Determinants of Intention to Forward Online Company-generated Content via Facebook

The aim of the present study is to analyze consumer intention to forward online company-generated content (CGC) from the retail banking sector via Facebook, the leading social network, based on (a) the mediating effect of attitude to forwarding online CGC on the relationship between brand experience and the intention to forward online CGC and (b) the mediating effect of Facebook perceived usefulness on the influence of subjective norms and source trustworthiness on the intention to forward online CGC. A sample of 404 users was recruited, and data collected by means of questionnaires distributed via email and personal interviews. The dataset obtained was analyzed using Structural Equation Modeling. The findings demonstrate the mediating effect of attitude as the main antecedent of intention to forward online CGC. Relevant scholarly and managerial implications are drawn for financial entities and the agencies that provide them with corporate and marketing communications services.


INTRODUCTION
Technological developments, the emergence of social media (such as Facebook, Twitter, and YouTube), and the increasingly diverse range of communication devices available, such as smartphones and tablets, have contributed to an unprecedented change in the way relationships are established and nourished in the current social environment (Zhang et al., 2014).
Social media are a group of Internet-based applications that facilitate the creation and exchange of user-generated content (UGC) (Charlesworth, 2018). Users share information and opinions by forwarding content about products and brands, among other functionalities (Liang et al., 2011). Consumers are therefore no longer dependent on the information provided by companies through traditional channels or on their websites (Chen and Shen, 2015). Furthermore, the rapid growth and adoption of social media globally have made them the focal point for business decisionmakers (Mir and Rehman, 2013).
The recently-released 2018 Global Digital suite of reports from We Are Social and Hootsuite reveals that there are now more than 4 billion people around the world using the Internet. Facebook is currently the social network with the greatest number of users (currently standing at 2167 million), followed by YouTube (1500 million users) and WhatsApp (1300 million) (We are Social and Hootsuite, 2018). These volumes indicate the importance of social networks for firms.
Over the last decade, scholarly research has advocated the value of social media interactions for companies, and has proposed a range of conceptual frameworks to support the achievement of key business goals, such as to: Manage brands (Gensler et al., 2013) and customers (Malthouse et al., 2013); co-create content This Journal is licensed under a Creative Commons Attribution 4.0 International License (Nyangwe and Buhalis, 2018); develop metrics (Peters et al., 2013); manage brand crises (Salvador and Ikeda, 2018); and assess marketing potential in the social media context (Yadav et al., 2013). However, there are very few studies that have analyzed company-generated content (CGC) on social networks (specifically, on Facebook). The present work therefore makes a dual contribution to the literature: In the nature of the study itself, which analyzes intention to forward online CGC, and in the nature of the antecedents it proposes.
Based on these considerations, the present investigation proposes the inclusion of brand experience, subjective norms, source trustworthiness, consumer attitude, and perceived usefulness of the communication medium as variables related to CGC, to examine the relationships between such variables and users' behavioral intention to forward (share) CGC. Brand experience is not a new concept, having been tested by many authors and applied to different cases and contexts. Developed over several decades, the concept is currently more crucial than ever before, since customers savvy in Internet and social media use are much more focused on the brand experience than on the benefits and functions of branded products and services. In turn, those organizations achieving a superior brand experience generate additional competitive advantage, customer loyalty, and uniqueness in the market, thanks to customer engagement with the brand (Bapat, 2017). The role of subjective norms has been analyzed by several authors, and the present study adopts the definition provided by Fishbein and Ajzen (1975) to test their effect on the behavior of forwarding and sharing CGC.
In addition, the study examines the relationship between the perceived usefulness of Facebook and the decision to forward online CGC. Trustworthiness as a source of credibility needs to be tested to understand how best to target the selected audience with the persuasive message. However, this gives rise to an important question: Does the content of the persuasive message motivate the audience sufficiently to forward online CGC through Facebook?
This paper is structured as follows: Section 2 introduces the literature review and hypothesis development; section 3 details the methodology; Section 4 presents the data analysis and research results; section 5 provides a discussion of the implications of the results; and, finally, section 6 explains the main limitations of the study and potential areas for further research.

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT
The purpose of the present study is to measure the effect -direct and/or indirect -of subjective norms, trustworthiness as source of credibility, brand experience, and attitude on intention to forward online CGC via Facebook.
This study focuses on the analysis of brand experience, subjective norms, attitudes, and usefulness and these concepts are incorporated in the research model to reveal their impact on intention to forward online company generated contents. In this regard, it is to be expected that positive or negative experiences with such brand will be directly related to customer's senses and behaviors towards such brand. Thus, we suggest that prior experience with a brand will guide customer's behaviors when they are exposed to advertisements or posts generated by the company via social media, negative or positive forwarding deriving from such customer brand experience.
With these premises, the researcher will examine how brand experience affects judgments, attitudes, and behavioral patterns, namely consumer intention to share and forward online company generated contents. More specifically, to achieve the research purposes we will measure the impact of brand experience, subjective norms, attitudes, and usefulness in the financial sector.
Firstly, we examine whether experience will affect customer attitudes and intention to forward online company generated contents. Secondly, we focused on the concept of perceived usefulness to look into its impact on the intentions of customers to forward online company generated contents. Trustworthiness is a relevant driver of the intention to forward the online company generated contents. Finally, subjective norms are analyzed in this study to reveal whether customers influence each other to forward the online company generated contents via the use of Facebook and Facebook community page.

Subjective Norms
Social influences, in the form of subjective norms, are used as factors in technology acceptance models and in their later adaptations (Venkatesh and Bala, 2008). Subjective norms suggest that behavior is instigated by one's desire to act in line with how influential individuals (e.g. friends, family, or society in general) believe one should act, or act themselves (Bearden et al., 1989). In other words, subjective norms are the perceived social pressures an individual faces when deciding whether to behave in a certain way. In the context of the present study, the subjective norm is the degree to which the client of the financial entity perceives that sharing content from that entity via the social network will improve their status (Rejón-Guardia et al., 2011). From this point of view, many authors have identified a direct and positive link between subjective norms and perceived usefulness (Zhanga et al., 2011;Liébana-Cabanillas and Alonso-Dos-Santos, 2017). Moreover, the theory of reasoned action (TRA) (Fishbein and Ajzen, 1975) and the theory of planned behavior (TPB) (Ajzen, 1991) state that subjective norms positively influence behavioral intention. In this regard, Pavlou and Fygenson (2006), Li et al. (2008), Shin (2009), andLiao (2017), among others, have indicated that subjective norms positively affect behavioral intention. Zhou (2011) proposed that subjective norms positively affect virtual community members' participation intention. Therefore, those users of the social network who feel social norms more strongly will have a greater intention to share CGC via Facebook. Based on the above rationale, the following hypotheses are formulated: H 1 : Subjective norms have a positive influence on the perceived usefulness of the medium (Facebook). H 2 : Subjective norms have a positive influence on behavioral intention to forward online CGC.

Trustworthiness
Trustworthiness here refers to "the perceived willingness of the source to make valid assertions" (McCracken, 1989. p. 311) and reflects an audience's belief that the communicator provides information in an honest, fair, and sincere manner (Ohanian, 1991). In the information systems literature, the positive effect of perceived trustworthiness on perceived usefulness has been widely acknowledged in the context of particular systems (e.g., Benbasat and Wang, 2005). Trust is defined as "the willingness of a party to be vulnerable [trustor, giving trust] to the actions of another party (trustee, receiving trust) based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party" (Mayer et al., 1995. p. 712). Prior literature has verified the relationship between trustworthiness, perceived usefulness (Walter et al., 2015;Wu and Lin, 2017), and attitude Chang et al., 2017). Thus, it is expected that the trustworthiness of the online content generated by Middle Eastern banks on Facebook will influence the customer's attitudes and behavioral intention to forward that content via Facebook. On this premise, the following hypotheses are proposed: H 3 : Perceived trustworthiness of the CGC has a significant positive effect on perceived usefulness of the medium (Facebook). H 4 : Perceived trustworthiness of the CGC has a significant positive effect on attitude to forwarding online CGC.

Brand Experience
Experience has been studied from various different perspectives, such as product experience, shopping and service experience, customer experience, and brand experience (Gentile et al., 2007;Bapat and Thanigan, 2016).
For customers, brand experience has been found to be an important factor that can help improve their engagement with, and commitment to, the brand. The firm attempts to achieve competitive advantage through experience by developing a higher level of loyalty from customers, who are willing to pay a higher price for a brand that offers a superior brand experience (Bapat, 2017). Brands that facilitate a greater number of experiential affordances on their Facebook pages achieve higher levels of consumer engagement, encouraging brand post "likes" and "shares" (Tafesse, 2016).
According to Schmitt's (1999) definition, experiential marketing sees consumers as rational and emotional human beings interested in the achievement of pleasant experiences; while traditional marketing views consumers as rational decision-makers who care about functional features and benefits. In addition, Schmitt (1999) listed five different types of experiences that marketers can create for customers (namely, sensory, emotional, cognitive, behavioral, and relational), while  adopted four dimensions of experience (affective, behavioral, intellectual, and sensory). Moreover, customers seem to value brands that are able to provide them with attractive experiences. In other words, consumers expect something distinct from brands and search for something that engages their senses and touches their hearts ).
According to , brand experience can be defined as the "sensations, feelings, cognitions and behavioral responses evoked by brand-related stimuli that are part of brand design and identity, packaging, communications and environments."  believes that the concept of brand experience relates to the essence of branding more than other analytically-and cognitivelyoriented brand concepts such as brand equity, brand value, brand associations, brand attitudes, or brand personality. Based on the unique characteristics of, and main differences between, services and products, Nysveen et al. (2013) examined the brand experience for service brands with five dimensions (affective, behavioral, intellectual, sensory, and relational), while  examined product brands via the aforementioned dimensions (affective, behavioral, intellectual, sensory). The latter theoretical framework has been adopted for the present study, via the examination of the impact of the four dimensions of brand experience on customer attitudes and their behavioral intentions to forward online content. Huang (2017) found that affective and intellectual experiences exert positive effects on brand love, while sensory and behavioral experiences exert positive effects on brand trust, the latter not being influenced by intellectual experiences.
In the present research we will examine the relationship between customers' experience with a given brand and their intention to forward CGC. Several scholars have examined the impact of: Brand experience on different variables, such as brand loyalty (Huang, 2017); consumer-brand engagement on brand experience and behavioral intentions (Risitano et al., 2017); experiences with store brands on purchase intention (Diallo and Siqueira, 2017); and brand experience in profiling consumers and predicting consumer behavior (Zarantonello and Schmitt, 2010), among others. From an analytical perspective, we have observed a correlation between the latent first-order factors analyzed in this study. Although there is a correlation, we can also discuss isolated factors (Satorra, 2002). In other words, we can examine sub-dimensions of a more extensive factor (Del Barrio-García and Luque-Martínez, 2012) formed by the affective, behavioral, intellectual, and sensory dimensions." Different authors have corroborated the relationship between brand experience and attitude (Zarantonello and Schmitt, 2013;Mourad and Valette-Florence, 2017), on the basis of which the following hypothesis is proposed: H 5 : The more favorable consumers' brand experience (affective, behavioral, intellectual, and sensory) of their particular bank, the more positive their attitude to forwarding the online CGC of that bank.

Perceived Usefulness
Perceived usefulness is defined by Davis (1989) as "the degree to which a person believes that using a particular system would enhance his or her job performance". Perceived usefulness is the expected value of using a particular information system (Huang, 2017).
Different studies have demonstrated that perceived usefulness has a direct relationship with attitude Upadhyay et al., 2018;Santhanamery and Ramayah, 2018) -Santos, 2017). In the context of the present research, we consider that the perceived usefulness of Facebook will influence users' intention to forward content generated by Middle Eastern banks. Based on this premise, the following hypotheses are formulated: H 6 : Perceived usefulness has a positive influence on attitude to forwarding online CGC. H 7 : Perceived usefulness has a positive influence on behavioral intention to forward online CGC.

Attitude to Forwarding Online CGC
Attitudes reflect the favorable or unfavorable feelings people have toward a given behavior (Fishbein, 1963;Premkumar et al., 2008). Therefore, attitudes develop over time, as a person acquires experience. The different theoretical models (the technology acceptance model, the TRA, and the TPB) show that attitude is a fundamental antecedent of the intention to perform a given behavior (Ajzen and Fishbein, 1980;Pee et al., 2008;Hoque et al., 2018). Fishbein and Ajzen (1975) consider attitude to be a multidimensional construct, comprising cognitive, affective, and conative (behavioral) aspects. The cognitive component refers to what a person knows about a product or service (based on experiences, beliefs, and opinions). The affective component refers to the individual's personal tastes regarding the product or service (feelings, emotions, and evaluations). And the conative component refers to behavioral intention (purchase intention, purchase response, and rejection response). In the present case, the attitude to forwarding online CGC will influence the intention to forward online CGC, as other works have established (Amin et al., 2011;Hsu et al., 2013;Ayeh et al., 2013;AlSaleh, 2018). The following hypothesis is therefore proposed: H 8 : Attitude to forwarding online CGC has a positive influence on behavioral intention to forward online CGC.
All the hypotheses formulated in this study and included in the proposed conceptual model are summarized in Figure 1.

Sample Design and Data-gathering
Data were gathered by means of a questionnaire distributed via email and personal interviews. A total of 530 questionnaires were distributed among customers of financial entities operating in the Middle East, resulting in 404 valid responses and a 76.3% response rate. The description of the respondents' profile is shown in Table 1.

Measurement Instruments
The measurement instrument for the present study was adapted from previous published works (Appendix 1). A structured questionnaire was developed to cover the variables analyzed in the study (personal data, independent variables, and dependent variables), which were measured using 7-point Likert scales.
The instrument was translated from English into Arabic by professional translators to ensure uniformity and consistency. The researchers used email and personal interviews to distribute the questionnaires. To test the hypotheses proposed in the model, data were collected using the convenience sampling procedure. The questionnaire was divided into two major sections: The first to gather demographic information, and the second to measure variables proposed by the model. The measurement instrument used to assess brand experience (affective, behavioral, intellectual, and sensory) was drawn from . The scales to measure perceived usefulness, attitude and use intention were adapted from Davis (1989). The subjective norms scale was adapted from Bansal and Taylor (2002); and finally, the trustworthiness scale was adapted from Ohanian (1990).

Common Method Bias
We also tested the influence of common method variance by conducting Harman's single factor test. Recent research has used confirmatory factor analysis as a more sophisticated test for the hypotheses than a single factor accounting for all the variations in the data (Podsakoff and Organ, 1986;Podsakoff 2003). Harman's single factor test revealed that no single factor that accumulated the total variance emerged and the first extracted factor explains 48.66% of the variance. When the model includes all of the factors, it justifies 75.89% of the explained variance. These results

Pilot Test
A pilot study was performed among 32 academics to test the quality of the questionnaire items, using the same data collection procedures explained in the above section. The findings led to one item of the trustworthiness scale being eliminated, due to the fact that the words 'trustworthy' and 'reliable' share the same meaning in Arabic language.

RESULTS
Following the relevant methodological checks, using the methods and models described in the present work, we conducted an analysis of the measurement model and the resulting structural relationships.

Reliability and Validity of Measurement Items
To verify the suitability of the measurement scales used in the study, we applied various types of analyses of reliability and validity, both exploratory (using SPSS 24.0) and confirmatory (AMOS 18). Measurement instruments are deemed to be valid when they truly measure what they are intended to measure. And they are considered reliable when they provide stable, consistent scores and the measurements match those taken using equivalent or very similar methods.
First, the factor loadings were analyzed. These measure the direct relationship between the latent variable and the variable under observation. Factor loadings should be significant and present a standardized value of over 0.7. Variables presenting lower values should be eliminated, except in the case of indices that do not fulfill this requirement but do contribute to the validity of content (Hair et al., 2010). Some authors, such as Bagozzi and Yi (1988), propose that the value threshold should be 0.6.
Reliability was measured using composite reliability (CR) analysis and Cronbach's alpha (α) for internal consistency. Convergent and discriminant validity were also measured, using average variance extracted (AVE) and correlations between constructs. In the case of the Cronbach's α indicator, the optimum criterion is that of a value surpassing 0.7 (Hair et al., 2010). In the present analysis, all of the variables fulfill this criterion. CR measures the internal consistency of the set of indices of each latent variable (Mangin and Mallou, 2003), that is, the degree to which the latent variable explains the variance of its indices. Values above the threshold of 0.70 (Del Barrio-García and Luque-Martínez, 2012) are recommended. The AVE represents the part of the variance shared by the items overall and the latent variable. In the present case, it is recommended that values are in excess of 0.50 (Del Barrio-García and Luque-Martínez, 2012). Table 2 shows that all the variables fulfilled the criteria for reliability and AVE proposed in the literature.
Finally, discriminant validity was tested by verifying the correlations between each of the variables and their constructs.
According to the criterion of the confidence interval test, none of the confidence intervals of 95% for the individual elements of the latent factors should contain 1 (Anderson and Gerbing, 1988).
In view of the results of these analyses, it can be affirmed that the scale presents convergent and discriminant validity.

Assessing the Model Fit
After analyzing the reliability and validity of the scales, we proceeded to assess the fit of the proposed structural model, prior to testing the research hypotheses As recommended by Hair et al. (2010), the most appropriate estimation procedure was deemed to be the maximum likelihood method with the resampling or bootstrapping technique (using 500 replications). In this resampling technique, the corrected Bollen-Stine p-value was used and the standard correction of factorial errors with a reliability level of 95% was applied. The analysis of absolute measures, incremental and parsimony, suggests that the model fit is reasonably good (Bollen, 1990) (Table 3).

Hypothesis-testing Results
Once the validity of the model fit was validated, the final result was reflected in a graph based on the coefficients standardized for the behavioral model that measures the Intention to forward online CGC through Facebook (Table 4 and Figure 2). First, the hypotheses relating to subjective norms could not be entirely rejected. H 1 , which proposed a relationship between subjective norms and perceived usefulness, was confirmed (β = 0.69; P = 0.000); this finding is in line with the results of previous studies (e.g., Liébana-Cabanillas and Alonso-Dos-Santos, 2017). H 2 , which proposed a positive and significant relationship between subjective norms and attitude presented a lack of empirical evidence (P > 0.05), and could therefore not be confirmed.
Nor was there sufficient empirical evidence to (P > 0.05) to support the hypotheses proposing that trustworthiness exerts an effect on perceived usefulness and attitude (H 3 and H 4 , respectively). This result is in contrast to the findings of the extant literature.
H 5 , which dealt with the possible influence of brand experience on attitude, can be confirmed (β = 0.865; P = 0.000). We therefore conclude that brand experience exerts a direct effect on attitude, as also affirmed in previous studies (Mourad and Valette-Florence, 2017).
The hypotheses dealing with perceived usefulness could not be entirely rejected. On the one hand, H 6 , which proposed a positive relationship between perceived usefulness and attitude, can be confirmed (β = 0.209; P = 0.000). This result is in line with those of Santhanamery and Ramayah (2018). By contrast, H 7 , which linked perceived usefulness and intention to forward online CGC through Facebook, found no empirical support (P > 0.05).
Finally, H 8 , which proposed a direct and positive relationship between attitude and intention to forward online CGC through   Facebook, found empirical support (β = 0.999; P = 0.000), which echoes the findings of the extant literature (AlSaleh, 2018;Jamshidi and Hussin, 2018). It is worth noting in particular the importance of the relationship between these variables in the proposed behavioral model.

Theoretical Implications
The main objective of this research was to examine the relationship between brand experience and customer intentions to forward CGC, which is important for both marketers and academicians. A better understanding of this relationship will be crucial in guiding both practitioners and scholars to make an informed decision regarding customers' attitudes and intentions to forward online content.
This study suggests a rational route to intention to forward online CGC, and it demonstrates the mediating role of attitude to forward online CGC. Based on the previous finding, brand experience was found to affect the consumer's behavioral intention to forward CGC. Such results are in line with those obtained by , who indicated that brand experience has a behavioral impact and an indirect effect on consumer satisfaction and loyalty mediated through brand personality. Moreover, the results are consistent with the findings obtained by Iglesias et al. (2011), Shamim and Butt (2013), and (Ding and Tseng, 2015). In fact, brand experience was found to affect customer loyalty only through affective commitment as a mediator variable and to have a strong relationship with consumer-based brand equity and brand attitude when the effect is mediated by brand credibility/ brand attitude. A strong relationship between brand experience and brand loyalty was found to be mediated by hedonic experience.

Practical Implications
This study suggests a managerial approach to encouraging customers to forward CGC among their friends and families. Companies wanting their customers to forward their online should shape the customers' attitudes to forwarding online CGC through the maximization of their brand experience. In practice, this could be achieved, for example, by formulating a marketing strategy focused on customers and developing a marketing campaign based on providing services to match customers' needs, wants, desires, and demand. Companies should also track customer satisfaction, loyalty, after-sales service, and negative and positive word-ofmouth via social networks, responding to customer requests or concerns in a timely fashion so that they are less likely to switch to competitors' brands. Finally, practitioners and marketers are encouraged to care about the customer brand experience, as the content posted on social media by customers is much more credible than any CGC.

Limitations and Future Research
Some limitations of this study need to be acknowledged. First, the generalizability of the findings should be taken into consideration.
In this regard, the characteristics of the sample represent a relevant limitation, since data were gathered from customers of financial entities operating in a specific geographic area (the Middle East). Thus, future research is needed to test the proposed model in other geographical contexts. The second limitation derives from the sector selected for the field of study, therefore researchers are encouraged to conduct research across other industries, to further investigate the influence of some of the variables included in this study, whose effect on intention was found not to be significant. Furthermore, it would be valuable to investigate the impact of brand experience in other service categories. Finally, future studies are called-for to test the same relationships in banking services through a cross-cultural study comparing Islamic and non-Islamic customers.