Cause-Related Marketing Inﬂuence on Consumer Loyalty in a Medium-Sized City

: The purpose of this study is to analyze the e ﬀ ect of the performance of a cause-related marketing action on consumer loyalty by a company. In addition, the study explores the moderating e ﬀ ect of the publicizing medium. The proposed theoretical model was tested based on data gathered from a face-to-face questionnaire completed by 421 respondents living in a medium-sized city. The results validated the proposed model and showed that the functional and image ﬁt between social actions and companies are key antecedents of perceived corporate ability (CA) and company credibility. It was shown that CA directly inﬂuences customer satisfaction, that credibility indirectly inﬂuences customer satisfaction through perceived corporate social responsibility, and that satisfaction directly and positively impacts customer loyalty. Moreover, the inﬂuence of functional and image ﬁt in the model were shown to be moderated by the type of publicizing medium. Speciﬁcally, the e ﬀ ect of functional ﬁt on corporate ability is greater for traditional media (TM) than for social media (SM). On the other hand, the e ﬀ ect of image ﬁt on corporate ability is greater for SM than for TM. The theoretical and practical implications of these results are discussed.


Introduction
Social studies conducted over the last decades have shown that consumers are becoming more socially and environmentally aware in their purchasing decisions. Companies have observed how the consumer's concern for social issues influences their purchasing decisions. In parallel, small-and medium-sized cities have adopted a management model based on innovation, diversification, and environmental sustainability [1]. The citizens of small-and medium-sized cities demand that the companies operating in their territories do so with consistent sustainable development that minimizes the negative impacts on society [2]. In addition, from the end of the first decade of this century, many cities have taken advantage of technological advances to achieve sustainable economic development and improve quality of life, in line with the recently coined concept of the "smart city". In this context, companies have shown a growing interest and commitment to implementing corporate social responsibility (CSR) [3][4][5]. CSR has been defined as the management of stakeholder concerns about ethical, social, and environmental issues to generate corporate benefits [6]. That is, the issue is not solely ethical or ideological, but also economic, because "doing good" leads to "doing better" through the positive effect on stakeholders [7]. More than 8000 companies from more than 150 countries have signed the United Nations Global Compact on human, workers, and environmental rights (and some

Conceptual Framework
The marketing literature has established that the knowledge a consumer has about a company or brand influences his/her behavior toward it [21,22]. An up-to-date view of the classical conditioning paradigm would regard this as cognitive associative learning, which suggests that individuals can learn from the relationships between environmental events [23]. According to associative learning, consumers learn consciously or unconsciously from the associations between stimuli, which influence what (s)he thinks or feels. Previous studies have used the classical conditioning paradigm as an appropriate theoretical framework to explain the positive transfer of social cause values to the socially responsible positioning of companies [13,17,24]. According to this paradigm, in cause-related marketing, there is an association between two stimuli: the cause and the brand/company [25]. For the association to take place in the mind of the consumer, companies have to make the action taken known. In this sense, researchers have identified that the main antecedent of the persuasive capacity of communication through the associative learning process is the fit between the social cause and the brand/company [17].
Perceived fit has been defined as the degree of similarity and compatibility between a company and its social actions [26]. Bigné et al. [17] differentiated between two types of cause-company fit: functional and image fit. Functional fit is the relationship between the functional characteristics of the main product of a company and the characteristics and intentions of the social cause. Image fit, on the other hand, is the similarity between the image and positioning of the social cause and the company. For example, the donation of vaccines by a pharmaceutical company would be perceived as having high functional fit, while image fit would compare consumer perceptions of Médecins Sans Frontières with the Sanofi Pasteur laboratory. Moreover, the literature has suggested that, in cause-related marketing messages, the credibility of the brand or company is one of the most powerful aspects that prevents consumers from being suspicious about the motivations of the brand or company [27]. In the context of cause-related marketing, credibility has been defined as "the extent to which a consumer perceives that the brand expresses sincerity and goodwill (trustworthiness) and has the skill and experience necessary (expertise) to associate to the specified social cause" [27].
According to Brown and Dacin [11], consumers classify the information that they might possess about companies into two types of corporate associations: perceived corporate ability (CA) associations, and corporate social responsibility (CSR) associations. These authors defined CA as the "company's expertise in producing and delivering product and/or services offerings," and CSR as "the character of the company, usually with regard to important societal issues" [11] (p. 70) CRM actions can affect consumer perceptions of both CSR and CA, so it is convenient to study them together to better understand the influence of social actions on consumer behavior [14,15,28]. Perceived CSR and perceived CA can influence consumer satisfaction, which has been defined as an overall assessment that the consumer makes of his/her experience of buying and consuming a product or service [29]. Finally, one of the most important expected outcomes of company strategy is loyalty intention, which has been defined "as the consumer's intention to behave in the retailer's interest, which can manifest itself in different ways, such as stating that they will continue buying at that retailer, that they will recommend it or that they will increase their purchases there in the coming months" [30]. Loyalty is important because small increases in customer retention rates can cause large increases in company profits [31].

Hypotheses Development
Previous studies have shown that the companies that carry out social activities with a high fit with their main activity are better regarded by consumers [3,32,33]. A strong association between the company and the social cause will evoke positive feelings among consumers due to associative learning [18]. Previous studies that examined fit unidimensionally showed that, if the fit is high, the company's message is assumed to be credible, and this increases the company's credibility [34][35][36]. A greater company-social cause fit minimizes the consumer's judgments of selfishness on the part of the company and increases its credibility [27,37]. Bigné et al. [17] found that when the fit was high, consumers needed less cognitive effort to form their judgments and the companies were considered to be more credible. Becker-Olsen et al. [34] confirmed that when the fit between the company and the social cause was low, the consumer felt that the company was less credible. Thomas et al. [36] suggested that the companies that align their social actions with their business are more successful because consumers recognize this compatibility and, therefore, regard them as more credible. Consequently, the following hypotheses are proposed: Hypothesis 1 (H1). Functional fit positively affects perceived corporate credibility. Hypothesis 2 (H2). Image fit positively affects perceived corporate credibility.
Social actions influence the consumer's perceptions of the quality and reliability of products [35,38,39]. Consumers not only want quality products and services at a lower price, but they also want them to be produced and sold in an environmentally friendly way that follows ethical standards [8]. Companies are considered more expert and their products of higher quality when they have a firm commitment to carry out social actions [40]. For example, Du et al. [9] showed how the use of natural ingredients and environmentally friendly practices by the Body Shop company led consumers to consider their products to be of high quality. Therefore, functional and image fit influences perceptions of corporate ability [35,38]. Consequently, the following hypotheses are proposed:
Consumer attitudes are better toward companies perceived as credible [37]. Greater company credibility favors the consumer's perceptions of its CSR [40,41]. Credibility reduces consumer skepticism about the true motivations of the social actions carried out by companies [27,42]. If companies manage to convince consumers that their relationships with social causes are credible, consumers will have better perceptions of their CSR [16]. Previous studies have empirically demonstrated the positive effect of company credibility on consumer perceptions of their CSR [13,15,17].

Hypothesis 5 (H5). Corporate credibility positively affects perceived CSR.
When choosing a product, consumers are not only worried about their immediate satisfaction, but are also concerned that the production and selling of the product is responsible [4]. CSR actions carried out by companies create a favorable context for consumers to develop positive attitudes toward them [11,18,28]. When a company carries out a social action, consumers perceive that its products and/or services have greater value, which increases their satisfaction [14,29,43,44]. CSR initiatives positively influence consumer perceptions of the benefits they receive from their relationships with companies [7]. Previous studies have empirically demonstrated that the performance of CSR activities by companies positively influences consumer satisfaction [14,29,44]. In accordance with the above, the following hypothesis is proposed: Hypothesis 6 (H6). Perceived CSR positively affects customer satisfaction.
CA associations are indicators that the consumer uses to evaluate companies [11]. Consumers do not normally have sufficient technical knowledge to evaluate a company's products, so they use their perceptions of the company's CA as an indicator. CA positively influences consumer attitudes toward companies [28] and their relationships with them [9]. In addition, companies with high CA are well regarded in the marketplace, so consumers will be happy to purchase their products, as this will contribute to improving their own identity [45]. Scholars have empirically demonstrated that consumer perceptions of a company's competence or ability positively influences their satisfaction with its brands [43,46]. Therefore, the following hypothesis is proposed: Hypothesis 7 (H7). Corporate ability positively affects customer satisfaction.
Satisfying customers increases their loyalty. Satisfaction is an overall assessment of perceived value and service quality; satisfied customers are more likely to repeat their buying behavior [47]. Satisfied customers are less motivated to look for alternative suppliers and are, thus, more likely to repurchase from the same provider [30]. Many studies have provided empirical evidence of the positive influence of customer satisfaction on customer loyalty [14,43,48]. Therefore, the following hypothesis is proposed: Hypothesis 8 (H8). Satisfaction positively affects customer loyalty.

Moderating Effect of the Publicizing Medium
The marketing communications landscape (including advertising) is evolving at a very dynamic pace as media and audience fragmentation processes and technological advances (i.e., sophisticated mobile devices) contribute to the creation of new types of interaction between companies and consumers and between companies and other stakeholders [49]. Organizations need to understand which communication tools are more effective for branding strategies [16]. Communications can be made through many different media. Social media (SM) is perhaps the most important new medium. It has re-shaped and transformed the nature of communications and the relationships between organizations and their stakeholders [21]. Undoubtedly, in such a rapidly changing environment, it is more important than ever for companies to understand how consumers feel about the messages they receive via the various media platforms they use. This will enable them better to understand how and through which channels they should communicate their socially responsible activities in order to generate more favorable cognitive and/or affective responses [4,8].
Regarding the choice of media, the extant literature suggests that credibility is one of the most important elements of persuasive messages and that it plays a key role in attitude formation and, ultimately, purchasing behavior [50]. Rapid technological developments have provided consumers with greater control over the information they receive and share, especially in the online environment. In this regard, Schultz [51] suggested that the push-pull marketing communications schema shows that consumer attitude toward advertising is a crucial factor in determining when and whether the messages are received. Several studies have shown that consumers prefer advertising in print media to advertising in digital media [52]. Similarly, Kelly et al. [53] found that advertising avoidance is more likely to occur in online social networks than in traditional media (TM). These authors called for more research to shed light on "how advertising, which is designed as a mass media tool, might rethink itself" [53] in an environment regarded as a personal space. Moreover, these authors identified two key barriers to advertising effectiveness: relevance and credibility. These variables are related, "because if advertising is perceived as not being relevant, neither the medium nor the message can be considered credible, and participants become skeptical" [53]. Similarly, Johnson and Kaye [54] showed that the credibility of advertisements is negatively affected if consumers distrust the medium. In addition, Moore and Rodgers [50] showed that the perceived trustworthiness of the medium affects the way consumers perceive the credibility of the messages received via that medium. More importantly, a lack of trust in the communication medium is directly related to a reduction in the attention that the consumer pays to both content and advertising [54].
Some studies have attempted to classify media into groups based on consumer perceptions of their credibility, but the results have been fairly contradictory. On the one hand, Johnson and Kaye [54] found that both online and TM were seen as having the same level of credibility. On the other hand, more recent studies have identified newspapers not only as the most credible medium in general [55], but also as the most credible medium for advertising campaigns, followed by other TM, such as television, magazines, and radio. Indeed, the Internet was found to be the least credible medium in which to advertise and the medium that consumers regard with the highest level of skepticism [50,54,55]. Similarly, in their recent study on political communications, Johnson and Kaye [56] showed that social networking sites were the least credible medium among the nine traditional and online media analyzed, and that the consumer's trust or distrust in these media was the strongest antecedent of their credibility. Moore and Rodgers [50] showed that skepticism toward advertisements is lower in the print media (newspapers and magazines) and greater in the new internet-based media. Furthermore, consumers are very skeptical about the credibility of online social networking sites advertising, and the industry trends emphasize the Internet's lack of credibility as an advertising medium [53]. This phenomenon might be related to the fact that the Internet is more 'task-oriented' than TM; thus, advertisements that distract consumers from their online activities may lead them to completely avoid advertising [57].
In addition, the cognitive and affective characteristics of the advertisements that promote CRM activities evoke positive and negative emotions in consumers that influence their attitude and, therefore, their behavioral intentions [58]. Previous studies have shown that TM informational and informational-emotional advertising appeals to creating positive attitudes toward brands and purchasing intentions [59]. However, the literature has shown that the informational value of an advertisement on SM does not influence the attitude, or, therefore, the behavior, of the consumer [60].
Thus, although there is a lack of research into the moderating effect of the publicizing medium on the model's relationships, according to the above discussion, it seems logical to assert that TM, as Sustainability 2020, 12, 3632 7 of 16 opposed to SM, will have a higher impact on the relationships between fit (functional and image), credibility, CA, CSR, satisfaction, and loyalty intention. Thus, the following hypotheses are proposed: Hypothesis 9 (H9). The impact of the communication medium used to advertise CRM activities on the proposed relationships between functional fit, image fit, corporate credibility, perceived CA, perceived CSR, satisfaction, and loyalty intention is higher for traditional media than for social media.

Procedure
An experiment was performed to evaluate the proposed research model. The data were collected through personal questionnaires. The experimental design allowed for the creation of fictitious situations based on the study objectives; moreover, the researchers were able to exert high control of the study, which enhanced its internal validity. To improve the external validity and generalization of the results, various combinations of social cause, company, and means of communication were used. The questionnaire allowed for the structured collection of data, which facilitated its encoding for later processing. The experiment involved the insertion of an advertisement in a newspaper and on a social network, in which, a retail food sector company associated itself with a non-profit organization (NPO) in a simulated CRM activity. This sector was chosen as the research context because it is very sensitive to CSR [61]. The study consisted of two phases. In the first, using a focus group of 10 consumers who habitually buy food in supermarkets and hypermarkets, a social action and a NPO were selected that were perceived as having a high fit (functional and image) with the retailer. Specifically, from a series of CSR action options, the focus group chose the donation of a percentage of a retailer's sales of a particular foodstuff to a food bank organization. The advertisement text and the retailer's and NPO's logos stated: "C (company name) commits to donating 5% of its milk sales to the (NPO name) food bank." In the second phase, a questionnaire was developed with the model variables and an advertisement was designed for each of the six retail food companies with the highest market share in Spain. Prior to the data collection, a pre-test was carried out with a sample of 25 consumers; this made it possible to improve and clarify the wording of the questions.
For the evaluation of the research model, two independent samples of consumers were selected. The participants each viewed only one advertisement and answered the questionnaire in reference to the retailer where they shopped most frequently. Sample 1 was shown the advertisements inserted in the newspaper, while sample 2 was shown the same advertisement inserted in a social network.

Procedure
An experiment was performed to evaluate the proposed research model. The data were collected through personal questionnaires. The experimental design allowed for the creation of fictitious situations based on the study objectives; moreover, the researchers were able to exert high control of the study, which enhanced its internal validity. To improve the external validity and generalization of the results, various combinations of social cause, company, and means of communication were used. The questionnaire allowed for the structured collection of data, which facilitated its encoding for later processing. The experiment involved the insertion of an advertisement in a newspaper and on a social network, in which, a retail food sector company associated itself with a non-profit organization (NPO) in a simulated CRM activity. This sector was chosen as the research context because it is very sensitive to CSR [61]. The study consisted of two phases. In the first, using a focus group of 10 consumers who habitually buy food in supermarkets and hypermarkets, a social action and a NPO were selected that were perceived as having a high fit (functional and image) with the retailer. Specifically, from a series of CSR action options, the focus group chose the donation of a percentage of a retailer's sales of a particular foodstuff to a food bank organization. The advertisement text and the retailer's and NPO's logos stated: "C (company name) commits to donating 5% of its milk sales to the (NPO name) food bank." In the second phase, a questionnaire was developed with the model variables and an advertisement was designed for each of the six retail food companies with the highest market share in Spain. Prior to the data collection, a pre-test was carried out with a sample of 25 consumers; this made it possible to improve and clarify the wording of the questions. For the evaluation of the research model, two independent samples of consumers were selected. The participants each viewed only one advertisement and answered the questionnaire in reference to the retailer where they shopped most frequently. Sample 1 was shown the advertisements inserted in the newspaper, while sample 2 was shown the same advertisement inserted in a social network.

Measurement Instrument
As indicated above, the data for the evaluation of the research model were collected through a questionnaire. The model constructs were measured using scales validated in previous studies (see Appendix A) that had been proven to have good reliability and validity in the context of CSR. Functional and image fit and corporate credibility were measured using 7-point semantic differential scales-the first two adopted from Becker-Olsen and Hill [39] and Bigné et al. [17], and the third from McWilliams and Siegel [41] and Bigné et al. [17]. The remaining constructs were measured using 7-point Likert-type scales, where 1 meant "totally disagree" and 7 meant "totally agree." The CA and CSR measurement scales were adopted from Berens et al. [38] and Walsh and Bartikowski [9]. The satisfaction measurement scale was adapted from Cronin et al. [47]. Finally, loyalty intention was measured using the Maxham and Netemeyer scale [62]. In accordance with previous studies (see Homburg et al. [39]; Oliver and Swan [63]; Jarvis et al. [64]; Lafferty et al. [65]), the epistemological relationships-that is, the links between the theoretical constructs and the empirical data-were considered as reflective in all cases.

Data Collection and Sample
Malaga, with a population of 569,000, is a city in the Andalusian region (Spain). Situated on the Costa del Sol (Coast of the Sun) of the Mediterranean Sea, it is the Southernmost largest city in Europe. Despite its importance within its own geographical area, it is considered a medium-sized city because its population is much smaller than that of large European cities, such as Berlin, London, and Madrid. Malaga is recognized internationally for its energy efficiency projects, its promotion and attraction of research, development and innovation (R&D&I), and its business acceleration support [23].
The study population was consumers from 18 to 65 years old residing in Malaga who had frequently shopped in a supermarket or hypermarket in the previous year. The questionnaire was administered using a face-to-face method. The samples were selected by quota sampling by age and gender in an intercept approach. Quota sampling is a fast and inexpensive method that assigns quotas (e.g., age and gender) in the sample according to the study population proportion. Field work was done from January to February, 2020. The response rate was 87%. The sample size was 421 valid questionnaires: 213 for sample 1 (TM) and 208 for sample 2 (SM) ( Table 1).

Data Analysis
The research model was evaluated through covariance-based structural equation modeling (CB-SEM), following the recommendations of Anderson and Gerbing [66]. First, the goodness of fit of the measuring instruments was analyzed using confirmatory factor analysis (CFA). Next, the compliance of the psychometric properties of the model were verified. Thereafter, using CB-SEM, the structural relationships of the model were evaluated. A multi-group analysis was performed to assess the moderating effect of the publicizing medium. STATA 15 software was used for the estimate. Table 2 shows that all the model variables exceeded the recommended minimum values of the reliability tests. Cronbach's alpha (α) was greater than 0.7, the composite reliability (CR) was greater than 0.7, and the average variance extracted (AVE) was greater than 0.5 [67]. The measures of validity were also adequate. The standardized loading coefficients were greater than 0.5 and their averages greater than 0.7 [68] (Table 2). In addition, the discriminant validity of the measurement model was confirmed, as none of the confidence intervals of the correlations contained the value 1 [66,68] and the AVE was larger than the shared variance [67] (Table 3). Regarding the goodness of fit measures (Table 2), the root mean square error of approximation (RMSEA), which is considered a very robust index [69], was below the recommended limit of 0.08 (0.039). The comparative fit index (CFI) and the Tucker-Lewis index (TLI) presented values greater than the recommended minimum of 0.9 (0.973 and 0.97, respectively) [70]. The only fitness measurement unfulfilled was χ2, but it is very sensitive to sample size and, very often, when large samples are used, the model is rejected [71]. Therefore, the research model fit well with the sample data. Figure 2 shows the results of the structural estimation model for TM and SM. The standardized path coefficients of the relationships between the model variables were significant at the 1% level, except the relationships between image fit and CA in the TM model (p-value < 0.05) and between functional fit and CA in the SM model (not significant). Therefore, the results confirm that functional fit and image fit are positively associated with corporate credibility and perceived corporate ability in both models (TM and SM), although the relationship between functional fit and corporate ability is not significant in the SM model. Corporate credibility has a strong effect on perceived CSR, which in turn positively impact on customer satisfaction in both models. Moreover, the findings show that perceived corporate ability has a stronger impact on satisfaction than CSR. Finally, in both models (TM and SM), customer satisfaction shows a positive and strong influence on loyalty intention. on loyalty intention. Table 4 shows the hypotheses testing. The hypotheses were accepted in both media (H1 to H8 supported), with the exception of H3 in SM. The CFI was higher than the recommended minimum value of 0.9 in both models (TM: 0.914; SM: 0.921). The TLI was greater than the recommended minimum of 0.9 (TM: 0.908; SM: 0.915). Last, the RMSEA did not exceed the recommended maximum of 0.08 (TM: 0.073; SM: 0.066).

Assessment of the Structural Model and Hypotheses Tests
A multi-group analysis was conducted to assess whether the model relationships had significantly different values based on the communication medium. Lagrange multipliers were used to evaluate whether there were significant differences among the structural coefficients based on the χ2 differences between the two models (Table 4).
The analysis confirmed significant differences in two model relationships: between functional fit and CA and between image fit and CA. The relationship between functional fit and CA was significant in the TM but not in the SM. The influence of image fit on CA (H4) was significantly greater in the SM (β = 0.6287) than in the TM (β = 0.2558). On the other hand, the causal relationship between functional fit and corporate ability was higher in the TM (β = 0.3729). Finally, the most intense direct causal relationship in the traditional means of communication was between satisfaction and loyalty (β = 0.8155), while in the social, it was between credibility and CSR (β = 0.7787). This demonstrates that in the social medium, it is essential that the social actions carried out are credible.    Table 4 shows the hypotheses testing. The hypotheses were accepted in both media (H1 to H8 supported), with the exception of H3 in SM. The CFI was higher than the recommended minimum value of 0.9 in both models (TM: 0.914; SM: 0.921). The TLI was greater than the recommended minimum of 0.9 (TM: 0.908; SM: 0.915). Last, the RMSEA did not exceed the recommended maximum of 0.08 (TM: 0.073; SM: 0.066). A multi-group analysis was conducted to assess whether the model relationships had significantly different values based on the communication medium. Lagrange multipliers were used to evaluate whether there were significant differences among the structural coefficients based on the χ2 differences between the two models ( Table 4).
The analysis confirmed significant differences in two model relationships: between functional fit and CA and between image fit and CA. The relationship between functional fit and CA was significant in the TM but not in the SM. The influence of image fit on CA (H4) was significantly greater in the SM (β = 0.6287) than in the TM (β = 0.2558). On the other hand, the causal relationship between functional fit and corporate ability was higher in the TM (β = 0.3729). Finally, the most intense direct causal relationship in the traditional means of communication was between satisfaction and loyalty (β = 0.8155), while in the social, it was between credibility and CSR (β = 0.7787). This demonstrates that in the social medium, it is essential that the social actions carried out are credible.

Discussion and Conclusions
An increasing number of companies have the establishment of CRM programs among their objectives to contribute to the creation of sustainable environments. However, despite the resources that companies are dedicating toward this objective, there has hitherto been limited research into the effects that CRM activities have on consumer behavior. In particular, few studies have analyzed how the communication medium used to publicize CRM actions influences consumer behavior. The authors' aim in the present study was to address this concern by analyzing the moderating effect of TM vs. SM on the relationships between perceived cause-company fit and customer loyalty. To this end, a model of consumer behavior was proposed and evaluated, which examined the dual nature of the fit (functional and image) between companies' social actions and consumer loyalty, through corporate credibility, corporate ability, perceived CSR, and satisfaction. This research is original in analyzing the effects of the means of communication used to promote CRM activity on the relationships between the consumer's perceptions of cause-company fit and their loyalty intention. In addition, the model includes causal relationships that are studied little in the context of the CRM, such as the proposed relationships between functional and image fit and CA and company credibility, and between CA and satisfaction. Taking all this into account, it is considered that the present study makes an important contribution to the theory and practice of marketing.
The data used to evaluate the model were obtained from a sample of residents in a medium-sized Spanish city noted internationally for its sustainable development activities, such as its smart city program. The model was evaluated through covariance-based structural equation modeling (CB-SEM), and the moderating effect of the publicizing medium was evaluated through a multi-group analysis. Based on the results, the study makes several important contributions to the CRM literature.
Firstly, the proposed model has high validity, as all the proposed relationships are statistically supported, except for H3 in the social medium. Therefore, it can be affirmed that fit, CA, corporate credibility, and perceived CSR are important antecedents of consumer loyalty, mediated by satisfaction. Several authors have previously noted the positive impact of CA, credibility, and CSR on loyalty [13,14], but very few studies have highlighted the key role of satisfaction as a mediator between perceived CSR and loyalty, and between CA and loyalty. This role was observed, with no significant differences being noted, in both the TM and SM.
Secondly, the dual conceptualization of fit (functional and image) allowed us to identify the only two differences between the relationships in the two models: between functional fit and CA (H3) and between image fit and CA (H4). Specifically, the effect of functional fit on CA is greater in the traditional than in the social medium; on the other hand, the effect of image fit is greater in the social than in the traditional medium. These results extend the findings of previous studies [17,37] and support the validity of the dual conceptualization of fit over studies that have considered it a one-dimensional construct [13,15,28,35,72]. In addition, this study contributes by highlighting the role of the fit between social actions and companies as an antecedent of consumer loyalty.
Thirdly, the study contributes by increasing the knowledge of the effect of the publicizing medium on consumer behavior. The results showed that the effects are perhaps more uniform than expected, given that significant differences were observed in only three model relationships. This is one of the first studies to show significant differences due to the moderating effect of the publicizing medium on the relationships between functional and image fit and corporate credibility and CA, extending the results of previous studies [13,14]. The high degree of uniformity of the model may be due to the increasing difficulty that the consumer faces in differentiating between traditional and online media, as the TM have now adopted online channels as an important part of their dissemination strategies.
Fourthly, residents of a medium-sized smart city perceived companies' social activities favorably. This is an important contribution, as hitherto, no research has been carried out to study consumer perceptions of companies' CSR activities in a city with a clear orientation toward sustainable development. However, the sample did not allow a comparison of possible differences between residents in cities with varied levels of orientation toward sustainability.
The results also have a series of implications for the management and publicizing of corporate social responsibility activities. Firstly, companies must carry out social activities because, not only do they contribute to more sustainable development, but also because they obtain a return through increased consumer loyalty. Loyalty, because of its important impact on profits, is one of the most precious company assets, as previous studies have indicated [31]. Secondly, companies must carry out social actions because they positively influence the consumer's perception of their CA. A better CA perception contributes to improving how the company and its products are viewed in the market. Thirdly, in their choice of social actions, companies must pay special attention to both the functional and the image fit between the actions and the company. In this sense, if the action has good fit (functional and image) with the company, it would be more advisable to place related advertisements in newspapers than on social networks. On the contrary, if only the image fit is good, it would be more advisable to place the advertisements on social networks. Fourthly, if the budget is limited and the prices are equivalent, it would be more advisable to place advertisements in newspapers than on social networks, as the effect on loyalty is greater. Fifthly, for social actions to positively influence loyalty, various aspects must be addressed, such as company credibility, CA, perceived CSR, and consumer satisfaction. None of these can be neglected because they all influence loyalty.
Based on the above points, it is considered that this study contributes to improving the knowledge of consumer responses to the performance of social actions by companies. However, there remain some aspects that would benefit from further research. The research was conducted only in a medium-sized smart city, so there is the potential for single dataset bias, and there is limited scope to generalize the results. Ideally, the same methodology might be applied, but in some differently-sized towns with different orientations toward sustainable development. Furthermore, future studies could also extend the analysis by evaluating the effect of other variables, such as consumer-company identification [72], perceived value [73], and trust [74], since this has been found to impact the consumer's attitude towards companies.   CA1. The company offers good quality products and services CA2. The company is sound and reliable CA3. The company has capacity for innovation and development CA4. The company is well managed CA5. The company offers products with a good price-quality ratio Corporate Social Responsibility CSR1. The company supports activities with social causes CSR2. The company is concerned about the environment CSR3. The company tries to improve the quality of life of the local communities in which it does business CSR4. The company treats its employees well CSR5. The company supports needy children Satisfaction SAT1. My choice to buy from this company was right SAT2. The facilities are adequate SAT3. I think I did the right thing when I bought from this company SAT4. This company is nice SAT5. I like to buy from this company SAT6. This company gives me a very positive impression SAT7. I am interested in buying from this company Loyalty intention LOY1. In the future I intend to shop in this establishment LOY2. I am likely to shop in this establishment LOY3. In the near future, I will not shop in this establishment (R) LOY4. In the future I will continue shopping at this establishment LOY5. I am loyal to this establishment (R): Reverse score.