The influence of homophily and social identity on eWOM in streaming consumption behaviour

Abstract

Purpose

The purpose of this paper is to identify the influence of homophily and social identity in eWOM adoption and influence on music consumption in streaming platforms. The study also proposes a framework for eWOM influence on product consumption.

Design/methodology/approach

Data were collected through an electronic questionnaire with a total of 524 valid responses. A partial least square methodology was used to approach the data analysis.

Findings

The findings indicate that both homophily and social identity have an influence on eWOM effect. Additionally, homophily shows influence on social identity being an important value to the total influence. It is also confirmed how eWOM has direct influence on music consumption.

Practical implications

Social information of consumers becomes more relevant than ever, as the results contribute to highlight its importance as a message adoption influencer and product consumption driver.

Originality/value

This study contributes to social communication theory by identifying homophily and social identity as valuable assets during eWOM generation. The research brings a new angle to social identity theory in the social network environment and influences eWOM adoption in music streaming consumption.

Objetivo

El propósito de este estudio es identificar la influencia de la homofilia y la identidad social en la adopción de eWOM y su influencia en el consumo de música en plataformas de streaming. El estudio también propone un marco para la influencia de la eWOM en el consumo de productos.

Metodología

Los datos se recopilaron a través de un cuestionario electrónico con un total de 524 respuestas válidas. Se utilizó una metodología de Mínimos Cuadrados Parciales para abordar el análisis de datos.

Resultados

Los hallazgos indican que tanto la homofilia como la identidad social tienen una influencia en el efecto de la eWOM. Además, la homofilia muestra influencia en la identidad social siendo un valor importante para la influencia total. También se confirma cómo la eWOM tiene influencia directa en el consumo de música.

Originalidad

Este estudio contribuye a la teoría de la comunicación social identificando la homofilia y la identidad social como activos valiosos durante la generación de eWOM. La investigación aporta un nuevo enfoque a la teoría de la identidad social en el entorno de las redes sociales e influye en la adopción de eWOM en el consumo de música en streaming.

Implicaciones prácticas

La información social de los consumidores se vuelve más relevante que nunca, ya que los resultados contribuyen a resaltar su importancia como influenciador en la adopción de mensajes y como impulsor del consumo de productos.

目的

本文旨在确定同质性和社会认同对电子口碑在音乐流媒体平台上的采纳的影响。研究还提出了一个关于eWOM对产品消费影响的框架。

方法论

通过电子问卷收集了524份有效回复, 并采用了偏最小二乘法对数据进行分析。

发现

研究结果显示, 同质性和社会认同对eWOM效果产生影响。此外, 同质性对社会认同产生影响, 对总体影响具有重要价值。另外, 研究还确认了eWOM对音乐消费的直接影响。

创新性

本研究通过确认同质性和社会认同在eWOM生成过程中的重要价值, 为社交传播理论做出了重要贡献。研究为社交网络环境下的社会认同理论提供了新的视角,并对音乐流媒体消费中的eWOM采纳产生了影响。

实践意义

在当前社会环境下, 消费者的社会信息比以往任何时候都更加重要。本研究的结果有助于凸显社会信息在信息采纳和产品消费方面的重要性, 进一步强调了其作为影响因素和驱动力的重要性。

Citation

Oquiñena, I., Sánchez, J. and Monfort, A. (2024), "The influence of homophily and social identity on eWOM in streaming consumption behaviour", Spanish Journal of Marketing - ESIC, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/SJME-05-2023-0131

1. Introduction

Word-of-mouth communication, also known as WOM or its electronic version eWOM, has been identified as one of the most influential sources of consumer decision-making (Xiao et al., 2022). eWOM is defined as any kind of communication exchanged between consumers of a good that is available through the internet (Rosario et al., 2020). People spend an increasing amount of time searching for information or communicating with other customers to share their experiences and help the purchase process (Dwivedi et al., 2020). The implementation of eWOM in retail platforms has become a necessity in many cases, being a preferred source of information for consumers. According to a 2022 Nielsen media marketing report, 89% of consumers trust information received by other users, even if they do not know them, more than information received by the company. In fact, in Spain, 80% of internet users report consulting opinions from other online shoppers during the purchase decision process (AIMC, 2021).

eWOM has actually become an important phenomenon in the digital age, as it has the potential to influence consumer behaviour and purchase decisions (Liu et al., 2022). The academic importance of eWOM lies in its potential to provide insights into consumer behaviour and attitudes, as well as offer new avenues for studying the impact of marketing and communication strategies on consumer perceptions and decision-making (Moradi and Zihagh, 2022). Research on eWOM has contributed to a deeper understanding of the social and psychological factors that influence online communication, such as trust, credibility, usefulness and attitude towards product consumption (Verma et al., 2023). From a practical perspective, eWOM has become a crucial element of marketing and branding strategies for businesses, as it can help build brand awareness, generate buzz and foster customer loyalty. Positive eWOM can enhance a brand’s reputation and attract new customers, while negative eWOM can damage a brand’s image and lead to lost sales (Purnawirawan et al., 2015). Therefore, understanding the factors that influence eWOM can help businesses develop effective strategies for managing their online reputation and improving customer satisfaction. Overall, eWOM is a complex and dynamic phenomenon that has important implications for both academic research and practical applications. As online communication continues to grow in importance and influence, studying eWOM will remain a critical area of research for understanding consumer behaviour and informing marketing and communication strategies.

Most industries have had to change their business model to adapt to upcoming digital needs such as tourism which totally adapted selling platforms integrating social media (Xu and Lee, 2020) or hospitality where customers eWOM has become one of the main promotion tool (Ransbotham et al., 2019). There is multiple evidence on the importance of eWOM in culture, and much of the research has focused on analysing the influence of consumer opinion (recommendations, reviews, etc.) and its effect on the consumption of cultural goods. Some researchers such as Godes and Mayzlin (2004) have demonstrated the effect of eWOM on television consumption; others such as Chevalier and Mayzlin (2006) have focused on book sales and there is also much evidence on cinema attendance such as Basuroy et al. (2020) or Yu et al. (2019).

The music industry is not exception being recently affected by a change of consumption related to digital consumption and social media interaction (Aguiar and Waldfogel, 2018a, 2018b). Consumer preferences have evolved in line with changes in technology, with music culture being particularly affected both by an increase in its offer and by opening up to new segments Peukert, 2019). The music business has also adapted from an offer mainly based on albums sales to subscription streaming models that give access to complete collections of musical content (Bula Tello, 2021). Thus, the consumption influence analysis has been changed to increase social exhibition where eWOM influence play a new role into his consumption. Moreover, due to its intangible and hedonist nature, analysis of the music industry is indispensable to understand how others’ opinion’s influence expectation building. It should be also taken in account that those variables related with cultural identification as homophily or social identity should be especially important to asses in eWOM impact due its hedonist cultural environment, as included in other studies such as Grappi and Montanari (2011). In this regard, the purpose of this paper is to use the music streaming market as an example to analyse how social variables can influence eWOM message adoption and product consumption.

Considering the three main axes of the communication process: the communicator, the receiver, and the message widely used by previous studies (i.e. Babić Rosario et al., 2016; Erkan and Evans, 2016; Cheung and Thadani, 2012), this research aims to understand which variables could affect eWOM consumption and its influence. Taking into account how subjective is the consumption of hedonist and intangible products as music the variables selected concern social perception such as homophily and social identity (Aguiar and Waldfogel, 2018b). Social variables have been related to eWOM consumption in many ways, such as sharing common aspects (Nofal et al., 2022) or belonging to similar social groups (Hogg, 2016). However, its effect and differences have not been properly studied. Moreover, regarding the message, the literature has demonstrated that the information created by consumers gives more credibility and strengthens the purchase decision process (Chatterjee, 2001). Thus, the importance of eWOM on the influence of product consumption has been shown (Bu et al., 2022; Flavián et al., 2016) as well as its relevance in several industries related to cultural products (Basuroy et al., 2020; Chen and Lei, 2021).

The study is built around the social communication theory (Hovland, 1948) which has been the main framework for most eWOM studies of the literature review establishing the main influence variables during eWOM adoption and influence. The receptor influence pivots around the homophily perception and the social identity theory (Tajfel, 1974) which have been studied to understand common traits and sense of belonging effect in eWOM message communication.

The paper is structured as follows. Firstly, in 2. Literature review and hypothesis development we present the communication theoretical framework upon which the study is based and analysed, developing the different hypotheses according to the principle of communicator, receiver and message. In 3. Methodology and 4. Results the methodology, sample and results are explained. Finally, in 5. Discussion the paper develops the main theoretical and practical implications derived from the results obtained.

2. Literature review and hypothesis development

2.1 Communication theorical framework

eWOM has a dual effect in its information exchange process, so that each interaction will have a reaction on the part of the other interlocutor. Consequently, its analysis must move from information exchange to information exposure and finally to its assessment (Rosario et al., 2020). The analysis of the theoretical framework of eWOM has generally focused on either the effect on the consumer or on the receiver (Kannan, 2017). However, their approach is generally biased as the effect of eWOM will not only be affected by the communicator itself but by the context or receiver (Cheung and Thadani, 2012). Additionally, eWOM consumers are themselves users exposed to the same type of information and in many cases generators of the same type of content (Rosario et al., 2020). Therefore, to have a complete model of the information exchange process and its effect on influence, a comprehensive approach that integrates the key factors influencing communication is necessary.

The social communication framework (Hovland, 1948) establishes the analysis of the communicator, the receiver and the message as a common research model in conversation studies. eWOM literature has used the social communication framework in several theorical models such as those from Babić Rosario et al. (2016) or Erkan and Evans (2016) which are formulated around the relationship of sender and receiver and the factors that influence the exchange of information between them; also those from You et al. (2015) and Cheng and Ho (2015) which focus the framework on stimuli influence and its role on the communication process.

Most eWOM theorical frameworks are then based on the communication axes that integrate the message. However, the model that would best adapt Hovland’s theory would be the one proposed by Cheung and Thadani (2012) which compiles evidence of eWOM around five axes identified as key in the communication process. The axes identified are the communicator, the receiver, the stimuli, the response and finally the contextual factor that interact during the whole process. The model proposed by Cheung and Thadani (2012) has been used as a reference to a complete analysis of eWOM communication and emulated in current studies and reviews such as the eWOM review by Kusawat and Teerakapibal (2022). The current study will develop a theorical framework around eWOM influence on consumption to answer the current gaps considering the three main axes of every communication process: the communicator, the receiver and the message. The situation of eWOM influence addresses an assessment challenge that drives the need to correctly frame the variables that influence within digital platforms such as music streaming consumption. The consumption nowadays is more focuses on information filtering and social content identification (Aguiar and Waldfogel, 2018b). Consequently, the variables that could be more important to assess will be those associated with risk reduction (homophily), cultural identity (social identity) or eWOM message impact (Currarini and Mengel, 2016). The current study will develop a theorical framework around eWOM influence to answer the current gap around social communication taking in account the three main axis of every communication process: the communicator, the receiver and the message (Hovland, 1948).

2.2 The communicator: homophily

Homophily is defined as the similarity of characteristics shared by the communicator and the receiver (Bu et al., 2022). eWOM communication in a homophily situation means that communication and interactions between similar people will be more frequent than between more different people (Sarker et al., 2022). Connections between people will affect multiple casuistry, considering demographic factors such as age, gender, region or social factors such as status, lifestyle, profession or culture. Therefore, a perception of common traits can potentially lead to greater message adoption and more intense interaction between subjects. Homophily influence has been challenged many times on eWOM. Brown et al. (2007) affirmed that homophily and social tie differs in online communication from offline. The online communication raises the doubt that an open context of social interaction could dismiss importance of homophily. The research of Ismagilova, Slade, Rana and Dwivedi (2020) states that homophily indeed is one of the less important variables on consumption influence its role being somehow less relevant in certain circumstances. Additionally, Babić Rosario et al. (2016) identified a similar situation where homophily influence showed variability depending on the eWOM platform, suggesting that those platforms with more intangible and experiential products could be more important to eWOM influence based on homophily. Thus, on streaming platforms where access to the catalogue is extensive, the main existing limitation is time and its consequent opportunity cost. Consequently, the identification of users with similar preferences can help the information provided by eWOM to serve as a filter and affect message diffusion (Furutani et al., 2023). Then, users with similar consumption patterns can probably be one of the most efficient ways to ensure satisfaction in the selection of cultural goods (Hagen and Lüders, 2017). The connection of similar cultural tastes generates a social relationship according to the perceived experience of the receiver, implying a bond of union that will lead to a greater adoption of the message (Chih et al., 2020).

Therefore, it could be defined that the perception of similarity between users, also called homophily, could be one of the main influencing factors in the influence of eWOM. The main motives identified would mention both the reduction of risk and facilitation of the consumption decision process, as well as a tool to identify a potential greater trend of interactivity between users. For previous contributions and considering the importance of homophily on users’ engagement in communication process, H1 establishes this relation:

H1.

The perception of homophily on the communicator will lead to a greater interest and adoption of the generated eWOM.

2.3 The receiver: social identity

Intangible goods are often associated with the consumers identity (Prior, 2013); therefore, in most eWOM influence, it is necessary to identify the social identity influence embedded in the receiver to understand its role during the communication process. While homophily would explain the behaviour associated with the perception of similarity between individuals, social identity would go a step further associated with a feeling of belonging or aspiration to a particular social group (Hogg, 2016).

Social identity theory has been an important contributor in social behaviour and interrelation the work of Tajfel (1974) has become a common framework in current social and psychological investigations related to eWOM and social networks (Rodgers and Rousseau, 2022). Its application in social projections and cultural value products is related to those elements associated with the user’s personality and individual culture. In this paper the theory is relevant as a cultural good appreciation such as music is biased by cultural education and will create bonds of social identity between different groups. Therefore, differences in music consumption could be perceived as a social distinction and therefore be an important barrier between consumers identification.

Musical taste is a remarkable barometer in social processes and which has a special incidence in the construction of our social identities (Prior, 2013). While homophily would explain the behaviour associated with the perception of similarity between individuals, social identity would go a step further associated with a feeling of belonging or aspiration to a particular social group (Hogg, 2016). The distinctions separating homophily and social identity are not always clear; there are several studies such as those from Froehlich et al. (2023) or Evans et al. (2023) that address social identity by the perception of similar traits among users, therefore a null differentiation with the homophily effect. However, there is some consensus identifying potential explanatory factors such as cultural or group-derived influence that would have an effect independent of homophily.

Social identity will be therefore a sentiment associated with the message receiver and a potential influencer on message adoption. Identity is usually related to those elements associated with the user’s personality and individual culture. Appreciation of a cultural good such as music is biased by cultural education and will create bonds of social identity between different groups (Bourdieu, 1984). Therefore, differences in consumption could be perceived as a social distinction and therefore be an important barrier between consumers identification.

Consequently, for previous contributions and considering the importance of similar traits perception, H2 establishes this relation:

H2.

The perception of homophily among users will positively influence the creation of a sense of belonging and social identity with the group.

The observation of differential behaviour between social groups will affect the receiver by encouraging the adoption of messages from those interactions that intensify their involvement in the social group to which they belong. The behaviour of one group can potentially create an influence on other users according to the information provided differentially from other groups. That is, the generation of differentiating messages with respect to other groups will be conducive to generating a potential effect on consumer behaviour to gain acceptance of the group (Bagozzi and Dholakia, 2002).

Therefore, social identity theory identifies a tendency to adhere to perceived social norms of the reference group that would reinforce such behaviour patterns (Hogg, 2016). The feeling of reward is preceded by a need for self-identification and to feel accepted within the individual's aspirational social circle just as brands can satisfy or at least benefit from such a desire (Chu et al., 2019).

The influence of social groups may potentially be greater at stages where social identity is still developing, such as in adolescence (Hoffman and Umaña‐Taylor, 2023). The imitation of behaviours and influence of messages is associated with a need to form a social identity associated with those social groups to which they aspire to belong (Casaló et al., 2020; Chan, 2008). On the other hand, social learning theories suggest that individuals' behaviour is also associated with social reward or punishment derived from their peers (Bandura, 1986).

Social identity will be based on value driven guidance for eWOM influence and therefore an important trigger to consumption. However elaboration of the likelihood model research has identified that depending on the product importance a peripheral route such as platform or message aspect could displace social identity influence (Wang et al., 2022). Researchers differs on product and culture differences to confirm identity influence (Cheng and Ho, 2015). However, at the end of the day the message receiver implication is subjective and in an experiential product, such as music, this could be important to understand the importance of social identity influence.

In short, social identity involves the confirmation of the normative influence of user behaviour in accordance with social identification with a group of belonging. In a context of maximum visibility of cultural behaviour currently exposed through social media platforms, eWOM and online interactions should be studied according to potential associations with social identity (Aguiar and Waldfogel, 2018a).

The congruity theory, which also supports behaviour influence on individuals based on the common trend of perception, is also relevant (Osgood and Tannenbaum, 1955). The congruity research has been relevant in many recent web 2.0 studies where eWOM message influence and valence has been explained through the desire to maintain balance between their cognition (Walther et al., 2012). Social identity therefore will be an important factor that would affect consumers behaviour to keep congruity between their ideas and the established way of thinking of their group of belonging.

Therefore, for previous contributions and considering the importance of social influence on user interaction, H3 establishes this relation:

H3.

The receiver's feeling of social identity with the group belonging to the communicator will lead to a greater adoption of eWOM.

2.4 The message: electronic word-of-mouth influence and response

eWOM has played a key role in virtual businesses from the beginning, as consumers on the internet cannot interact with the products they are interested in as they can when purchasing them in a physical shop. To overcome this barrier, sales websites try to offer complementary information that reduces the risk of purchase, such as comments or evaluations from other consumers (Romero and Ruiz-Equihua, 2020). In this way, the information created by consumers facilitates the purchase decision process as it provides more credibility, facilitating comparison and usefulness to the end consumer (Chatterjee, 2001).

Recent researches have highlighted the importance of eWOM on the influence of product consumption (Bu et al., 2022; Ruiz-Equihua et al., 2021), the relevance has been shown in several industries related to cultural products such as music festival (Chen and Lei, 2021) or cinema attendance (Basuroy et al., 2020). However, there is no specific analysis concerning eWOM on streaming platforms and its influence on consumption.

The importance of the impact of eWOM in a pure electronic consumption, such as music streaming platforms could be especially important as the decision-making process and consumption are instantaneous and driven by the social platform itself. Furthermore, the use of eWOM has become more popular; its impact has become especially relevant in sectors whose goods are experiential and therefore difficult to value except through other experiences. Studies by Zinko et al. (2021) and Cheng and Ho (2015) highlight the relevance and particularities of eWOM in sectors where it is difficult to compare goods and where the experience of other users is attractive when it comes to facilitating the decision-making process.

In brief, experiential goods such as music are associated with an intrinsic difficulty in knowing the product information in depth until it has been experienced, which is why the dissemination of a message with previous experiences such as eWOM is a very important tool in its marketing (Xu and Lee, 2020). Consequently, for previous contributions and considering the importance of eWOM influence in music consumption, H4 establishes this relation:

H4.

The influence of eWOM will be positively related to product consumption.

In summary, the theoretical model that we propose regarding the influencing variables on consumption is reflected in Figure 1.

The proposed model would capture the variables of homophily and social identification that would influence both the communicator and receiver in the influence of eWOM and ultimately its impact on product consumption.

3. Methodology

3.1 Sample and data

The research was mainly based on the music streaming consumption behaviour of the Spanish population, the streaming platforms currently being a common well-known consumer good and used all over the country (Kantar, 2022).

A previously pilot test has been performed with 30 students to test the questionnaire and once it was tested a total number of 524 valid responses were obtained, in the final questionnaire research (Appendix). The distribution of the study population according to sex showed that, 53.4% declared themselves to be women and 45.2% men; the remaining 1.2% either did not answer the question or indicated another variable. The average age is 35.7 years, with a generational distribution of 47.1% Millennials (23–41 years old), followed by 27.3% Generation X (42–56 years old), 16.4% Generation Z (22 years old or younger) and 9.2% Baby Boomers (57 years old and older) (Reeves and Oh, 2008).

72.9% of respondents claim to be subscribed with paid accounts with music streaming platforms, the majority also being subscribers to TV streaming platforms. The average consumption of music on streaming platforms is 1.9 h per day. Therefore, the data confirms the penetration of streaming on Spain. All the descriptive information could be checked in Table 1.

Finally, we collected data about information search interests about music, movies and TV: 60% of respondents actively search for information, social media being the most important source for them. Consequently, it was confirmed that the sample takes especial interest in product information searching on social platforms.

3.2 Variable measure

This study uses the scale of Faraji-Rad et al. (2015) to measure homophily perception where the similarity identification was analysed to understand message influence on communication. Social identity will be measured using the scale of Trudel et al. (2016) which was used to assess the importance of a product's link to its social identity and its effect on its consumption or disposal.

Finally eWOM influence is measured using the scales of Wu et al. (2017) and Nishikawa et al. (2017) where information interest is analysed; eWOM influence was measured using the scales of Mao (2016) and Chang (2017) where message influence was assessed in consumption influence. All the constructs measurement results could be checked in Table 2.

3.3 Partial least square methodology

A partial least square methodology has been used to approach the data analysis for several reasons. Firstly, the choice of theoretical variables that are not directly observable, such as homophily and social identity needs a method flexible enough to work with (Hair et al., 2021). Secondly, it was necessary to validate multiple relationships between variables, then the complexity of the model needed a strong validation model, such as Smart PLS (Roldán and Sánchez-Franco, 2012). Thirdly, the smart PLS analysis allows a proper assessment even if the sample quantity is reduced. Consequently, following the reasons already mentioned, the use of Smart PLS software suited on the research goals (Roldán and Sánchez-Franco, 2012). CB-SEM methodology has also been considered to assess the structural relationship but as it would be addressed a composite factor model it was preferred to use PLS-SEM (Dash and Paul, 2021).

The data analysis has two steps: the fit model, and the structural model. The fit model will be performed to measure how reliable and valid is the theorical model proposed. The structural model will be an assessment of the variable’s correlations and its statistical significance.

3.4 Measurement model

The fit model tests the reliability and validity of the data. The first step was to validate the reliability of the scales and as can be seen in Table 3 the reliability of all constructs is high, exceeding the limit of 0.7 for both Cronbach's alpha and composite reliability (Henseler et al., 2016). Therefore, it could be stated that all the scales represented have a high consistency in their measurement.

The second step is to confirm the measurement validity then average variance extracted (AVE) confirms the convergent validity as all the results are higher than 0.5 and the HTMT ratio confirms a discriminant validity as every intersection has a value lower than 0.85 (Henseler et al., 2015).

4. Results

Initially, no multicollinearity is identified, given that the estimation of the coefficients is based on regressions, as in a multiple regression, the presence of multicollinearity between the antecedent variables of each of the endogenous constructs must be avoided (Hair et al., 2021). Moreover, VIF assessment shows positive results in all the variables so common bias method (CBM) is checked for full collinearity (Kock, 2015).

The path coefficients result in Figure 2 states that the correlation between the variables identified on the theorical framework confirms the hypothesis variable relationships as all being positively related.

The analysis followed the bootstrapping method; 1,000 bootstraps samples ran at 95% confidence interval. The results in Table 4 offer that all p-values are significant, so the hypothesis stated are all confirmed.

Therefore, every hypothesis has been supported. H1 related to homophily influence on eWOM effect has a path coefficient of 0.274 and a p < 0.0001 confirming that mutual characteristics perception would help increase interest in eWOM communication. H2 and H3 are also supported with a path coefficient of 0.574 and 0.371 and a p < 0.0001, respectively, being possible not only that social identification has influence on eWOM effect but also that homophily perception would help to increase social identification. Finally, H4 is also supported with a path coefficient of 0.787 and a p < 0.0001 being the highest path coefficient confirming eWOM importance on music streaming consumption.

The R-square value (variance accounted for) for the dependant constructs, has to exceed 0.1. Figure 2 shows how every dependent variable meets the requirement, eWOM influence being 0.619 which indicates that more than 60% of the variance of the influence to product consumption is explained by the effect of eWOM (Hair et al., 2021). This result reaffirms the robustness of eWOM influence on music consumption and highlights the importance of eWOM on music streaming industry.

5. Discussion

This paper shows the influence of eWOM on a digital market such as music streaming consumption, solving gaps of communicator and receiver influence, highlighting the importance of homophily and social identification during the communication process. It is also presents a theorical communication framework of eWOM influence that would help to better assess how social variables could affect the message adoption and consumption influence. Through the framework, eWOM message adoption is confirmed as playing an important role in product consumption influence, with eWOM platforms being an important tool to better understand the music consumption framework. For instance, given the development and reach of social eWOM networks such as TikTok or Instagram where music is highly associated with images and video, the analysis of social media nowadays is quite relevant to understand current cultural consumption. We can reflect on how social media work and how those interactions could lead in future to changes in consumption and decision-making processes in many industries, as they have in the music streaming industry. In its framework of analysis, it is possible to identify how social variables, such as homophily and social identity play an important role with social media being an important tool to be analysed in purchasing behaviour.

Social media is currently well established in most of our current social interactions, being a key communication tool to keep relationship with our contacts. Moreover digital social engagement has been tested as quite influential and efficient with digital natives (Munsch, 2021). In short, social networks have become an instrument that is highly associated with lifestyle and consumption (Dwivedi et al., 2020). However, the effect social contacts potentially in new generations whose personalities are developing could be more associated with social identity and be more susceptible to influence (Ruiz-Equihua et al., 2021; Southgate, 2017). Most industries as with music streaming have recently changed to no longer facing a problem of access to information, but of information filtering; therefore, eWOM is starting to be integrated in social media platforms as a filtering tool to facilitate the decision process.

5.1 Theorical contribution

The main contribution of this study is the confirmation of eWOM influence in the music business through the test of two social variables namely homophily and social identity into the same model in a novel context of streaming platforms. Results from the PLS structural model test shows first how eWOM influence in music has a high influence on product consumption as was previously confirmed on other industries (Tariyal et al., 2022; Yu et al., 2019). Furthermore, how homophily is a variable that not only affects the eWOM effect but also the social identification of potential information receivers is also tested. Therefore, the implications of social similarity could have implications not only to increase interest in the information displayed around music, but also in how consumers could associate their own personalities around the content creator (Davis and Agrawal, 2018). Social identity is also confirmed as highly influential in hedonic products, being aligned with other studies such as Chiang et al. (2017). Hedonic products such as streaming music or music festival attendance may be associated with self-comparison and continued with self-identity. The cognitive effect of hedonic products consumptions could be related to self-esteem as social comparison is associated with self-identification behaviour and finally with a potential product consumption.

Another major contribution of this research is related to the social identity theory (Tajfel, 1974) confirming how social identity influence is still relevant in recent contexts such as social media networks. The perception of consumption of identity goods could have an important impact on those groups that feel represented. Therefore, music consumption and eWOM generation as Bourdieu (1984) pointed out could be a social classifier that affects personal image and involves social identity development. It also supports the congruity theory of Osgood and Tannenbaum (1955) as the consumer would tend to agree and follow what the social group message. Social identity and congruity theories have been major contributors in social behaviour studies explaining group tendencies and interactions. This paper helps better understand how both theories continue to be relevant in social networks where group communication and message adoption is being reinforced.

This paper also contributes to extend the academic research around social communication where common traits perception between communicator and receiver has been identify as an influence marker. The foundations of social communication concepts have helped to establish research of the dependent variables and confirms its current relevance communication influence in recent industries such as streaming platforms.

Finally, the findings agrees with latest bibliographical review of influencing factors on TV streaming platforms by Mulla (2022) where it is confirmed how streaming platforms consumption is associated with perceived enjoyment and socialization. The digital interactions around streaming platforms could affect brand loyalty and at the end of the day the user association with its product offer. Therefore, the eWOM generation in social media is a variable that should be studied in digital platforms consumption research as it could be highly influential on consumers behaviour.

Overall, the academic implications highlight how social variables like homophily, and social identity have an important role on eWOM effect supporting the social implication in culture communication. Furthermore, the eWOM influence framework builds on current research around eWOM (Akdim, 2021) confirming its influence in new digital consumption markets such as music streaming.

5.2 Managerial implications

Research results identify social networks as extremely relevant to product consumption such as music streaming platforms. From one side, digital platform should consider how contacts relationships could affect product consumption, in turn affecting decision-making processes. From the other side, content creators should also investigate how social platforms and influencers could trigger consumption of new products, being a clear alternative to traditional media such as radio or television.

Marketers should analyse eWOM generation and social connections to identify trends and groups generation that could affect future consumer behaviour. Therefore, consumption profile information could be extremely relevant as it could help to identify social trends and new product features preferences. Social media and streaming platforms could exchange profile information to create potential synergies to develop the consumers profile and customize service offers. Social influencers could be also addressed to promote indirectly not only specific products but also information and consumption platforms. The social identification of such group opinion leaders could be especially important in addressing new generation consumers such as Generation Z (Southgate, 2017). Storytelling could create content that enhance the connection of consumers with influencers, leading to social identity purchasing links (Farivar and Wang, 2022). The social research of individuals is a major practical contribution that enhances the social profile data value of consumers.

Social networks that compile lifestyle, preferences and consumer behaviour should start to be more important in the culture business strategy, as theirs insights could help better understand not only the decision-making processes of individuals but also their influence environment. The big data assortment of those networks could be integrated into AI development to better understand customer needs and better integrate culture communication into social projection and individual needs.

In line with the results of the study eWOM generation should be part of the marketer’s strategy. Digital content as music must be developed specifically and designed to be used in social media format with the intention and to be easily accessible to be used in short and effective format such as stories or shareable content. Producers should adapt to the social media environment with regular small content display to keep awareness of potential consumers and consumption leadership. In short, the result of this research confirms how the social component is deeply rooted in the consumption of intangible cultural goods such as music. Therefore, the theorical framework proposed should be considered for future research lines and managerial implications.

5.3 Conclusions

The result of this study confirms the relevance of homophily in message diffusion and influence. The perception of similar traits between users has been confirmed to be an important aspect to increase not only eWOM adoption but also a social belonging. Therefore common traits perception can be especially conducive to stimulate the attraction of consumption of experiential goods, such as audiovisual cultural goods (Chih et al., 2020). The homophily is identified as an uncertainty reducer and therefore an influential factor in the consumption of experiential goods whose intangibility is subject to a certain risk (Filieri et al., 2018; Ismagilova et al., 2020). Consequently, it is logical that a variable such as homophily is important in experiential consumption assuming social and relational embeddedness among users. In fact, it could be an explanatory factor in the creation of close circles and a variable to follow up in the social behaviour between groups (Barnett and Benefield, 2017).

The evidence regarding the influence of social identity corresponds with the results obtained in the literature review on social identity. Reviewed studies such as Chu et al. (2019) point out that the need for personal development predisposes users to engage in social networks and thus in eWOM adoption. Observation of behaviour associated with an individual's reference group may positively predispose to proactive adoption of such information. Figures such as “influencers” can play a key role in the dissemination of music, since their reach is not only greater than that of other users, but the involvement of their contacts favours greater adoption of their message. Therefore, it is logical that the greater the sense of social identity towards the eWOM sender, the greater the adoption of the eWOM content generated.

Finally, the literature review conducted on the influence of eWOM openly supports its relevance on current product consumption (Ruiz-Equihua et al., 2021), the results of this study extend its influence on new platforms such as streaming platforms where music consumption can be practically instantaneous and accessible (Aguiar and Waldfogel, 2018b). In this case, it should be considered that music has a hedonistic nature whose intangible essence makes the creation of expectations important in reducing the risk of consumption. Therefore, one of the main justifications for these results is based on the perception of utility and, in this case, on the reduction of uncertainty (Tariyal et al., 2022). A summary of the conclusions could be checked in Table 5.

5.4 Limitations and future research lines

The study used a small sample may not be representative of the total population. The limitations of the sample affect may affect the conclusion’s development being reasonable needed further research, indeed an additional sample could lead to a better understanding of regional differences or different cohorts’ trends in digital consumption. In addition, the whole research has been performed entirely in Spain. It would be interesting to perform similar studies in other countries with different social-culture-political environments as it has been confirmed that social factors could differ in social media behaviour (Alsaleh et al., 2019). A cross-cultural study could be conducted to identify potential differences between western and eastern countries or between different technological development environments (Kusawat and Teerakapibal, 2022). It could be possible to identify other industries related to cultural environment affected by the streaming phenomenon as it could probably be affected by a similar situation. Future research should address differences between music and other video streaming platforms to identify potential differences between hedonic products, such as cinema or TV.

As this study does not differentiate between different social groups another research line could be to study eWOM influence on differences between generations. New generations could be more influenced due social identity as their personality is still being developed and social role models and media impact has a high exposure. Future research could also address other variables that explore other social implications, such as social ties or credibility, as have been proved on other studies to be extremely relevant in eWOM influence (Moradi and Zihagh, 2022; Nofal et al., 2022).

Finally, complementary methodologies, such as a focus group or in-depth interview could be addressed to develop more the reasons for influence to understand better how social identification and homophily are related to music consumption.

Figures

Figure 1.

Theorical framework of eWOM influence on streaming platforms

Figure 2.

Theorical framework path coefficient and R-square result

Table 1.

Descriptive statistics

Items	Cases (%)
Gender
Female	280 (53.4)
Male	237 (45.2)
Other	7 (1.3)
Education
Primary school	2 (0.4)
High school	56 (10.7)
University degree	228 (43.5)
Post-graduate degree	153 (29.2)
PhD	56 (10.7)
Other	26 (5)
Generation
Z	86 (16.4)
Millennial	247 (47.1)
X	143 (27.3)
Baby boomer	48 (9.2)
Music consumption per day
<0.5h	158 (30.2)
0.5—1 h	144 (27.5)
1—2 h	91 (17.4)
>2 h	131 (25)

Table 2.

Constructs measurement

Constructs	Outer loadings
AWOMM1 ← WOM	0.843
AWOMM2 ←WOM	0.847
AWOMM3 ← WOM	0.798
AWOMM4 ← WOM	0.871
AWOMM5 ←WOM	0.853
HOMO1 ← HOMO	0.918
HOMO2 ← HOMO	0.959
HOMO3 ← HOMO	0.944
IS1 ← SOCIAL	0.747
IS2 ← SOCIAL	0.862
IS3 ← SOCIAL	0.886
IS4 ← SOCIAL	0.813
IS5 ← SOCIAL	0.821
IS6 ← SOCIAL	0.816
IS7 ← SOCIAL	0.828
IS8 ← SOCIAL	0.827
IWOMM1 ← CONSUM	0.843
IWOMM2 ←CONSUM	0.857
IWOMM3 ← CONSUM	0.806
IWOMM4 ← CONSUM	0.813

Notes: WOM = eWOM influence; HOMO = Homophily; SOCIAL = Social identity; CONSUM = Product consumption

Table 3.

Results reliability and validity tests

Constructs	Cronbach's alpha	CR	AVE	HTMT ratio
				WOM	HOMO	SOCIAL	CONSUM
WOM	0.898	0.925	0.710	0.843
HOMO	0.935	0.958	0.885	0.426	0.941
SOCIAL	0.933	0.945	0.682	0.494	0.573	0.826
CONSUM	0.849	0.898	0.689	0.787	0.460	0.477	0.830

Notes: WOM = eWOM influence; HOMO = Homophily; SOCIAL = Social identity; CONSUM = Product consumption; CR = Composite reliability; AVE = Average variance extracted

Table 4.

Results of the hypothesis testing

Hypothesis	Relations	Collinearity assesment VIF	Path coefficient	Standard deviation (STDEV)	T statistics(|O/STDEV|)	p-values
H1	HOMO → WOM	1.489	0.214	0.051	4.231	0.000
H2	HOMO → IS	1.000	0.573	0.035	16.321	0.000
H3	SOCIAL → WOM	1.489	0.371	0.048	7.790	0.000
H4	WOM → CONSUM	1.000	0.787	0.017	46.343	0.000

Notes: WOM = eWOM influence; HOMO = Homophily; SOCIAL = Social identity; CONSUM = Product consumption

Table 5.

Conclusions and theorical and managerial implications

Conclusions	Theorical and managerial implications
Homophily influences music eWOM adoption and social identity perception	Social similarity perception increases interest in message adoption but also in how consumers could associate their own personalities around the content creator. Social media and streaming platforms could exchange profile information to create potential synergies to develop the consumers profile and customize service offers
Social identity influences music eWOM adoption	Observation of behaviour associated with an individual's reference group may positively predispose to proactive adoption of such information. Figures such as “influencers” can play a key role in the dissemination of music, as their reach is not only greater than that of other users, but the involvement of their contacts favours greater adoption of their message
eWOM influences music consumption	eWOM influences music consumption, which in streaming platforms could lead to instantaneous access and consumption. Music has a hedonistic nature whose intangible essence makes the creation of expectations important in reducing the risk of consumption

Appendix. Research Questionnaire (detail of variables selected questions)

 

Research variables

Content about music shared by your contacts: music consumption influence

Regarding the comments or publications made by your contacts about music. rate from 1 to 7 the degree of relevance that the content shared by your contacts generates for you, with 1 = do not agree at all and 7 = totally agree. (Question to evaluate CONSUM):

• The music posted by my contacts is attractive to me.

• The music posted by my contacts helps me to conceive whether the songs are good or to my liking.

• I am willing to listen to the song if my contacts' comments invite me to think that I will like it.

• I find it hard to resist listening to songs shared by my contacts.

• I am influenced by the music my contacts share.

Use of musical information: eWOM adoption

What is the presence and use you make of the musical information you get from your contacts? Please rate from 1 to 7 with 1 = not at all agree and 7 = strongly agree (Question to evaluate WOM):

• I think about the music that my contacts share.

• Comments or opinions about music posted by my contacts can affect my opinion about songs.

• Comments or opinions about music posted by my contacts make me think.

• Comments or opinions about music posted by my contacts can be helpful.

Contacts: homophily

Regarding the users or contacts, you usually consult or from whom you receive information about music, television or movies. Rate from 1 to 7 with 1 = I do not agree at all and 7 = I totally agree (Question to measure HOMO):

• The contacts from whom I receive information are like me.

• The contacts from whom I receive information behave similarly to me.

• The contacts from whom I receive information are similar to me.

• The contacts from whom I receive information are like me in the way they communicate.

• I am familiar with the way my contacts express themselves.

• I feel competent with the language my contacts use.

Feelings about my contacts: social identity

The next block of questions aims to identify your feelings or sensations with respect to the group of contacts you usually consult or from whom you receive information about music, television, or movies. Rate from 1 to 7 with 1 = I do not agree at all and 7 = I totally agree. (Question to measure SOCIAL):

• I feel part of the social group of my contacts.

• It is important for me to be part of the social group of my contacts.

• Being part of the social group of my contacts is linked to my identity.

• Being part of the social group of my contacts is important for my image.

• I feel committed to belonging to the social group of my contacts.

• I feel connected to my contacts.

• My way of being is strongly related to my contacts' way of being.

• I consider it very valuable to belong to the social group of my contacts.