The influence of social media on purchase intention: The mediating role of brand equity

Abstract

The paper examined the relationship between social media and purchase intention and the mediation role of brand equity within Ghana’s fashion industry. The study was quantitative and employed the survey methodology to sample the views of 500 fashion customers. Statistical Package for Social Sciences (SPSS) and the structural equation modelling (SEM) technique were used using AMOS software version 22.0 to determine the hypothesized relationships of the study. The study findings revealed that surveillance, Information sharing and remuneration have significant and positive effects on brand equity. However, the relationship between social interaction and entertainment have negative and insignificant on brand equity. The positive significant relationships proposed to exist between brand equity and consumer purchase intention were all accepted. The findings can contribute to the scant empirical works that social media on brand equity and purchase intention in a single study. Recommendations were further made for management in the clothing industry, policy makers, and future researchers.

Keywords:

• Fashion
• social media
• uses and gratification
• purchase intention
• brand equity

PUBLIC INTEREST STATEMENT

Social media communication has to be expansively investigated since it plays a relevant role in creating brand equity and influencing consumer decisions. In this study, empirical research was conducted. The social media aspects looks at the inspirations and satisfaction consumers get by employing uses and gratification theory to understand how social media becomes a forbearer of brand equity and how that eventually affect consumers’ decision to effect purchase. The study used brand equity to moderate the influence of social media on purchase intention. The study concludes that brand equity plays a significant role in influencing buyer intentions. The study again found three dimensions of social media to be significant, but two were insignificant. Tentatively, this result may enrich social media and brand equity literature and particularly contribute to the formation of purchase intention. Essentially, the outcomes may be used to evaluate company promotional mix elements, specifically the development of communication strategy.

1. Introduction

The history of Ghanaian fashion is traced back to the pre-colonial era when indigenous Ghanaians used barks of trees and hides of animals to clothe themselves (Adu-Akwaboa, 1994). Today, Ghanaian Fashion has undergone a complete overhaul because of globalization. In order to remain competitive, IT based of awareness creation and sales are necessary. As a means of getting the attention and meeting specific needs of customers, experts in advertising have begun the use of social media website appeals in crafting advertisement in order to motivate users to share brand image (Waters, Canfield, Foster & Hardy, 2011; Gallaugher & Ransbotham, 2010; Dickey & Lewis, 2010). In addition, research by CliqAfrica and Avante (2016) postulate that the presence of fashion brands in Ghana is increasingly growing on social media platforms. According to Kemp (2020), Ghana had about 6.00 million social media users as of January 2020. The total number of social media users in the country increased by 629 thousand (+12%) between April 2019 and January 2020. Social media penetration in Ghana stood at 20% in January 2020. This is therefore an indication that Ghanaian fashion brands are also embracing social media (Aseidu, 2017). Fashion includes any market or product which contains a stylist element and will possibly exist within a period of time (Aseidu, 2017). The worldwide web was invented 28 years ago (Zhung, 2017). According to Karman (2015) the revolution of the internet has transformed the way data and information are gathered, way of communicating and how businesses are conducted globally. Accordingly, Stojanovic, Andreu and Curras-Perez (2018) indicated that in the last two decades social media intensity has triggered a marketing communication revolution providing different opportunities for interface between brands and users. The web 2.0 technology makes it simple for firms and customers to create and display content, share information, and recommend products to other users (Hutagalung, Dalimunthe, Pambudi, Hutagalung4 & Muda, 2017). The subsequent development of the social media has transformed the way people interact and share information with each other. This study focuses on Instagram (social media). Instagram is a mobile request and application that enables users to crack their mobile photos into images, and then shared on the Instagram application (Treitel, 2020). Those images can as well be shared on other social media network sites such as Facebook, Twitter, Tumblr and foursquare. According to Ershad and Aghajani (2017), Instagram is seen to be the fastest rising worldwide network site. It has over 400 million continuing monthly users and 80 million shared pictures with 3.5 million likes on daily basis (Hawi & Samaha, 2016). Treitel (2020) opined that Instagram has become widespread and more popular in people’s way of life, which can have adverse psychological effect. Today, people spend so much time on social media such as Instagram.

According to Agnihotri et al. (2016); Huang, Roth & Madden (2016); Zhan, Sun, Wang & Zhang (2016), businesses could make use of social media as a tool for creating and sustaining relationships with customers, improving brand equity and increase firm’s performance.

Brand equity is an important concept for understanding the objectives, mechanisms and the net impact of marketing efforts (Reynold & Philips, 2005). Aaker (1991) defines brand equity as a set of brand assets and liabilities linked to a brand, its name and symbol that add to or subtract from the value provided by a product or service to a firm and/or to that firm’s customers. According to Yoo and Donthu (2001), brand equity marks the difference in consumer choice between the focal branded product and an unbranded product given the same level of product features. Hence, brand equity generates value for consumers and helps brands create defensible competitive positions that cannot be easily transferred to other brands (Pitta & Kastsanis, 1995). Brands have resorted to several strategies including advertising, sales force, public relations, slogans and symbols in the attempt to develop brand equity (Aaker, 1991). Yazdanparast et al. (2015) opine that social media-based marketing activities tend to be an effective marketing strategy to build brand equity in this digital era. Consequently, brands have incorporated social media marketing into their marketing strategies to build lasting relationships and connections with consumers (Erdogmus & Cicek, 2012).

The function of social media as a forebearer of brand equity in the fashion industry is a scarcely researched theme (Keller, 2016). The interaction between users and brands has greater importance for brands since social media users depends on the advice of other user (Stojanovic et al., 2018), and this interaction amongst them stimuluses consumer decision-making process (Chan, 2019; Hinz et al., 2011). Recently, firms have become aware of the need to recognize the impact of social media communication and brand perception (Kumar et al., 2016; Stojanovic et al., 2018). Incontrovertibly, past studies have demonstrated a deeper and good understanding of social media communication (Swaminathan, 2016), yet very few studies have been conducted on interface of users with fashion brands (Stojanovic et al., 2018), neither have combined social media, brand equity and purchase intention in a single study. Again studies that have been conducted in areas of social media have used theories such as social presence theory (SPT) (Karikari, Osei-Frimpong & Owusu-Frimpong, 2017), theory of planned behaviour (TPB), theory of reasoned action (TRA) and technology acceptance model (TAM) (Ajjan & Harstshorne, 2008; Chang et al., 2015) to highlight intention to use social media or its adoption.

Therefore, scholars have researched social media communications in varied spheres for instance, the effect of social media and brand equity (Zahoor & Qureshi, 2017), predictors of social media analysis (Guede, Curiel & Antonoviaca, 2017), antecedents of social media marketing application (Brech, Messer, Schee, Rauschnabel & Ivens, 2017), effectiveness of social media in improving purchase intentions (Dehghani & Tumer, 2015), creative strategies in social media marketing (Ashley & Tuten, 2015), user interaction on social media and purchase intention (Hutter, Hautz, & Dennhardt, 2013), foundation of social media communication (Constantinides, 2014), impact of social media communication on brand loyalty (Balakrishman, Dahnil & Yi, 2014), and effect of social marketing on brand loyalty (Erdogmus & Cicek, 2012).

Even though there is huge number of studies also on social media and brand equity (Gurhan-Canli et al., 2016; Keller, 2016; Schivinski & Dabroski, 2015), there is inadequate research investigating social media, brand equity and purchase intention of Instagram users. Again, a study is yet to be sited comprehensively researching into social media, brand equity and purchase intention of fashion products in developing markets like Ghana. Consequently, the focus of this study is to investigate the association amongst social media, and consumer purchase intentions mediated by brand equity in the context of fashion product categories in Ghana. The main problem of the research is to study whether social media is positively contributing to brand perception with specific reference to Ghanaian social media (Instagram) users.

The rest of the sections are as follows: Section 2 looks at comprehensive review of literature on social media (and uses and gratification theory), followed by conceptual model. The next section focuses on research methodology, followed by summary of findings. Finally, we present discussions, conclusions, implication for practice and theory and limitations and suggestions for future studies.

2. Literature review and development of hypothesis

The theories underpinning this research are the uses and gratification theory and theory of planned behaviour.

2.1. Theory of planned behaviour

TPB was developed by Ajzen (1985), which is an extension of the theory of reasoned action (TRA) by Fishbein & Azjen (1975). Bojei and Abu (2016) referred to TPB as the degree to which an individual has either positive or negative assessment of a specific behaviour. TPB says that, many human behaviour can be predicted from behaviour intention and that consumers can behave in a planned manner and deliberately (King, 2019). In this study, the output variable (purchase intention) is anchored on the framework of theory of planned behaviour (TPB), because purchase intention is incorporated under the concept of behavioural intention (Ahmed, 2014). In TPB, intention is perceived as a direct forebear of genuine behaviour (Lin & Chen, 2010). Since most of the existing SM research has been conducted in Asia (Lee & Ma, 2012) and the United States (Choi, 2016), there is still a lack of emerging perspective. This study focuses on Ghanaian social media users who, compared to the developed markets social media users, are still very confined when it relates to using social media to promote brands (Newman, Fletcher, Levy, & Nielsen, 2016). Study in the Ghanaian context can therefore help to shed more light on the importance and universality of previously recognized interrelations. The TPB models the influence of motives and attitudes on intention and, subsequently, intention on behaviour.

2.2. Gratification Theory (UGT)

The UGT was developed (Katz, Blumler & Gurevitch, 1973) to assess user inspirations and gratifications of a particular media (Musal, Azmi & Ismail, 2016). UGT was one of the first approaches to study the dynamic role of the audience in media choice, recommending that people actively search for, identify with and employ media to fulfil specific gratification needs (Ku, Chu, & Tseng, 2013). This perspective recognizes that consumers are active social media (on Instagram), rather than passive, recipients of media. Considered one of the most effective frameworks, scholars have used U&G theory to explore a variety of social media sharing behaviour, including sharing links (Baek et al., 2011), news (Hanson & Haridakis 2008; Lee & Ma, 2012) and photos (Malik et al., 2016; Sung et al., 2016). In particular, this study explores the aim of use of Instagram users through the UGT though with some mediator variable in the background. Limited research works have been conducted to study the continuous use of SNS intention by following various theories, such as motivation theory, expectations-confirmation theory, social capital theory, and so on. Recently, most SNS studies have implemented the UGT to assess the intention to continue using SNS (Al-Jabri et al., 2015; Hsiao et al., 2015).

2.3. Social media

Social media are the different online platform and communities that encourage social interaction where users can exchange their views, experience and observations (Schroeder, 2017). Social media come in multiple form like Facebook, weblogs, wikis, LinkIn, Twitter, podcast and others (Yazdanparast, Joseph & Muniz, 2016). Social media offers significant opportunities for firms to engage buyers individually (Harrigan & Miles, 2014). Pena (2019) said, social media is an engagement platform that enable trust building that is important to the formation of brand equity. Social media has been regarded as a real mechanism that contributes to the organization’s marketing goals and strategies in respect of customer communication, customer relationship management (CRM) and customer involvement (Alalwan, Rana, Dwivendi & Algharabat, 2017).

2.4. Instagram as social media

From UGT perspective, there many motives for using Instagram including medium appeal, surveillance, social interaction, entertainment, coolness, documentation, creativity, self-expression and convenience (Alhabash & Ma, 2017; Sheldon & Bryant, 2016), time passing, task management, self-popularity motives (Kircaburun et al., 2018). Instagram is found to make people close to the families and friends from educational and informational, self-promotion and passing time motives/gratifications (Manasijevic et al., 2016; Sendurur et al., 2015) as well as posting selfies and photos (Kim & Chock, 2017); hence, UGT fits well in this study. Khan (2017) showed that users gain gratifications through viewing, commentating, liking and sharing of videos on Instagram. This current study considered entertainment, surveillance, social interaction, remunerative and information sharing.

2.5. Brand equity

In the 1990s, brand equity emerged. The development of brand equity is a complex process (Couto & Ferreira, 2017). Sasmita and Suki (2015) noted that a brand is the most valuable strength and assets of all firms. Regardless of the intangibility of brand equity, it provides value to the firm, leading to customer loyalty, creating and expanding opportunities, competitive advantage, negotiating power and higher margins (Perera, Wijesekera, Priyadarshani, Kodippili & Jayasuriya, 2019). Social media appears as a significant predictor of brand equity for general brand and high involvement products and services (Colicev et al., 2018). The consumer brand equity dimensions of this research are anchored on Schivinski and Debrowski (2015) categorization including brand quality, loyalty, awareness and association.

2.6. Purchase intentions (PI)

Purchase intention is indicated as buying behaviour (Martins, Costa, Oliveira, Goncalves, & Branco, 2019). A considerable amount of papers has published on purchase intention on inline platform (Rahim, Safin, Kheng, Abas & Ali, 2016). Purchase intention is a mixture of consumers’ interests in and likelihood of purchasing the product (Kim & Ko, 2012). Determining PI adopts buyer future behaviour based on their attitudes (Kim & Ko, 2010_b). On social media platform, there is exchange of information, ratings and responses that enables people to have trust and overcome the insecurities, which result in willingness to buy firms product (Bong, 2017). Consumers will usually undertake a process of appreciating the intended product or service to purchase, research, and analysis of the product before pledging to buy it (Lim et al., 2016). Online buying intentions become key element that can influence the usefulness of online stimuli (Shaouf, Lu & Li, 2016; Elwalda, Lu & Ali, 2016; Lu, Fan & Zhou, 2016). PI is the basis of demonstrating buying behaviour (Martins et al., 2019).

2.6.1. Social media and brand equity

SMC comprises marketing-related activities such as sharing online and blogging (Yazdanparast et al., 2016) and consists of applications, platforms and tools that allow customers to connect and communicate with one another (Pinto & Yagnik, 2017). Marketing managers are well aware of the need to understand the impact of social media communication on brand perception/equity (Kumar et al., 2016). SMC is a type of content that is shared on social media networks with goal of increasing awareness and persuading customers (Hanaysha, 2016). It is said the goal of advertising is to create brand equity in many ways including customer satisfaction, perceived quality, brand image, brand awareness and brand equity (Kumar et al., 2016). Advertising in social media helps to recall a brand and brand recognition particularly during purchase (Raza, Bakar & Mohamad, 2017).

2.6.2. Surveillance and brand equity

Surveillance refers to the need for information and evidence about influencers, which can assist people (Davis, 2009). Achieved via a passive social media use mode aimed at increasing one’s knowledge of the immediate social environment or just knowing what’s happening within the social network. It includes searching for specific types of people using advanced search characteristics; social network surfing, browsing statuses and profiles (including photos) of friends and friends of friends alternatively as those of other total strangers (Ruggiero, 2000). Asmah (2011) indicated that people use some type of media to monitor what is going on within and outside their realm. Hence social media is what individuals use to monitor what others are doing (Whiting & Williams, (2013). The traditional media is unable to report the whole shebang due to vastness of space, fear and censorship, whilst in social media people can upload videos, images and comment on such platforms (Musal et al., 2016). This is what culminates into gorilla journalism (Papoola, 2014). From the discussions above, it is hypothesized that

H1: Surveillance has a significant positive relationship with brand equity.

2.6.3. Social interaction and brand equity

UGT social interaction the context denotes to the need to build and develop relationship s and connect with others. The fundamental nature of social media is participatory, via which the sharing of content is both a form of expression and means of relationship building (van House et al. 2005). Social interactivity is widely regarded as an essential factor in determining a variety of affective and behavioural outcomes such as satisfaction, attitude, decision-making and encompassment (Coyle & Thorson, 2001; Fortin & Dholakia, 2005; Stewart & Pavlou, 2002). When a user has higher level of interaction in a social networking community, he or she will be more inclined to recommend the brand community, because of the benefit from engagement (Woisetschläger et al., 2008). Ayeh et al. (2013) enumerated common interest, value, and passion as the basis for consumer interaction. Cheng and Guo (2015) stated that posts of remark or praise, would entice more replies. Hence, number of replies might be viewed as mark of personal achievement (Dolan, 2016). In Dessaert (2017), people’s interest in interacting online was the main factor influencing social media engagement. Therefore, it hypothesized that

H₂: Social interaction has a significant positive relationship with brand equity.

2.6.4. Remunerative and brand equity

The level of remuneration offered to the consumer via social media content has been studied as a driver of consumer decisions to contribute to online communities (Muntinga et al., 2011). Consumers engage in social media use as they expect to gain some kind of reward such as an economic incentive, job-related benefit or personal wants (Muntinga et al., 2011). Social media content that offers a reward or remuneration includes monetary incentives, giveaways, prize drawings or monetary compensations (Füller, 2006). This type of content is anticipated to gratify users’ needs for remuneration and rewards within social media. Scholars have recommended that social media communication that is rewarding may cause lower levels of engagement, compared to informative content (Cvijikj & Michahelles, 2013; Dolan et al., 2016). In contrast, Cvijikj & Michahelles (2013) opined that social media content that is remunerative is a significant predictor of commenting behaviour of users. From the above description, it is hypothesized that

H₃: Remuneration has a significant positive relationship with brand equity.

2.6.5. Information sharing and brand equity

Use for information is the third “consumer motivations” for using social media that were identified by a researcher (Nyland & Near, 2007). This dimension describes the information resource use of social media. That is to say, how users can find information they want or need on social media. These uses were compiled and adapted from diverse studies on general internet uses and social network uses (Shao, 2009; Nyland & Near, 2007; Ruggiero, 2000; Recchiuti, 2003). Further, achieving numerous forms of information has been recommended as the most significant motive consumers use the internet (Maddox, 1998) and levels of information and attitude to the website have been found to be positively related (Chen et al., 2002). Whilst the importance of delivering information via advertisements has been recognized for traditional media (Rubin, 2002), the role of informational content in the online, social domain has only lately received attention. Probing for and receiving information about a brand is one of the main gratifications of consumer participation in online brand communities (Muntinga et al., 2011; Raacke & Bonds-Raacke, 2008). Hence, the researchers hypothesized that

H₄: Information sharing has a significant positive relationship with brand equity.

2.6.6. Entertainment and brand equity

Entertainment is the feeling of being sociable, cheerful, and happy, and referred to the way social media serves as a means for entertaining and escaping pressure (Lee, 2012; Kim, 2004). Lee and Ma (2012) believed the value of media entertainment was attributed to the ability to satisfy users’ needs for escapism, enjoyment, emotional release, and anxiety relief. Ducoffe (1996) found that entertainment directly affects user attitudes towards advertising: A favourable advertising message not only positively affects user attitude towards the advertising but also increases user brand attitude. Users seeing advertising to be sociable show more engagement behaviour and display positive attitudes towards the brand (Mir, 2011; Ünal et al., 2011). Hence, it hypothesized that

H₅: Entertainment has a significant positive relationship with brand equity.

2.7. Brand equity and purchase intention

Communications on social media contribute to enhancing customer attitude toward the brand, brand equity and increase WoM (Cambria et al., 2012). Chi Yeh and Yang (2009) showed that perceived brand quality mediates between perceived value and consumer purchase intention. Products that are associated with a famous brand name have influence of consumers viewing the brand as high quality, hence relating value altogether, which impacts on purchase intentions (Chi et al., 2009). In Halim and Hameed (2005), it was observed that consumers’ value perception will affect their purchase intention, indicating that, the higher the value, the higher the purchase decision. In another study, purchase decisions are influenced by brand involvement, brand image, product quality, brand loyalty, product attributes, brand awareness, and product quality (Chen & Change, 2010). Finally, using social media platforms to design, increase post and managing different sites to engage customers leads to increase awareness, product interest and likely influence purchase intention (Batra & Keller, 2016) Figure 1 and 2. From the discussions above it is hypothesized that

Figure 1. Conceptual model of the study

Figure 2. The structural model of the study

H₆: Brand equity has a significant positive relationship on purchase intention.

2.8. Social media, brand equity and purchase intention

Ashley and Tuten (2015) said branded social campaigns offer more touchpoints to influence ongoing communication between the brand story and consumers during the day which can increase brand relationship, persuade online customers to engage on social media content and assists managers to uncover common themes within customers’ feedback. Both UGC and FCC via social media have positive relationship with behaviour intentions. For instance, several researchers have found social media communication to be a positive influencer of consumer purchase intentions (Aaker. 1991, ; Goodrich, 2011; Keller & Lehmann, 2003; Martin et al., 2003; Shivinski & Dabrowski, 2013, 2015, 2016; Yoo et al., 2001). Khadim, Zafar, Younis and Nadeem (2014) postulated that FCC and UGC and positive feedback about the brand on network sites positively influence the perception of consumers and intention to buy. Hence, it hypothesized that

H₇: Social media has a significant positive influence on purchase intention through brand equity.

3. Methodology

3.1. Research strategy

According to Preibusch (2013), investigator uses a survey as a tool to collect data from institutions or individuals. The survey is suitable for a study that measures both the independent variables and the dependent variables and is a helpful way to collect participants’ information rapidly (Amin & Ahmad, 2012; Lee et al., 2014b; Nagengast et al., 2014). A survey is a method of gathering data about people to describe, compare, or explain their knowledge, attitudes, and behavior. Survey is a data collection strategy where data are collected from individuals, called respondents, by having them respond to questions (Monette et al., 2005). The approach enables the establishment of extensive opinions under natural situations (Auka, Bosire & Matern, 2013). Chandra (2004) noted that surveys are comparatively economical (especially self-administered surveys); surveys are valuable in describing the characteristics of an outsized population and no other method of observation can give this universal capability. In addition, when considering the limited time period of data collecting and the cost it is reasonable to apply the survey method for a study (Sekaran, 2000; Malhotra, 2007; Sekaran& Bougie, 2016).

3.2. Research design and research choice

This study used quantitative design. The goal of this research entails testing of hypothesis and is a systematic and scientific approach to quantify data (Sarma, 2015). The reason for using quantitative is based on research goal and research precedence. The study intended to establish the relationship between social media, brand equity and purchase intentions via regression analysis. It also provided avenue for testing the six hypotheses. We used quantitative correlational design to determine the relationship between social media, brand equity (CBBE) to in the current study. Cross-sectional was used in study by collecting data ones off (Sekaran & Bougie, 2016). Survey data will be collected from one organization at one point in time. Cross-sectional research typically uses diverse survey techniques to gather data and also uses various quantitative or multiple methods (Saunders, Lewis & Thornhill, 2012).

3.3. Population, sampling and sample size

The participants of this study were both men and women who follow, like and share information about brands on Instagram (online). These respondents took part in the study by completing online question via Qualtrics (qualtrics.com). The population was limited to only social media users on Instagram and those who follow fashion and related brands. Respondents must have the strategies and knowledge required provide adequate responses to the research questions (Marshall & Rossman, 2016; Yin, 2018). After all, the goal of selecting respondents is to acquire an in-depth understanding of the study phenomenon (Yin, 2018). A sample size greater than 30 and less than 500 is suitable for many studies (Abranovic, 1997; Ahmad, 2017; Roscoe, 1975; Sekaran, 2003). We used convenience sampling. Consequently, to improve generalization, the study was conducted in a natural situation, where the respondents may be honest in their answers (Polit & Beck, 2019). The researcher found convenient sampling technique useful in this study in order to increase the possibility of number of respondents. Finally, it was selected because of its objective of easy accessibility of 500 participants whilst overcoming time constrains and cost reduction (Zikmund, 2013).

3.4. Data collection

Structured online questions were used for this quantitative research, since it is a preferred method for finding out the different aspect of participants’ behaviour. Questionnaire method increases speed of data collection, cost effectiveness, and higher levels of objectivity compared to other data collection methods (Dudovskiy, 2013). The researcher used internet-based (online) questionnaire by distributing it through short message service (SMS) or/and over the email in the form hyperlink (weblink) (Saunders et al., 2016. Many research papers and researches have stressed the increasing popularity of online questionnaire (Dillman et al., 2014; Saunders, 2012), whilst most of them have used it (Schivinski & Dabrowski, 2016; Schivinski et al., 2019). The various literature reviews became the foundation for developing the instrument (Ai & Wu, 2018; Moore, 2016; Newton & Sundin, 2016). It provides adequate time to the respondents (1–3 weeks) and enables respondents to respond to the questionnaires with their ease and comfort. All items in the instrument using Likert, nominal, ordinal and ratio scales were considered closed. Close-ended questions help the research to code easily and allow the participants to make quick choice decisions amongst the alternative provided (Sekaran & Bougie, 2016). The instrument used sought behavioural questions that check purchase intentions and reason for using social media. The questionnaire was divided into two sections. Section A sought demography data such as sex, age, education and occupation. Section B sought information pertaining to the research matter (uses and gratification variables, brand equity and purchase intentions), where closed-ended questions were used. Likert scaling was adopted in studies because it has been commonly used by investigators as a tool for measuring beliefs, opinions, and attitudes. It also makes easy the analysis of data (Coleman, 2011).

3.5. Measurement instrument

Surveillance and social interaction items were adapted from Sung et al. (2016), information sharing (Kim and Johnson, 2012), remuneration and entertainment (Fuller, 2006). The dependent variable (PI) was measured using three items adapted from Yoo, Donthu and Lee (2001). In this study, the mediator is brand equity comprising brand awareness, brand loyalty, brand association and perceived quality. It is measured by adapting four items from (Schivinski et al., 2016). We used 5-point Likert scale: strongly disagree = 1 to strongly agree = 5 (Meyer, Gracey, Irlbeck & Akers, 2015).

3.6. Data analysis

As a requirement for SEM, confirmatory factor analysis was conducted, discriminant validity analyses were done. Reliability test was done via Cronbach’s alpha to understand the internal consistency of the instrument. In this study, result estimation will be produced by testing hypotheses. The mediator (brand equity) provides more evidence about why or how social and purchase intentions are interrelated. The research tested brand equity mediation impact by following Baron and Kenny (1986) in order to justify the link between the independent variables and the dependent variable. (a) social media (IV) should significantly affect brand equity (mediating variable), (b) brand equity should significantly affect purchase intention (dependent variable), (c) social media should significantly affect purchase intention and (d) the effect of social media on purchase intentions should no longer be significant (full mediation) or weaken (partial mediation) after the researcher controls for brand equity.

4. Results

4.1. Demographic profile of respondents

According to Pallant (2011), it’s advisable to include such demographic indicators such as age, gender, educational level and income when conducting studies involving human respondents. The demographic profile of respondents required for the study is outlined in Table 1. Respondents were profiled according to gender, marital status, age, academic qualifications and state of jobs. Table 1 includes the demographic profile of those respondents engaged in the study. Amongst the respondents, 310 females representing 62% and 190 males of 38% were engaged in the study with respect to the gender distribution of the respondents.And the total number of respondents was 190 (38%) were between the ages of 30–39. This was followed by the age range of 20–29 with 170 responses totaling 34%. The age ranges from 40 to 49, and the study revealed that, there were 120 responses resulting in 24% and at least 20 people representing 4% of the respondents were between the ages of 50 and 59. Five respondents were 60 and above years of age distribution, totaling 1%. The last demographic indicator to be identified has been the respondent’s job status. There were 295 constituting 59% respondents were employed with wages and salaries, whilst 154 (30.8%) were student who use Instagram. Those Instagram users who are out of work but not currently looking for work were 4 (0.8%) respondents, whilst those who are out of work and looking for work were 18 respondents representing 3.8%. Eventually, 24 respondents representing 4.8% were self-employed.

Table 1. Demographic profile of respondents

Group	Indicator	Frequency	%
Gender	Male	190	38
	Female	310	62
Age	20–29	170	34
	30–39	190	38
	40–49	120	24
	50–59	20	4
	60 and over	5	1
Occupation	A student	154	30.8
	Retired	5	1.0
	Employed for wages	295	59.0
	Self-employed	24	4.8
	Out of work but not currently looking for work	4	.8
	Out of work and looking for work	18	3.6
Education	SHS	7	1.4
	Diploma	25	5.0
	Degree/HND	233	46.6
	Masters	190	38.0
	PhD	45	9.0
Marital Status	Single, never married	141	28.2
	Married	347	69.4
	Widowed	11	2.2
	Separated	1	.2
Total		500	100

The respondents’ marital status was equally well captured and at face value, it is evident that (141) 28.2% singles were involved in the study belonged to the youthful group, in other words, they are referred to as the “Generation Y,” the working class and mostly urban dwellers who usually spend a great deal of time and resources on social media for new items including clothing, and this has a positive thesis reflection. The findings were in consistent with the research by Kemp (2020), who observed an upwards increase of users in Ghana to have increased by 629 thousand—representing 12% between April 2019 and January 2020. Three hundred forty-seven (347) were married, comprising 69.4%. Whereas 11 respondents were widows, totaling 2.2% and finally, 1 person representing 0.2% was separated from the spouse. As regards the respondents’ educational level, those with SHS as their highest level of education were 7, representing 1.4%. The educational average of the Diploma was 25, 5%. Holders of HND/degree combined were 233 respondents totaling 46.6%. Postgraduate stage had 190 master’s holders, representing 38%, and finally 45 respondents representing 9% holds PhD.

4.2. Assessing model fit

The confirmatory factor analysis was utilized in measuring the Goodness-of-fit (GOF) indices. The measurement model was given as “ꭓ²/df = 2, CFI = 0.942, TLI = 0.972, RAMSEA = 0.001”. Furthermore, the structural model was estimated and has been captured in table 4.8. This clearly posit that the GOF of the model met the stated cut-off points. That is, the factors of the independent variables fit for predictions on the dependent variable.

4.3. Reliability and validity test

Reliability test was performed to ensure accuracy level for measuring the intended latent build. The Cronbach’s alpha technique was utilized in testing the reliability of the instruments. To enhance the statistical level of confidence in quantitative study, reliability testing for answering of the formulated hypotheses is necessary. Cronbach’s alpha of reliability of the factors should exceed the minimum required level of 0.70 (George & Mallery, 2016) and this can be found in Table 2. Additionally, the factors loadings and composite reliability values provide evidence of convergent validity (Bagozzi & Yi, 1988). Further, if the average variance extracted (AVE) of the factors is greater than 0.5 then, thus, the revelation of convergent validity (Babin & Zikmund, 2016). Hence, this applied this analysis to show the present of convergent validity and reliability which can be found in Table 3.

Table 2. Reliability and validity

Items	Loadings	AVE	CR	Cronbach’s alpha
Surveillance		0.54	0.93
SV1	.705			.815
SV2	.779			.911
SV3	.738			.812
SV4	.741			.913
SV5	.686			.912
Social interaction		0.61	0.95
SI1	.787			.811
SI2	.760			.922
SI3	.794			.712
SI4	.779			.911
SI5	.781			.810
Information sharing		0.53	0.92
IFO1	.765			.711
IFO2	.785			.921
IFO3	.723			.952
IFO4	.625			.912
IFO5	.767			.913
Remuneration		0.60	0.98
REM1	.752			.913
REM2	.785			.912
REM3	.791			.942
REM4	.770			.912
Entertainment		0.59	0.94
ENT1	.752			.919
ENT2	.818			.918
ENT3	.791			.916
ENT4	.709			.917
ENT5	.767			.917
Brand equity		0.59	0.98
BQ1	.859			.915
BQ2	.735			.918
BQ3	.764			.912
BQ4	.700			.917
Purchase intention		0.62	0.95
PI1	.856			.912
PI2	.786			.913
PI3	.794			.961
PI4	.707			.912
PI5	.716			.914

Table 3. Discriminant validity

Constructs	AVE	1	2	3	4	5	6	7
SV	.540	.735
SI	.610	.589**	.781
IFO	.530	.663**	.695**	.728
REM	.600	.288**	.577**	.493**	.774
ENT	.590	.277	.011	.026	.087	.768
BQ	.590	.359**	.392**	.471**	.506**	−.003	.768
PI	.590	.340**	.436**	.475**	.495**	−.010	.729**	.768

Notes: SV, social media surveillance; SI, social interaction; IFO, information sharing; REM, remuneration; ENT, entertainment; BQ, brand equity; PI, purchase intention.

Inferable from Table 2 shows the reliability of the Cronbach’s alpha of the instruments and factor loadings. The values range from 0.911 to 0.917 indicating internal consistency within the data set. Also, this shows that all the variables have met the minimum threshold of 0.70 and above (Hair, Babin, Anderson, & Tatham, 2010). It is also evident from Table 2 that the loadings of the AVE were above the threshold 0.5, which fell between 0.625 and 0.859. That is an evidence of convergent validity (Fornell & Larcker, 1981). Additionally, the factors recorded a composite reliability of 0.92–0.98 which exceeded the minimum cut-off point of 0.70 (Hair et al., 2010).

Several studies have indicated that factor loading should be greater than 0.5 for better results (Truong & McColl, 2011; Hulland, 1999), whilst Chen & Tsai (2007) were also considered as a cut-off for appropriate loadings in the tourism context. Furthermore, Ertz, Karakas & Sarigollu (2016) found the factor loadings of 0.4 and above for their Confirmatory factor analysis when investigating pro-environmental consumer behaviour. So, we have not only chosen this criterion on the above ground but also as 0.6 is higher than the cut-offs for factor loading studies. In line with this, the standardized factor loading of all ranges of items was above the threshold limit of 0.6 and above also suggested by Chin, Gopal & Salisbury (1997) and Hair et al. (2006). Given these arguments, the lambda values (Factor Loadings) have exceeded all the thresholds.

4.4. Discriminant validity

To fulfil the requirement of discriminant validity, the square root of the factors of AVE should be greater that inter-factor correlation (Fornell & Lacker, 1981).

Notes: SV, social media surveillance; SI, social interaction; IFO, information sharing; REM, remuneration; ENT, entertainment; BQ, brand equity; PI, purchase intention.

The result in Table 3 shows that the result of the square root of AVE was more than the correlation numbers. Hence, discriminant validity is evidential within the constructs of the study which shows that the factors are distinct from each other. Again, the values of the correlation were not above 0.80, indicating no multicollinearity within the data set (Hair et al., 2010).

Table 4 provides the summary of hypotheses tested in study. The beta values, t and p values, and the hypotheses decisions were all captured. Also, the threshold of accepting or rejecting hypotheses was provided a footnote.

Table 4. Hypotheses test

Hypotheses path		Beta	t Values	p Value		Decision
H1	SV → BQ	0.116*	2.530	0.011		Accepted
H2	SI → BQ	−0.072	−1.550	0.121		Rejected
H3	IFO → BQ	0.269***	4.340	0.000		Accepted
H4	REM → BQ	0.304***	8.680	0.000		Accepted
H5	ENT → BQ	−0.054	−1.140	0.254		Rejected
H6	BQ → PI	0.165***	11.620	0.000		Accepted
Indirect paths
	SV → BQ → PI	0.844	5.582	0.000
	SI → BQ → PI	−0.035	−0.057	0.265
H7	IFO → BQ → PI	0.939	12.620	0.010
	REM → BQ → PI	0.031	11.110	0.001
	ENT → BQ → PI	−0.021	−0.211	0.060

Note: Social media surveillance = SV; Social interaction = SI; Information Sharing = IFO; Remuneration = REM; Entertainment = ENT; Brand Equity = BQ; Purchase Intention = PI; ***P < 0.01, **P < 0.05, *P < 0.10

The measurement model obtained through CFA shows the GOF of the measurement and structural model. All the indices met threshold as indicated in the Tables 4 and 5. This makes the data usable for further analysis. The results revealed that SV has a significant positive effect on BQ ($\beta =0.116,\hspace{0.17em}p<0.05)$) supported H₁, SI negatively and insignificantly affect BQ ($\beta =-0.072,\hspace{0.17em}p>0.05)$ rejected H₂, IFO has a significant positive effect on BQ ($\beta =0.269,\hspace{0.17em}p<0.05$) supported H₃, and REM ($\beta =0.304,\hspace{0.17em}p<0.05$) has a significant positive effect on BQ supported H_4, ENT ($\beta =-0.054,\hspace{0.17em}p<0.05$) has a insignificant negative effect on BQ rejected H₅. Finally, BQ is statistically significant to PI (confirmed H₆.

Table 5. Measurement and structural model

GOF	Measured Level	Structural	Accepted Value		References
RMSEA	0.001	0.042	<0.10	Bentler (1990)
CFI	0.942	0.942	>0.90	Byme (2010)
GFI	0.955	0.954	>0.90	Hu & Bentler (1999)
NFI	0.960	0.958	>0.90	Hu & Bentler (1999)
RFI	0.952	0.952
TLI	0.972	0.982	>0.90	Bentler (1990)
ꭓ^2/df	2	3	>0.05	Schumaker & Lomax (2001)

4.5. The indirect effect

The AMOS software was applied to estimate direct and indirect effects with involved mediator in order to understand the mediation effect. The model was used to investigate the mediating effects of gratification dimensions on output variable. As illustrated in Table 4, the indirect effect of gratification of social media through brand equity to purchase intention were evaluated.

Surveillance and purchase intention through brand equity: The effect of surveillance on purchase intention through brand equity was significantly positive (β = 0.844 and p = 0.000).

Social interaction and purchase intention through brand equity: The effect of social interaction and purchase intention through brand equity was insignificant and as well as pose a negative relationship (β = −0.035 and p = 0.265).

Information and purchase intention through brand equity: The effect of information and purchase intention through brand equity was significant and as well as pose a positive relationship (β = 0.939 and p = 0.010).

Remuneration and purchase intention through brand equity: The effect of entertainment and purchase intention through brand equity was significant and as well as pose a positive relationship (β = 0.031 and p = 0.001).

Entertainment and purchase intention through brand equity: The effect of remuneration and purchase intention through brand equity was insignificant and as well as pose a negative relationship (β = −0.021 and p = 0.060).

4.6. Discussions of study

The purpose of this quantitative, cross-sectional and survey study was to investigate the influence of social media communication on purchase intention of consumers in Ghana through brand equity. Whereas many of the previous research examined these constructs as single indications, this study compared all the gratification component and brand equity on the intention to buy. The study made use of six assumptions by way of hypotheses. Out of the five antecedence of the uses and gratification model (social media), only three have positive and significant effects on brand equity.

The study findings revealed that surveillance has a significant and positive effect on brand equity ($\beta =0.116,\hspace{0.17em}p<0.05$) supported H₁. This finding is in line with the works of Majeed et al. (2020), who found a positive and significant association between surveillance and customer engagement with the brand. Also, the relationship between social interaction and brand equity is negatively and insignificantly ($\beta =-0.072,\hspace{0.17em}p>0.05$), hence, rejected H₂. However, Abu-Rumman and Alhadid (2014) found positive effect of social media social interaction on brand value in an empirical research on mobile service providers in Jordan

Information sharing has a significant positive effect on brand equity ($\beta =0.269,\hspace{0.17em}p<0.05$), which supported H₃. Tresna and Wijaya’s (2015) study conforms with the current study, since they also found conversation and information sharing to have both positive and significant relationship with brand equity in an organization in the retail industry. The relationship between remuneration and brand equity is significantly positive ($\beta =0.304,\hspace{0.17em}p<0.05$) and hence supported H4. This study is contrary to Cvijikj & Michahelles (2013), who found that remuneration has positive but insignificant relationship with contribution to buyer intentions. Social media entertainment ($\beta =-0.054,\hspace{0.17em}p<0.05$) has an insignificant negative effect on brand equity rejected H₅. This is contrary to the study of Angela, Kim and Ko (2014), who found that social media marketing activities such as entertainment and social interaction and brand equity are positive and significant.

Finally, the relationship between brand equity and purchase intention is statistically positive and significant to ($\beta =0.1652,\hspace{0.17em}p<0.05$) confirmed H₆. This study is in line with Halim and Hameed (2005), who observed that consumers value perception will affect their purchase intention, indicating that, the higher the value, the higher the purchase decision. In another research, brand involvement, brand identity, product quality, brand loyalty, product attributes, brand recognition, and product quality influence purchasing decisions (Chen & Alter, 2010).

In social media, advertising appeal leads to positive brand association, which eventually leads to favourable behaviour intention (Adetunji et al., 2018). Abu-Rumman and Alhadid (2014) and Godey et al. (2016) found a significant relationship between social media entertainment and social interaction on brand equity and buyer purchase intention. This the reverse case for this study, where both social media entertainment and social interaction have negative and insignificant relationship with brand equity and buyer purchase intention. Overall, not all the components of social media communication lead to brand equity but brand equity definitely leads to customer purchase intention.

4.7. Conclusion

In this quantitative survey study, the researcher investigated the effect of social media on buyer purchase intention through brand equity to participate in social media activity at the different levels of social media participation for users. The current research established hypotheses from Katz (1959) based on the “UGT” and TRB. In the present study, too, the theory has helped explain the role of social media marketing strategies and tendency to online interaction as motivators for user engagement on social media with brand-related content. Building on the results of numerous studies based on the UGT system, the present study found a significant and positive impact of social media marketing strategies and tendency to online interaction on social media engagement of users. As a result, a strong emphasis on how businesses connect their customers online produces positive or negative interactions that further affect their brand equity. Domestic brands must also express their principles clearly to consumers in order to achieve a fair return on their investment. Given that it has never been easier to meet the target audiences at lower cost levels, domestic marketers need to participate in social media contact and embrace it as part of their marketing strategy. This can be done by tailoring digital promotion activities to match and meet the heart of their target audience, to which they can express their beliefs in order to boost sales and move up the marketing funnel. The attitude towards buying by customers can never be overlooked. Whether positive or negative, it impacts their intention to buy. The results showed that remuneration-to-brand equity was the best of all.

5. Implication for management

The findings of this research underscore the degree to which the social media behavior outcome indicators contribute to brand equity. In addition, the findings show that companies need to target related social media network activity outcomes and then build tailored strategies that can increase brand equity. Results can help companies realize that Instagram’s social media activity outcomes favor their brands more and can be used more efficiently in managing social media by considering the essence and direction of the influence of different social media network activity outcomes. Social networking is an essential source of knowledge that affects customer decision-making. Companies can use social media and other online applications to provide customers with up-to-date, appropriate and fresh information. Management of social media marketing would promote the brand community on which consumers rely as a significant source of knowledge in decision-making. In the end, this motivational system would establish the engagement between users.

6. Theoretical and managerial implications

A number of research in the previous literature validating the five dimensions of social media have accepted the used scale of measures of SMC behaviors enshrined in usage and gratification theory. The holistic effect of these dimensions determines the behaviors that inspire people to use social media. From a customer point of view, consumers engage on-line sites to access personalized resources and to receive the latest knowledge when sharing their insights and other tips. Second, SMC enhances brand equity, whilst the five established attributes of SMC are amongst the drivers of brand equity in the online context (Srinivasan et al., 2002).

Our results offer hands-on insights to practitioners. First, the greatest vital luxury fashion marketing strategy is to ensure that social media offer information and learning opportunities to customers. The study underscores the importance of providing up-to-date and pertinent information about the brand and its products in social media activities (SMA) to satisfy customer needs. Social media activities should be judiciously planned to provide mental and knowledge experiences, social communications, and personal integrative assistances that is likely to improve social status and reputation. By positioning SMA activities with visuals, thoughts, bodily experiences, problem-solving, and inquisitiveness, managers could rely on them by building and maintaining strong brand-consumer associations in social media environments (Brakus et al., 2009; Morgan-Thomas & Veloutsou, 2013).

7. Limitations and future studies

The first weakness was that, the researcher used relatively, a small number of items to calculate each of the structural model constructs discussed in this article. When replicating this study, researchers should consider the inclusion of objects in the measuring model. A potential investigator may consider conducting a mixed-method study to collect information-rich data through interviews whilst checking the meaning of variables through quantitative analysis. Finally, this study used a Ghanaian sample, making it impossible to generalize the findings to other countries. The majority of social media users in Ghana are still young people; thus, when replicating this research, one should take into consideration social, economic and cultural differences. A stronger confirmation and generalization of the results should be performed in different countries in future research in this area.

Section B: Questionnaire

SOCIAL MEDIA

Surveillance (SV)

I receive good information on brand X producers

Brand X producers keep me-up-to date with latest information.

Brand X producers offer customized information search.

Knowledge on the latest promotional information is always available

Brand X producers help me to learn about valuable products.

Social Interaction (SI)

Brand X producers give me opportunity to know other customers.

Producers reply to customer posted feedback/inquiries.

Producers use social media to allow me to interact easily with other customers.

Preferred brand social media allow me to share my opinion with others about products.

I use brand X producers Social media platform to easily connect to friends/others.

Information Sharing (IS)

I can easily share information with others via Brand X makers Social media page.

Provided latest Brand X promotional information.

I always get quick access to information on the brand X phone.

Remuneration (RM)

Brand X makers Social media page content is useful.

Social media offers me economic incentive (monetary incentives, giveaways)

Social media offer me prize drawings

Social media offer me job-related benefit

Social media offer me personal wants

Entertainment (ENT)

To pass time when I am bored

To entertain my self

Because is a habit just something to do

To occupy my free time

BRAND EQUITY

I can easily recognize this brand amongst other competing brands.

I trust the company who make this particular brand.

This brand would be my first choice.

In comparison to alternative brands, the likely quality of this brand is extremely high.

BRAND PURCHASE INTENTION

I would buy this product/brand rather than any other brands available

I am willing to recommend that brand to others to buy

I intend to purchase this product/brand in the future

No matter what I will buy this brand

This brand attracts my attention quickly

This quickly comes to mind when I have to make purchase

Additional information

Funding

The authors received no direct funding for this research.

Notes on contributors

Mohammed Majeed

Dr Majeed M. is a Lecturer (PhD) at Tamale Technical University, Tamale-Ghana. His current research interest includes branding, social media in service organizations. Majeed holds Doctor of Business Administration (DBA), MPhil and MBA Marketing. He lectures part time in many Ghanaian public Universities and a reviewer to many Journals in Management, hospitality, and marketing. Majeed has also published in good journals like Journal of Hospitality and Tourism Insights (Emerald).

Adu-Ansere Ashmond is a part-time lecturer at AAM-USTED in Kumasi, Ghana. His areas of research interest are marketing strategy, consumer behavior, social marketing, sustainability, green marketing and related areas.