Keywords

Introduction

Word of mouth (WOM) communication is a strategic marketing tool for building relationships with consumers, generating awareness and interest in products, and influencing consumers’ purchase behaviour (e.g., Chu & Kim, 2011; Lee et al., 2012). It has been defined as an “oral, person to person communication between a receiver and a communicator whom the receiver perceives as non-commercial, concerning a brand, a product or a service” (Arndt, 1967, p. 3). WOM can involve information and advice seeking when making a purchase (opinion-seeking) or the generation of information and advice by influencers, namely individuals who are able to affect the purchasing decisions of others through their opinions (opinion-giving).

As the world became digital, more and more people went online and started to exchange product information electronically (eWOM), thus influencing other peers’ preferences and experiences (Cheung & Thadani, 2010; Huang et al., 2011; Kietzmann & Canhoto, 2013; Ozuem et al., 2008). eWOM can be defined as “the positive or negative statement made by potential, actual, or former customers about a product or a company, which is made available to a multitude of people and institutions via the Internet” (Hennig-Thurau et al., 2004, p. 39). More precisely, there are three types of eWOM: opinion-seeking, opinion-giving and the sharing of third-party information (opinion-passing) (Flynn et al., 1996; Sun et al., 2006). These types of eWOM involve roles which do not have a clear distinction as each person can do all three; there is, though, one factor that is common to these three types, and that is of being based on user-generated content (UGC), namely on consumers’ online information generation, distribution and retrieval. As the source of information is perceived natural, genuine and honest, other consumers are led to consider its contents as trustworthy (e.g., Doh & Hwang, 2009; Hornik et al., 2015). Therefore, and similarly to WOM, eWOM emerges as a key driver in the buying process; it has a greater impact on customers’ purchasing decisions than other communication channels (e.g., Goldsmith & Horovitz, 2006; Lee et al., 2012). That is the reason why eWOM attracts the attention of scholars and practitioners in marketing; past literature has investigated the impact of eWOM on sales (e.g., Abubakar et al., 2017; Bulut & Karabulut, 2018; Goh et al., 2013; Gu et al., 2012; King et al., 2014; Zhu et al., 2020), the effect of positive or negative online comments/posts/reviews (e.g., Hornik et al., 2015; Hu & Kim, 2018; Yang et al., 2015), and the best strategy to induce consumers’ positive eWOM (e.g., Erkan & Evans, 2016; Reimer & Benkenstein, 2016; Yen & Tang, 2019).

In the plethora of Web 2.0 online communication channels, social networks (SNs) stand out because they enhance the information sharing process by allowing consumers to chat in real time with each other; for instance, through the creation of microblogging WOM that increases the speed of data exchange (Hennig-Thurau et al., 2015). The high levels of self-disclosure and social presence of SNs have enabled users to connect with other users by exchanging information, opinions and thoughts about products and brands (Chu & Kim, 2011). Accordingly, they are the perfect tool for eWOM as consumers freely create and share brand-related information in their favourite SNs composed by friends, classmates, colleagues and other acquaintances (Chu & Kim, 2011). This participation in online communities may positively or negatively impact on brand reputation/image and, thus, contribute to the process of branding co-creation (Kamboj et al., 2018). From their side, firms push to increase their presence on SNs (See-To & Ho, 2014) and develop online customer relationship management strategies aimed at engaging consumers and connecting them with brands (Azar et al., 2016; Chang et al., 2017). Among these, those that operate in the fashion sector have recognized the power of eWOM and turned towards marketing communication using social media in order to seize the opportunities of new communication models and survive the challenges of heated competition. This translates into a growing need to investigate consumers’ engagement in SNs communication during the product evaluation and purchase process.

Fashion products are particularly apt when studying social media usage as new style trends spread through network effects (Ananda et al., 2019; Easley & Kleinberg, 2010). When they are successfully adopted by a large number of people, they shape the perceived value of the product for other users, either positively or negatively. Moreover, fashion products are often used to build and communicate personal and group identities (Ahuvia, 2005; Wolny & Mueller, 2013). This feature, together with the fact that they can be very expensive, can lead to fashion products being classified as high-involvement goods. This has profound implications for peer-to-peer communications as it has been highlighted that high-involvement goods attract a significant amount of UGC and conversations online (Gu et al., 2012). Social media users often share style-related information with their peers with the expectations of receiving feedback on their stylistic choices and, in particular, on the social value of these choices (Lin et al., 2012). In light of this evidence, a better understanding of what motivates consumers to engage in social eWOM during fashion products’ evaluation process and how brands can encourage this engagement is undoubtedly of interest for both academics and practitioners. Although eWOM has received a lot of attention in the academic literature, a deep investigation into the influence of online products’ reviews through SNs on consumer’s decision-making processes is still needed. Through empirical research built on an online survey with a sample of 230 consumers, this chapter contributes to the literature on the spread of eWOM across SNs and its impact on purchase intention. More specifically, focusing on the fashion context, it investigates the effect of (a) involvement with SNs, (b) social cues, (c) accessibility and (d) informative value of reviews on SNs on social eWOM (opinion-seeking) and, contextually, the importance of eWOM in the pre-purchase decision.

Theoretical Framework and Conceptual Model

The study described in this chapter adopted the stimulus-organism-response (S-O-R) model to investigate the determinants of socialeWOM, focusing on opinion-seeking and its impact on the intention to buy the reviewed products. This model was developed by Mehrabian and Russell (1974) in the context of environmental psychology. Subsequently, it was applied in many areas of consumer behaviour with the aim of explaining the decision-making process (e.g., Chang et al., 2011; Chebat & Michon, 2003; Eroglu et al., 2001, 2003; Kang & Sohaib, 2015; Kim & Lennon, 2013; Ozuem et al., 2017; Rose et al., 2012). Some of the most recent applications are in the context of online consumer experience (e.g., Emir et al., 2016; Eroglu et al., 2003; Fang, 2014; Islam & Rahman, 2017; Kamboj et al., 2018; Mollen & Wilson, 2010; Qiao et al., 2019; Rose et al., 2012; Yan et al., 2018; Zhu et al., 2020). The S-O-R model postulates that Stimuli from the environment influence individuals’ internal reactions (Organism), which in turn lead to some behavioural Responses (Donovan & Rositer, 1982). With reference to consumers’ behaviour, the literature conceptualized stimuli as environmental inputs, including marketing mix variables (e.g., atmosphere, accessibility, social cues, customer service, information), which affect the attitudinal response. The organism element involves affective and cognitive reactions of individuals, which influence their final behaviour (e.g., Bagozzi, 1986; Bagozzi et al., 1999; Eroglu et al., 2001; Fiore, 2002; Frow & Payne, 2007; Zhang et al., 2014). It is usually operationalized in terms of perception, experience, evaluation and habits. The outcome in the S-O-R paradigm is the behavioural response, which can be classified as either approach or avoidance (Mehrabian & Russell, 1974). Approach behaviours include all positive actions that might be directed towards a particular setting (e.g., positive communications, intention to purchase or to act), whereas avoidance behaviours reflect the opposite responses, such as negative communications and no intention to purchase.

In order to suit the research objectives of the study, five antecedent variables were proposed as external stimuli (S) capable of influencing socialeWOM: (1) involvementwith SNs, (2) perceived accessibility of reviews, (3) informative value, (4) homophily and (5) social influence. The selection was made according to the relevant literature and to their expected relevance in the context under investigation. The habit of reading reviews of fashion products on SNs took the role of the organism dimension (O) in the S-O-R model. Meanwhile, the final response (R) is the intention to purchase the reviewed products.

The first focal antecedent is involvementwith SNs, measured in terms of time spent in reading and/or posting on SNs. Alhidari et al. (2015) found a significant effect of SNs involvement on consumers’ propensity to share their opinion on SNs (opinion-giving). Starting with this evidence, the research aimed to investigate the influence of such a variable on the opinion-seekingdimension of eWOM. A higher involvement with SNs should lead to greater familiarity with social environments and, therefore, should strengthen a consumer’s habit of reading fashion products reviews published by other users.

The second set of variables pertained to an individual’s evaluation of the accessibility and informativeness of other users’ reviews on SNs. Accessibility is the ease of using and understanding the use of SNs to collect information on fashion products, while informativeness represents the perceived value (convenience and usefulness) of reviews on SNs as a source of fashion products information. According to the technology acceptance model, the perceived ease of use and the perceived usefulness of a technology predict individuals’ attitude towards accepting it (Davis, 1989). Equally, it is supposed that the perceived ease of use and informativeness of reviews on SNs positively influence the degree of openness towards the reviews and the willingness to read them.

The last set of variables pertains to social cues, measured in terms of homophily and normative social influence. Homophily is defined as the degree to which individuals who interact with one another are congruent or similar in certain attributes (Rogers & Bhowmik, 1970), while normative social influence refers to “the influence to conform to the expectations of another person or group” (Deutsch & Gerard, 1955, p. 629). Prior research has suggested that homophiles tend to share information with one another (e.g., Rogers & Bhowmik, 1970). However, literature on social media has produced mixed results. Mainolfi and Vergura (2019) found a positive effect of homophily on opinion-giving through fashion blogs, while Kim et al. (2018) showed that homophily significantly influences attitude towards eWOM information. By contrast, Chu and Kim (2011) highlighted that information deriving from a socially similar source decreases the degree of involvement with eWOM, for both opinion-seeking and opinion-passing. Whereas susceptibility to social influence was found to have a positive impact on all the three dimensions of eWOM (Chu & Kim, 2011). In order to shed light on the relationship pathways between these variables with reference to fashion products, the present study aimed to test the effect of perceived homophily with SNs’ contacts and normative social influence on social eWOM adoption.

The last relation investigated is that between the organism dimension (habit of reading product reviews on SNs) and the behavioural response (intention to purchase the suggested products) conceptualized in the S-O-R model. Torres et al. (2018) found a significant effect of acceptance of eWOM information on consumers’ purchase intention, while Alhidari et al. (2015) highlighted that consumers’ propensity to share their opinion on SNs is positively related to the intention to purchase products reviewed on SNs. Similarly, Vahdati and Mousavi Nejad (2016) and López and Sicilia (2014) confirmed that eWOM, defined as opinion-seeking and opinion-giving, had a positive effect on purchase intention. In light of this evidence, a significant impact of eWOM adoption on fashion products purchase intention has been assumed.

The proposed structural model is shown in Fig. 10.1.

Fig. 10.1
A horizontal model diagram depicts that 5 components of stimuli are connected with social e W O M and then to purchase intention under response.

(eWOM Electronic word of mouth)

Conceptual model

Full size image

Empirical Research

Data Collection

An online self-administered questionnaire was filled in by a sample of 230 Italian people. All participants were informed that the study was on a voluntary basis and that information provided would be kept confidential. The respondents were first asked about their SNsusage (typeand involvement), followed by homophily, perceived ease of use, perceived usefulness and engagement in socialeWOM, which were operationalized as opinion seeking, normative social influence and purchase intentions. Finally, demographic information was collected. The items of the questionnaire were adapted from previous research, with some amendments made to fit the context of the present research.

Involvement was assessed using the Alhidari et al. (2015) 7-item scale (see Table 10.1). Homophily was measured through the four items proposed by Kusumasondjaja (2015), while the 8-item scale by Bearden et al. (1989) was used for detecting normative social influence. The items for perceived ease of use were adapted from Glover and Benbasat (2010). The concept of informativeness was assessed using the three items proposed by Taylor et al. (2011). The susceptibility to online product reviews scale by Bambauer-Sachse and Mangold (2011) was used for the measurement of social eWOM adoption (opinion-seeking). Finally, the scale for purchase intentions was derived from Mikalef et al. (2013). All items were measured on a 7-point anchored scale (from “completely disagree” to “completely agree”).

Table 10.1 Measurement scales and reliability indices
Full size table

Structural equation modelling with maximum likelihood method was employed for the analysis of the measurement model and of the conceptual model. Data analysis was performed using the IBM SPSS statistical software (SPSS Inc, Chicago, IL; release 25.0) and the LISREL software (release 8.80).

Sample Characteristics

The sample was represented by 70% women and 30% men, with a mean age of 32 (min = 18; max = 63). The respondents were well-educated: 54% had graduated or post-graduated and 38% completed high school; the remaining 8% had left school after the primary or secondary level. Out of the sample, 64% were single, 31% were married or cohabiting and 5% were widowed or divorced. The three most used SNs were Instagram, Facebook and YouTube, followed by Twitter and LinkedIn.

Research Results

As the skew and kurtosis statistics showed that the normality assumption was violated (χ2 = 2533.935, p < 0.001), the model was estimated using the Satorra–Bentler method (Satorra & Bentler, 1994). The fit indices indicated an acceptable overall fit of the measurement model to the data: Satorra–Bentler scaled χ2 = 942.160, df = 539, p = 0.000, comparative fit index (CFI) = 0.986, root mean square error of approximation (RMSEA) = 0.057, non-normed fit index (NNFI) = 0.984 and standardized root mean square residual (SRMR) = 0.049.

Convergent and discriminant validity were evaluated through the strength and significance of the loadings, the composite reliability (CR), the average variance extracted (AVE) and the Cronbach’s alpha (Bagozzi & Heatherton, 1994; Cronbach, 1951). All items loaded strongly and significantly on the hypothesized latent variables, ranging from 0.671 to 0.931. All constructs exceeded the recommended cut-off points for the adequacy of 0.70 for CR (Steenkamp & Van Trijp, 1991) and 0.50 for AVE (Fornell & Larcker, 1981). Finally, the data met Fornell and Larcker’s (1981) criterion: the average variance explained by each latent variable was greater than any of the squared correlations involving the variable, suggesting that discriminant validity was achieved. Cronbach’s alphas were also used to confirm the scales’ internal consistency. The index was very high for each construct, ranging from 0.92 to 0.97.

The results indicated an acceptable fit for the proposed model (Satorra–Bentler scaled χ2 = 950.853, df = 544, p = 0.000, CFI = 0.985, RMSEA = 0.057, NNFI = 0.984 and SRMR = 0.051). The model explained 57% of variance for socialeWOM and 70% for purchase intention. The significant parameters estimates are reported in Fig. 10.2. The analysis of the path coefficients showed that accessibility of fashion products’ reviews exerted a significant influence on social eWOM: a higher perceived ease of use of reviews on SNs translates to a greater habit of reading reviews of fashion products (β = 0.397, p < 0.05). By contrast, involvement with SNs and informativeness of fashion products reviews did not have a significant impact on social eWOM adoption. Turning to social cues, both homophilyand normative social influence significantly increased the habit of opinion-seeking (β = 0.188, p < 0.05; β = 0.307, p < 0.01). Finally, a strong relationship emerged between social eWOM and purchase intention (β = 0.836, p < 0.01). Reading fashion products reviews on SNs positively influences a consumer’s decision-making process, increasing the intention to purchase those products.

Fig. 10.2
A horizontal model diagram depicts that 5 components of stimuli are connected with social e W O M and then to purchase intention under response.

(eWOM Electronic word of mouth, ns not significant, *p <0.05, **p <0.01)

Structural model with standardized coefficients

Full size image

Discussion and Implications

The major aim of the study was to investigate the factors that can predict consumers’ engagement in socialeWOM, defined as opinion-seeking, and the impact of engaging in social eWOM on the intention to buy the reviewed product, focusing on fashion products. In the face of the increasing connectivity among SNs users, social eWOM—that is the sharing of content regarding brands/products/venues via online SNs—has also grown. Its pervasiveness and capability to affect users’ perceptions of companies and of their products make it a key driver in the buying decision process. Accordingly, both academics and practitioners are interested in exploring consumers’ engagement in social eWOM and understanding how to encourage the spread and influence of eWOM.

The research goals were achieved by adopting the S-O-R framework. Results showed that accessibility of reviews on SNs, perceived similarity with the SNs’ contacts and susceptibility to social influence positively impact social eWOM adoption. By contrast, involvement with SNs and the informative value of reviews do not translate to a greater habit of reading reviews of fashion products. Finally, the stronger this habit, the greater the intention to purchase the reviewed products.

The study enriches the literature on online products’ reviews and provides companies some guidance for the understanding of the role of social eWOM in influencing consumer behaviour.

At the theoretical level, it demonstrates that the S-O-R model is an adequate framework to investigate the decision-making process in the context of socialeWOM. Specifically, social cues and perceived ease of use of reviews on SNs represent the environmental inputs that affect the consumers’ involvement in opinion-seeking, which in turn influences the intention to purchase the reviewed fashion products.

From a managerial perspective, understanding the role of social eWOM in the consumer–product relationship helps companies to effectively incorporate SNs as an integral and significant part of their marketing communication mix. This is particularly relevant in the fashion industry because peer influence is of great importance. Market trends are created less by established fashion magazines or designers and more by opinion formers who have the power to shape the perception of brands’ image and value (Ozuem et al., 2016; Wolny & Mueller, 2013). The findings of the study encourage practitioners to take into consideration the social relationshipvariables that affect consumers’ eWOM behaviours. Community-based ties play a decisive role in creating a persuasive process driven by homophilyand normativepeer-to-peer influence. The tendencies to be connected to other SN users and to seek social approval appear as significant influencing factors within the process of opinion-seeking and creating purchase intentions. To take advantage of this influence path, companies should employ analytics data to select the more powerful reviews according to the similarity between opinion-giver and opinion-seeker profiles. In this perspective, offering users the ability to autonomously filter reviews according to their preferred parameters would enable the achievement of more effective results. The ease of use of SN channels also stands out as important in the propensity to read reviews. In this perspective, anything that simplifies the move from reading the product review on a social media page to purchasing it on the sales website or through shoppable posts is fit for purpose. By contrast, involvementin SNs does not emerge as a relevant driver in improving consumers’ propensity towards social eWOM, at least with reference to the opinion-seeking dimension. This means that familiarity with SNs is not important in persuading users to seek and rely on non-commercial communication in SNs. Even individuals who do not spend much time in posting and updating on SNs have the habit of reading product reviews on this type of social media. This speaks volumes about the current importance of SNs as a source of product and services information. Through eWOM, brands can reach a very large sample of consumers made up of regular and non-regular users of SNs who can both be effectively influenced by other users’ content.

Conclusion

In recent years, SNs have gained notable popularity in consumers’ information searches and subsequent purchase decisions. From their side, companies and brands have quickly embraced this communication media in order to reap the benefits of direct engagement with customers and peer influence. Indeed, social media platforms are one of the main online channels through which users exchange information and opinions about products and brands. This made social eWOM a key driver in the consumer decision-making process, which can influence products’ and brands’ image, reputation and equity (e.g., Casaló et al., 2007; Chae & Ko, 2016; Gummerus et al., 2012; Kamboj et al., 2018).

This study aimed at investigating consumers’ engagement in social eWOM—measured as search for information—on fashion products. These products were chosen because their consumption is influenced by symbols and images, and often serves to communicate personal and group identities (Ahuvia, 2005; Altuna et al., 2013; Wolny & Mueller, 2013). As SNs are vehicles for self-expression, they are appropriate tools for communicating information about the fashion shopping experience and, in this way, affirming identity and social belonging.

The results confirmed the basic role of eWOM in influencing the purchasing decision-making process and highlighted two main drivers of consumers’ habit of reading fashion products reviews on SNs: accessibility and social cues. This means that marketers who want to encourage consumers’ engagement in social interaction and induce positive eWOM have to take into consideration the key role of perceived similarity among users and of the seeking of social approval. At the same time, the easy accessibility of reviews is equally relevant. This is an important aspect not only for social networking service providers, who should ensure the ease and understandability of the use of SNs to collect product information, but also for companies and brands, which can facilitate the path from reading the product review to purchasing.

The results of this study are a stepping stone towards future research. Although eWOM emerged as a key resource in influencing and forming behavioural intentions, future research could investigate whether familiarity/involvement with the product/brand might influence the persuasion capability of eWOM and moderate the effect of the stimuli. Moreover, the study focused on the recipient perspective (opinion-seeking); however, analysis of the information sender perspective is also valuable (opinion-spreading). Finally, a comparison between different SNs would be opportune in order to explore any differences in the peer influence dynamics.