, it cannot be concluded that adjustment to only two of the five strategic dimensions represents poor fit, and gaps remain concerning fit categories. More research is required to overcome them.
Another critical feature of this study is the complex nature of the proposed service and its effect on purchase intentions. Despite a stimulus increase in the emotional node that generated the joint brand-cause advertising, the largest boost to purchase intentions came from message strength rather than information related to affective response. Perhaps the type of service used during the experiment determined that result. The literature suggests that when a sponsoring brand involves a hedonic product (e.g., ice cream, concert tickets, etc.), this affects purchase intentions more than when the sponsoring brand is a utilitarian product (e.g., laundry detergent, toothpaste, etc.) (Chang, 2008;Strahilevitz and Myers, 1998). From an advertisement assessment perspective, Pham (1998) argues that when a product advertised is a final product, affective criteria are used more than when a product is instrumental. The DIR gym chain, which is popular in Barcelona, is difficult to classify between hedonic and utilitarian, or between final and instrumental. Communications from the company emphasise healthy elements such as exercise, with entertainment and social components to them. According to what the valence affective model suggests, when an advertisement claims that a brand supports a social cause, the information generates positive feelings among customers, which in turn creates a state of wellbeing when purchasing the product (Andrews et al., 2014). However, according to Andreoni 26 (1989), these feelings respond to impure altruism since the benefactors derive some sense of usefulness by supporting social causes. Types of stimuli that transmit happiness cause people to focus on affective information (Schwarz and Clore, 1996).

Managerial implications
From the obtained findings, it is possible to draw some practical and interesting implications for managers in charge of advertising campaigns that support social causes. As companies have to convince consumers that their support for a social cause is sincere, consumers should define which social cause best fits the brand, since this could be a quick and easy way to get potential candidates to establish an 'unsuspicious' brand-cause alliance.
In addition, as pointed out above, an excessive role of the brand in promoting support for social causes may arouse suspicion, thereby creating the impression in consumers that the motivation behind funding social causes is to increase profits, thus giving rise to a negative assessment. Therefore, a discrete position of the brand, leaving the main role of communication to the cause, could increase the brand image much more than excessive prominence. On the other hand, this study also complements extant research on brand-cause fit, and expands knowledge regarding the strategic nature of fit relevant to designing a cause-brand, co-branding strategy, or choosing a sponsorship event. CRM has become general practice among firms, and managers must therefore consider establishment of robust, co-branding strategies capable of achieving goals. Thus, exploring the type of adjustment that provides brand-cause partnerships assists with improving the choice. Managers should determine which type of fit each alliance provides (i.e., if it is in the slogan, mission, target, promotion form, or geographical features). This knowledge guides the design of communication campaigns, improves consumer perceptions of fit, and increases purchase intentions.

Limitations and extensions
The low significance of brand attitudes in comparison to cause affinity when creating affective responses highlights the complexity of brand-cause relationships that needs to be considered in more complex models. Barone et al. (2007) suggest that complexity is due to decision contexts.
In this study, and following Lee's (2002) recommendations, the three most cited brands, and all cited by less than 15% of participants, were discarded to avoid ceiling and floor effects.
However, the asymmetric effects that were detected pose the following question: In a hypothetical situation in which there are two brands, one stronger than the other, and two causes, one with a higher affinity than the other, what type of alliance generates greater synergy? An extension of this study should examine what combinations (strong-strong, weak-strong, etc.) generate greater synergy. Another limitation resides in the use of a sample of students. Studies using student samples often lack external validity, limiting extrapolation to a population (McGrath and Brinberg, 1983). Since the purpose of this study was to assist with validating the construct that Zdravkovic et al. (2010) propose, while also demonstrating how synergistic effects operate, greater emphasis on internal validity was placed, controlling advertisement exposure and monitoring a homogeneous sample. Another limitation, due to economic restrictions, was that only one brand and cause were considered; examining more brands and causes would enrich Items Source Purchase intention (PI) (PI 1) Next time I need to buy a product/service with these characteristics, I will buy brand XXX.
(PI 2) It is likely that in the future I will purchase a product/service of brand XXX.
(PI 3) I will definitely buy a product/service of brand XXX. Putrevu and Lord (1994) Social cause affinity (SCA) (SCA 1) Activities in which this organization is working are important for me.
(SCA 2) I find the work done by this organization interesting.
(SCA 3) I like the initiatives/projects developed by this organization. Grau and Folse (2007) Brand attitude (BA) My attitude toward brand XXX is: