How does social media marketing affect mobile game companies’ brand equity in China?

(1)

How does social media marketing

affect mobile game companies’

brand equity in China?

BACHELOR DEGREE PROJECT THESIS WITHIN: Marketing NUMBER OF CREDITS: 15 ECTS

PROGRAMME OF STUDY: Marketing Management AUTHOR: Lingyu Ye, Mingze Gao

(2)

Bachelor’s Thesis in Business Administration

Title: How does social media marketing affect mobile game companies’ brand equity in China? Authors: Lingyu Ye & Mingze Gao

Tutor: Ulf Linnman Date: 2020-5-17

Key terms: Social media marketing, Brand equity, Mobile gaming, Chinese consumer market

Abstract

Background: The widespread use of social media has offered brands a great opportunity to engage with their customers and promote sales. In China’s mobile gaming market, the huge user base of social media marketing provides companies with an excellent opportunity to develop their brand image and cultivate further awareness. However, the research regarding social media marketing’s influence on mobile game’s brand equity is still limited from a literary perspective.

Purpose: Firstly, the research aims to identify the factors of social media marketing that influencing mobile game companies’ brand equity. Secondly, this research will evaluate the importance of identified impact factors.

Method: In order to investigate the research subjects and answer research question properly, the positivism research paradigm has been implemented in this research. In addition, the deductive reasoning approach is applied to formulate and test hypotheses to generate reliable results with a high level of generalization and prediction. The theoretical foundation of this research has been formulated through the frame of reference. Then, the quantitative research methods were applied to the collection and analysis of the data. The primary data has been collected through a survey and the samples have been randomly selected. The main method for the data analysis was regression analysis.

Conclusion: In general, the research has demonstrated that social media marketing is a useful tool to develop brand equity for mobile game companies in the Chinese consumer market. Brand popularity and information quality are considered the most important influential factors of social media marketing on mobile game company’s brand equity in China.

(3)

Acknowledgements

Firstly, the authors of the research would like to acknowledge every person that involved in the research process. In particular, the authors would like to especially thank their tutor, Ulf Linnman for the guidance and constructive feedback throughout the research process.

Secondly, the authors would like to thank the participants of the thesis seminars for providing valuable and helpful advices and feedbacks

Thirdly, the authors would like to thank all of the people that participated in the pilot study and the survey section of the study. This research could not have been completed without the contribution of those participants.

Lastly, the authors would like to express their gratitude to the instructors of Jönköping International Business School for providing the authors with practical and useful guidelines for this thesis project.

Jönköping, May 17, 2020

(4)

1. Introduction

1.1 Background

The widespread use of social media has offered brands a great opportunity to engage with customers and promote sales. In China, a statistical report produced by the CNNIC (2019) indicates that the country had 854 million internet users in June 2019, and that its Internet penetration had reached 61.2%. The latest data from eMarketer (2019) states that China had 819.9 million social media users at the end of 2019. As a marketing channel, this huge user base for social media marketing provides companies with an excellent opportunity to develop their brand image and cultivate awareness.

A report from CNNIC (2019) states that 99.1% people access the internet through their smart phones in China in June 2019, which made smart phones the most used Internet access device in the country. Boosted by the widespread use of high-performance smart phones and high-speed mobile internet networks, as well as consumers increasing requirements on mobile entertainment, the mobile game industry has achieved tremendous development in China recently. CNNIC (2019) indicated that China had 467.56 million online game users in June 2019, and that the mobile gaming industry had achieved 18,032.5 million U.S. dollars in revenue in China in 2019 (Statista, 2020), thus proving that the country has become an important market that mobile game companies cannot ignore. Due to the complex and competitive market environment (Su, et al., 2016), any business enterprise that wants to gain an advantage must work hard to optimize and ensure their game’s quality in order to satisfy consumers (Liu et al., 2019). In addition, considering the intense competition in the gaming industry, a positive opinion of a brand would be one of the key reasons for consumers to select their products or services (Naatu, 2016). The data from the CNNIC (2019) indicates that social media comprise the most used mobile internet service in China.

1.2 Problem

Social media marketing has become a crucial channel for mobile game companies to develop their brand, and correspondingly brand equity has become a popular concept that continues to attract growing interest in the mobile game industry. However, the research regarding social media marketing’s influence on the brand equity of mobile game in

(7)

2 China is still limited from a literary perspective. According to Godey et al. (2016), social media marketing can have a different influence regarding brand equity depending on different markets. As part of the internet-based entertainment industry, mobile game companies are usually considered to have a modern business culture, and social media marketing’s effect on firms with different cultures may vary (Parsons & Lepkowska-White, 2018). Therefore, it is necessary to study how social media marketing affects mobile game companies’ brand equity in China.

1.3 Purpose

Due to the current gap in the literature, the research has two main purposes. Firstly, the research aims to identify the factors of social media marketing that influencing mobile game companies’ brand equity. Secondly, this research will evaluate the importance of identified impact factors. This research focus on marketing activities performed by mobile game companies in China. This is because it will be crucial for mobile game companies to organize and evaluate their strategies towards increasing brand equity in China’s mobile gaming market. Therefore, the study aims to answer the following research questions (RQ):

RQ1: What factors of social media marketing influencing mobile game company’s brand equity in China?

RQ2: What is the most important influencing factor of social media marketing on mobile game company’s brand equity in China?

1.4 Delimitation

This research has focused on large mobile game companies that have an active social media presence in China, which requires that the firm have official accounts on the main social media platforms in China. The following types of social media platforms have been involved in this research since the marketing activities which take place on these platforms are easy to access and participate in: microblogging, video sharing, social Q&A, and online communities.

(8)

3 In 2018, the largest shareholder in China’s mobile gaming market is Tencent, with other major shareholders including NetEase, Sony Interactive Entertainment, and Perfect World (iResearch, 2019) also involved in this research’s scope.

2. Frame of reference

2.1 Methods for the frame of reference

For the purpose of designing this investigation, the literature regarding social media marketing and brand equity have been reviewed respectively. The literature has been collected through searching for the keywords of: “social media marketing,” “brand equity,” “mobile game,” and “China” on Primo and Google Scholar. In order to ensure the quality of this research, most of the literature that has been reviewed consists of academic articles from peer-reviewed journals. Due to the lack of academic literature regarding China’s mobile gaming market and social media platforms, the information from professional research institutions and the firms’ annual reports have also been utilized to provide fundamental understandings regarding the research subjects.

2.2 Social media platforms in China

The data from Statista (2019) indicates that around 48 percent of the Chinese population used social media in 2018, which made the country the largest social media market in the world. Since most of the global social media platforms are unavailable in China due to the country’s internet policy (Davison, et al., 2018), this research has involved following several Chinese domestic social media platforms, listed as follows:

Weibo

Weibo is a Chinese microblogging platform similar to Facebook and Twitter. As one of the most popular social media platforms in China, the latest data states that the platform had 462 million active users in December 2018 (Weibo Investor Relations, 2019). Weibo’s users can share text, images or videos with other user on the platform for “self-presentation and interaction with those with the same interests” (Shu, et al., 2017).

(9)

4 TikTok (Douyin)

TikTok is one of the most popular short video platforms in China’s mobile social media market (Wang & Gao, 2019), enabling users to film, edit, and share videos with other users on the platform (Chen, et al., 2019). The data from Wearesocial (2020) indicates that TikTok was ranked as the third-most downloaded mobile application in China throughout 2019.

Bilibili

Bilibili is one of the most popular video sharing platforms in China, through its unique danmaku interface, video viewers are allowed to “input “live” comments in a way that is directly overlaid onto the video” (Yang, 2020). The annual report from Bilibili (2020) indicates that the platform had over 67.9 million users in China at the end of 2019.

Zhihu

As an online Q&A site that similar to Quora, Zhihu was ranked as the fourth largest social media platform in China at the end of 2017 (CNNIC, 2018). Users are able to ask questions, provide answers, and interact with each other through rating and commenting on others’ answers (Stockmann & Luo, 2017).

Baidu Tieba

“Baidu Tieba is a BBS-like chat forum with an average of 50 million new posts posted per day” (Stockmann & Luo, 2017). The platform allows users to communicate with others that have similar interests by establishing different groups (Stockmann & Luo, 2017).

2.3 China’s mobile gaming market

A mobile game is defined as a video game that operates on operating system-based mobile devices, such as smart phones and tablets that run Apple iOS or Google Android (Su, et al., 2016). According to the survey from CNNIC (2016), from 2012 to 2016, with the increasing user base of smartphones in China, the number of mobile game users rose sharply and steadily, with 248 million users participating in the consumption of mobile games at the end of 2016, and this number only continues to grow. That is, in the period

(10)

5 from 2015 to 2016, the growth of mobile online game users exceeded 30 million. The growth trends regarding the user base and usage percentage of mobile games in China between 2012 and 2016 are displayed in Table 1.

Table 1: User base and usage percentage of mobile game in China 2012-2016

2012 2013 2014 2015 2016 User based (10,000) 7817 10751 13941 21535 24823 Usage percentage (%) 25.8 30.2 33.2 43.1 44.6

Source: China Internet Network Information Center (CNNIC, 2016)

In January 2020, over 74 percent of internet users in China chose to play video games on their smart phones (Wearesocial, 2020), which made smart phones the most popular gaming device in the Chinese market compared to PCs, gaming consoles, and other devices. iResearch (2019) states that the monthly playing frequency of mobile games has reached 57.24 in April 2019, while casual games, board and card games, shooting games, and MOBA are the most played types of mobile games in the Chinese market. As the largest mobile gaming company in China, Tencent has gained 26,035 million Chinese Yuan in revenue in 2019 from its mobile game titles such as Glory of Kings and Peacekeeper Elite (Tencent, 2020). In general, iResearch (2019) forecasted that China’s mobile gaming market will continue its growth trends and reach a total revenue of 270.36 billion Chinese Yuan revenue in 2021.

2.3.1 Representative product in China’s mobile game market

"Glory of Kings" is a mobile multiplayer online battle arena game designed by Tencent. The game was publicly tested in November 2015 and soon became the most profitable mobile game in China (eMarketer, 2017). In addition, this game has also been launched overseas where it was re-branded as "Arena of Valor." Throughout its development, "Glory of Kings" has become the most successful mobile game in China. Furthermore, the data from eMarketer (2017) indicates that the game had 200 million registered users

(11)

6 in 2017 and achieved 828 million U.S. Dollars in revenue in the first quarter of 2017. One of the reasons for the success of "Glory of the King" in the Chinese mobile gaming market is considered to be its original game character design, as various well-known characters from ancient Chinese history have been redesigned to become characters suitable for the game. These interesting characters were designed based on history, and traditional culture is highly important to the game’s ability to attract users' attention (Tencent, 2020).

2.4 Social media marketing

Social media are online content sharing sites and platforms that enable users to create, modify, share, and discuss internet content (Kietzmann, et al., 2011). The most used social media platforms at present mainly include Facebook, Instagram, Whatsapp, Twitter, and TikTok. Additionally, social media marketing (SMM) can be defined as the actions conducted on and over social media platforms to promote a product or service (Felix, et al., 2017), or increase the presence of the enterprise on the internet. Social media marketing allows enterprises to easily monitor the progress and achievements of a marketing campaign (He, et al., 2015).

2.4.1 Corporate perspective

From a corporate perspective, SMM has irreplaceable value for enterprises because of its obvious advantages over traditional marketing (As’Ad & Alhadid, 2014). Through SMM, enterprises are able to develop positive interactive communications with consumers (Kim and Ko, 2012), which thus made SMM an effective method to assist the firm with increasing their capability when it comes to consumer relationship management (Wang & Kim, 2017). SMM can also contribute to improving the relationship equity and brand equity of the enterprise (Kim and Ko, 2012). In addition, for enterprises that are seeking to develop a strong brand, SMM has been identified as an excellent approach to building brand awareness on the internet, especially if it can be used in an efficient manner (Monica Bîja & Raluca Balaş, 2014; Seo & Park, 2018). SMM can also be used to influence consumers' attitudes towards brands (Yazdanparast, et al., 2016). Due to the positive effect of SMM on firm’s business performance (Wang & Kim, 2017), Venciūtė (2018) states that enterprises should treat social media marketing as an organizational capability rather than only as a marketing method. However, Parsons and

(12)

Lepkowska-7 White (2018) found that firms with different marketing cultures and structures harbour different views towards social media marketing. Those firms that possess a modern marketing culture and effective digital communications guidelines are more willing to and more effective at developing and benefiting from their interactions with consumer and stakeholders through social media networks (Parsons & Lepkowska-White, 2018). Furthermore, firms should note that brands' active presence on social media may negatively influence the performance of social media marketing, which suggests that firms should pay attention to the quality of their social media content when attempting engagement with their consumers (Tafesse and Wien, 2018). In summary, the complexity of SMM requires firms to have a deeper understanding of different social media platforms to adjust their marketing activities in order to ensure that their social media marketing is effective (Zhu & Chen, 2015).

2.4.2 Consumer perspective

From a consumer perspective, the response of consumers to SMM is positively influenced by the strength of the SMM messages, including their argument quality, message popularity, and message attractiveness. Therefore, firms would be able to increase the influence of their SMM activities by strengthening the quality of their SMM messages (Chang, Yu, & Lu, 2015). In addition, Khan et al. (2016) have indicated that the influence of an SMM message can be measured by the number of likes, comments and shares it receives, and these indicators can be influenced by different types of content. For example, the study identified that vivid and interactive content have a positive impact on the number of likes they receive, while interactive content has a positive impact on the number of comments, and vivid content is positively related with the number of shares (Khan, et al., 2016). In business practice, SMM activities affect consumers’ loyalty towards a brand (Ismail, 2017) and their purchasing intention (Boon-long & Wongsurawat, 2015). For example, a consumer’s brand loyalty can be positively influenced if the brand offers advantageous campaigns, relevant content, popular content, and applications on various social media platforms (Erdoğmuş & Çiçek, 2012). Furthermore, the study conducted by Akar and Topçu (2011) indicates that consumer attitudes towards social media marketing have a significantly positive relationship with their income status. Besides, consumers from different countries may be impacted by

(13)

8 SMM in different ways. An international study found that luxury brands' social media marketing activities are more effective in building brand equity in China compared with France, Italy, and India, but are less effective at developing brand loyalty (Godey, et al., 2016).

2.4.3 Key factors in social media marketing

Although SMM has become a widely used marketing approach for mobile game companies, the researchers found that the literatures regarding mobile game company’s SMM activities are still limited. However, there are some existed literatures regarding the evaluation of enterprises’ SMM activities. Although these studies are not conducted with a focus on mobile gaming area, the researcher believe that these previous studies will contribute to this research by formulate a comprehensive understanding about SMM, which would help researchers to understand and evaluate mobile game company’s SMM activities. Among these literatures, following factors have been found to have a significantly effect in enterprises SMM activities:

2.4.3.1 Brand popularity

Brand popularity refers to the degree to which people know of and understand a company's presence on social media platforms, and which reflects a company's reputation and brand awareness on the internet. Huang and Sarigöllü (2012) indicated that brand awareness is positively related to the customer mind-set brand equity and market performance of the brand. This also means that brand awareness has a positive influence on the results of an enterprise’s social media marketing activities (Barreda, et al., 2015). Furthermore, the most important factor that defines the popularity of a brand is its fan base. The fans of a brand are defined as a group of people who follow the official account of a particular brand on social media platforms (Lipsman, et al., 2012), and fan base refers to the total number of fans that a brand’s social media accounts have. As such, social media marketing is a powerful tool for enterprises because of its ability to deliver branded content to a large audience through the brand’s fans and their friends (Lipsman, et al., 2012).

(14)

9 2.4.3.2 Information quality

The quality of the content that a brand posts on social media is usually considered to have a significant effect on the results of that brand’s social media marketing activities (Kim, et al., 2017). Information quality refers to the overall perceived value of the content posted by a brand on webpage (McKinney, et al., 2002). Furthermore, high-quality information may boost interactive communication between brands and their followers (Shi, et al., 2016). On the contrary, excessive, low-quality content will lead to an information overload that prevents the usage of social media (Chai, et al., 2009).

2.4.3.3 Involvement of influencer

Consumers’ attitudes towards brands and their purchasing intensions sometimes do not stem from their own ideas or perceptions, but rather are influenced by those around them (Jin, et al., 2019). Influencers such as celebrities and successful entrepreneurs often have a strong social influence, in addition to usually having a large number of followers on social media (Jin, et al., 2019). As a result, the information and content they post on social networks has an outsized impact, and their influence may continue to grow through word-of-mouth communication (Audrezet, et al., 2018). Tri Hanifawati et al. (2019) determined that influencers have a significant effect on consumer’s attitudes towards brands. Additionally, the trustworthiness of influencers enables them to affect people’s attitudes and perceptions (UzunoAlu & Kip, 2014). Influencers are usually perceived by customers as other ordinary customer on social media, so the information from influencers is most likely to be considered more reliable than information coming from marketers (Audrezet, et al., 2018).

2.4.3.4 Interaction with fans

The development of the internet has provided companies with the opportunities to exchange information and interact with fans on social media platforms (Jara, et al., 2014). By interacting with consumers, companies can better understand the consumers’ needs, on the other hand, the consumers are also able to learn more about the brand and its products through interactive communication (Zhang & Lin, 2015). In addition, the interactive communication taking place between brands and fans also enables the brands to foster closer consumer relationships (Zhang & Lin, 2015). On the other hand,

(15)

10 companies can undertake more comprehensive evaluations of their products through these interactions, in order to achieve more targeted improvements and satisfy the consumers’ needs (Jara, et al., 2014).

2.4.3.5 Intensity of sales promotion activities

Offline sales promotions have proven to be a useful method to boost sales and increase a firm’s market performance in business practice (Hultén & Vanyushyn, 2014). Similarly, sales promotion activities taking place over social media platforms also have a positive effect on consumers purchasing intensions (Handy Martinus & Liza Anggraini, 2018). In addition, companies can also post interactive promotional activities on social media, such as quizzes and sweepstakes, to promote the spread of their marketing information on social networks in order to reach more consumers and improve the effects of SMM activities. Furthermore, sales promotions also has a significant influence on consumers’ brand loyalty and brand associations (Sinha & Verma, 2018).

2.5 Brand equity

Lasser et al. (1995) states that brand equity is “the beliefs, images and core associations consumers have about particular brands”. During the years of development and maturation, brand equity has become a valuable concept in modern business. Brand equity is intangible assets for enterprises (Zaichkowsky et al., 2010), Kim (1990) believed that enterprises influence consumers' thinking and decision-making through corporate brands, with this process subsequently forming corporate brand equity. Aaker (1991) states that brand names and brand symbols are both considered a part of brand equity, and which influence the value of the enterprises by altering or improving the image of their products or services. According to Keller (2003), the additional effects and benefits of enterprises’ products endowed by brand name are also considered brand equity. Hence, brand equity plays a key role in promotion and publicity, influencing consumers' purchasing decisions. Further, brand equity can enhance enterprises' popularity and expand their competitive advantages (Jalilvand, et al., 2011).

(16)

11 2.5.1 Corporate perspective

From the corporate perspective, brand equity is the core of a brand, and its influential factors mainly include corporate image, corporate marketing mix, corporate scale and R&D capability (Keller & Lehmann, 2003). In addition, Keller (2003) highlighted that the promotion of brand equity through marketing activities will be affected by a number of factors, including the company’s profile, brand name and source, and product promotion methods. Ailawadi et al. (2003) believe that an enterprise’s brand equity is positively affected by a company’s the marketing mix, competitive advantages and product categories. Zhao et al. (2020) analysed source identification and brand equity from an integrated perspective, concluding that brand strength plays an important role. In addition, corporate marketing strategies, such as promotional strategy and method of product management mode, had a significant impact on brand equity (Kim, et al., 2003).

2.5.2 Consumer perspective

From a consumer perspective, the pricing or services of products have a significantly positive impact on consumer-based brand equity. The lower the price, the more conducive it is to the improvement of the brand’s equity (Lassar, Banwari, & Sharma, 1995). Furthermore, consumers’ awareness and loyalty towards a specific brand would be significantly impacted and influenced by social media marketing (Laroche, Habibi, & Richard, 2012). Another key point is consumer perceived product innovation and consumer’s attitudes towards brand are positively related with their perceive product innovation (Hanaysha & Hilman, 2015).

2.5.3 Multidimensional Brand Equity Scale theory

The Multidimensional Brand Equity scale (MBE) was developed by Yoo and Donthu (2001) based on Aaker’s (1991) five components brand equity model and Keller’s (1993) consumer-based brand equity theory. The MBE indicate that consumer-based brand equity can be measured with three dimensions including brand loyalty, perceived quality, and brand awareness/associations. The theory also explains how brand equity can be influenced by brand knowledge, purchasing experience, marketing campaigns, and brand image. MBE is a valuable and practical theory that have been used to analyse brand equity related issues in various industries such as hotel (Kim, et al., 2008), apparel (Li & Ellis,

(17)

12 2014), banking (Nath & Bawa, 2011), insurance (Nath & Bawa, 2011), cellular services (Nath & Bawa, 2011), and tourism (Hyun & Kim, 2019). Therefore, the researchers believe MBE would be an applicable theory in this research to define the elements of mobile game company’s brand equity. The three dimensions of brand equity in MBE theory are further discussed as following:

2.5.3.1 Brand loyalty

As a core aspect of brand equity in the MBE theory, brand loyalty defines consumers intension of repeat purchase regarding a particular brand (Nam, et al., 2011). Consumers are more willing to purchase a brand that they are familiar with (Shum, 2004). The firms that have a more loyal customer base will have a competitive advantage in market competition, and the customers that are more loyal to a brand will thus create more profits for the firm (Baldinger & Rubinson, 1996). Aaker (1991) identified five levels of brand loyalty—the first level represents price-sensitive consumers that have no loyalty towards brands, the second level represents habitual consumers who have no reason to switching to other brands, and the third level represents consumers that are satisfied with the brands, as these consumers usually have a relatively high switching cost, the fourth level represents loyal consumers that consider the brand almost as a friend, and the fifth level represents committed consumers who are extremely loyal to the brands.

2.5.3.2 Perceived quality

Perceived quality is a key concept of brand equity in the MBE theory that refers to a consumer’s overall perception regarding the quality of a brand’s offerings (Aaker, 1991). Consumers formulate brand perceived quality based on their own unique cognition and subjective evaluation of the brand when they experienced a product or service. Dolezalová et al. (2016) states that consumer’s perception of a product’s quality is a crucial factor that influences their purchasing decisions. The factors that affect product quality and service level are mainly including product practicality, reliability, durability, cost performance, social background, physical environment, and consumer’s cognitive ability. In the entertainment industry, perceived quality may be interpreted as a consumer’s perceived experience of using a product (Ginsburgh & Weyers, 1999). In

(18)

13 addition, perceived quality is not the actual quality of the products since it is a subjective judgement made by a consumer (Zeithaml, 1988).

2.5.3.3 Brand association

Brand association strongly affects brand equity because it defines a consumer’s ability to identify brands (Aaker, 1991). Aaker (1991) believes that anything that linked to a brand can be classified as brand association, which is extremely important to formulating consumers purchasing intension towards a brand. O’Cass and Lim (2002) identified different factors of brand association including price perception, brand personality, brand-elicited feelings and self-image, which have a similar effect on a consumer’s preferences and willingness to purchase a brand. By developing an understanding of consumers, companies are able to create brand names, logos, and slogans that can attract the attention of consumers and assist them in remembering brands.

3. Methodology and methods

3.1 Research paradigm

“A research paradigm is a philosophical framework that guides how scientific research should be conducted” (Collis & Hussey, 2014, p. 43). In order to investigate the research subjects and answer research question properly, the positivism research paradigm has been implemented in this research. Positivism paradigm “encompasses the empirical methodology, meaning data is derived from experiment and observation” (Mcgregor & Murnane, 2010, p. 421). By selecting such a paradigm for this research, the researchers are enabled to propose and test hypotheses in order to explain specific phenomena (Mcgregor & Murnane, 2010). Which is the most applicable philosophical framework regarding the research topic.

3.2 Research approach

In positivism studies, the deductive reasoning approach are widely applied to formulate and test hypotheses to generate reliable results with a high level of generalization and prediction (Collis & Hussey, 2014). Therefore, the deductive approach has been applied in this research. Quantitative research approach that based on statistical analysis regarding

(19)

14 quantitative data is usually carried out to observe and measure the variables of a phenomena within positivism (Collis & Hussey, 2014). In this research, the researchers intend to measure the relationships between social media marketing and mobile game company’s brand equity as a social phenomenon. Therefore, quantitative approach is considered as a more appropriate research approach compared with qualitative approach. It enables researchers to generate results from sample to the entire population (Collis & Hussey, 2014).

3.3 Population and sample size

As defined in the research questions, the population is defined as China’s internet users who play mobile games and use social media. It enables researchers to generate results with the maximized level of generalization. Although the exact population size is unknown in this research, the researchers have reason to believe that the population size may exceeds 100 million according to the statistical data regarding the user based of social media and mobile games in China. The data from CNNIC (2019) indicates that China had 467.56 million mobile game users in June 2019, which reflects that a large population may have been involved in the research. Therefore, the Cochran’s Sample Size Formula would be an appropriate method to calculate the sample size in this research:

𝑛 =𝑍 𝑝𝑞 𝑒

Due to the limited information regarding the subject, we must assume that 50% of China’s internet users who play mobile games and use social media, which enables us to have maximum variability, so the p equals 0.5 and q equals 0.5. The researchers have defined a 95% confidence level and a 3% margin of error in the research, which gives the researchers a z-value equals 1.96 and e equals 0.03. Hence, we have 𝑛 ≈ 1068, which means that the research requires at least 1068 samples to generate a result with 95% confidence level and 3% margin of error. In this way, although the sample may still small compared with the large population, the accuracy and reliability can be ensured.

3.4 Research variables

In the frame of reference, the researchers have identified five key factors in enterprises’ social media marketing activities based on previous studies within the field. Those factors

(20)

15 are considered to have significantly effect on the results of firm’s social media marketing activities. Therefore, those factors including brand popularity, information quality, involvement of influencer, interaction with fans, and intensity of sales promotion activities will be the independent variables in this research.

As a widely recognized method, the MBE theory enables researchers to measure mobile game company’s brand equity in more detailed and reasonable way. Therefore, the three components of brand equity in MBE theory including brand loyalty, perceived quality, and brand association are applied as dependent variables in this research.

3.5 Research model and hypothesis

Based on the independent variables and dependent variables that have been proposed above, the research model can be established as the Figure 1 shows. The research model provides a base for researchers to investigate and answer the research questions. More specifically, research model allows researchers to evaluate the relationships between independent variables (social media marketing) and dependent variables (brand equity) by proposing relevant hypotheses. The impacts of independent variables on dependent variables can only be evaluated if the they are correlated with each other, that reflects the importance of hypotheses in this research. In order to test the correlation relationships between each independent variables and dependent variables in a comprehensive manner, fifteen research hypotheses are formulated based on the research model. By proposing and testing these hypotheses, researchers are able to answer the research questions.

(21)

16 Figure 1: Research model

Hypotheses regarding brand popularity in correlation with brand equity: H1: Brand popularity has a positive impact on brand loyalty

H2: Brand popularity has a positive impact on perceived quality H3: Brand popularity has a positive impact on brand association

Hypotheses regarding information quality in correlation with brand equity: H4: Information quality has a positive impact on brand loyalty

H5: Information quality has a positive impact on perceived quality H6: Information quality has a positive impact on brand association

Hypotheses regarding involvement of influencer in correlation with brand equity: H7: The involvement of influencers has a positive impact on brand loyalty

H8: The involvement of influencers has a positive impact on perceived quality H9: The involvement of influencers has a positive impact on brand association

Brand Popularity Information Quality Involvement of Influencer Interaction with Fans Intensity of Sales Promotion Activities Brand Loyalty Perceived Quality Brand Association

(22)

17 Hypotheses regarding interaction with fans in correlation with brand equity:

H10: Interaction with fans has a positive impact on brand loyalty

H11: Interaction with fans has a positive impact on perceived quality

H12: Interaction with fans has a positive impact on brand association

Hypotheses regarding the intensity of sales promotion activities in correlation with brand equity:

H13: The intensity of sales promotion activities has a positive impact on brand loyalty

H14: The intensity of sales promotion activities has a positive impact on perceived quality

H15: The intensity of sales promotion activities has a positive impact on brand association

3.6 Survey design

Since this research utilized quantitative research methods, survey was the main source of primary data in this research, and the quality of the survey was extremely important to generating reliable results. Therefore, the survey had to be designed using a scientific and comprehensive approach.

Based on the formulated hypotheses in combination with some general factors regarding the audience’s backgrounds, the indicator structurer of the survey could be formulated as the Table 2 displays.

Table 2: Indicator structure of the survey

First-level Indicator Second-level Indicator

Background information

Gender, age, level of education

How long has the audience been using social media Frequency of social media use per day

Amount of time spent on social media per day Most-used social media platform

Social media marketing Brand popularity Information quality

(23)

18 Involvement of influencer

Interaction with fans

Intensity of sales promotion activities

Brand equity

Brand loyalty Perceived quality Brand association

Since the second-level indicators of social media marketing and brand equity are difficult to evaluate directly for the audience, the survey questions concerning these indicators have been designed based on their expressions proposed by researchers as Table 3 shows. The expressions of each second-level indicator items have been proposed based on the literary sources listed in Appendix 1.

Table 3: Expressions of second-level indicators regarding social media marketing and brand equity

Indicator Number Expression

Brand popularity - A

A1 You only follow the brand’s page that you like A2 You often check the brand’s posts in your leisure time A3 You talk about the brand’s posts with people around

you

Information quality

– B

B1 The brands you follow often post interesting content B2 The brands you follow care about the quality of what

they post

B3 You follow a brand because you care about the quality of information

B4 You think the brands you follow post useful information

B5 You think the brands you follow try to post information at the right time and in the right way

(24)

19 Involvement of

influencer – C

C1 You follow influencers on social media

C2 You think the branded contents posted by influencer help you understand the brands

C3 You think influencers would be great prolocutor for brands

Interaction with fans

– D

D1 You think interaction with brands on social media helps you better understand the brands

D2 You think it is important that brands interact with fans in a timely manner

D3 You think interaction with brands on social media helps you to better understand the products

D4 You feel good when interacting with brands

Intensity of sales promotion

activities – E

E1 You think the offers of a brand’s sales promotions are excellent

E2 You feel good participating in a brand’s sales promotion activities

E3 You think the discounts that are offered in a brand’s sales promotions are attractive

Brands loyalty – F

F1 You often purchase the products of a particular brand F2 A brand that you are familiar with would be your

priority when you want to buy something

F3 You would not consider other brands if you can buy a brand you like

Perceived quality – G

G1 The quality of products is the reason that you purchase a brand’s product

G2 The performance of product is the reason that you purchase a brand’s product

Brand association – H

H1 You are usually familiar with a brand before purchasing its products

H2 It is easy for you to identify a brand’s products H3 You can remember the features of a brand among

(25)

20 H4 You think the logo or slogan of a brand is easy for you

to remember

To conclude, the survey consisted of three parts. The first part introduced the subject and purpose of the survey, and the principles of privacy protection. The second part focused on the respondents’ demographic characteristics and their social media use. The third part was the main part of the survey that focused on the respondents’ experiences regarding the mobile game companies’ social media marketing activities, and measuring brand equity from the respondent’s perspective. In addition, the survey was designed using closed-ended questions to ensure the efficiency of the investigative activities and limit the risk that any potential errors may occur during the data collection and analysis. The survey was designed in English, and has been translated to Mandarin Chinese very carefully. The completed survey is shown in Appendix 2.

3.7 Pilot study

In order to develop a high-quality survey, a pilot study was conducted to evaluate the survey. The researchers have invited 100 students from Shenyang Aerospace University through social media to answer the survey. According to the respondents’ feedback, the researchers have corrected and excluded the unclear and improper words in the questionnaire, so that the authenticity of the data can be further guaranteed. Based on the results, an item analysis has been carried out in order to test the validity and rationality of each item in survey questions 8 and 9 as the Appendix 3 shows. Then, the Table 4 displays the results of the item analysis, and we can see that all items in survey questions 8 and 9 have passed the T-test and have a high significance level of discrimination. Which means that all items in survey questions 8 and 9 are valid and rational. Therefore, all these items should be accepted.

(26)

21 Table 4: Results of Item Analysis

Question Item p Decision Question Item p Decision A1 0.000 Accept D4 0.000 Accept A2 0.000 Accept E1 0.000 Accept A3 0.000 Accept E2 0.000 Accept B1 0.000 Accept E3 0.000 Accept B2 0.000 Accept F1 0.000 Accept B3 0.000 Accept F2 0.000 Accept B4 0.000 Accept F3 0.000 Accept B5 0.000 Accept G1 0.000 Accept C1 0.000 Accept G2 0.000 Accept C2 0.000 Accept H1 0.000 Accept C3 0.000 Accept H2 0.000 Accept D1 0.000 Accept H3 0.000 Accept D2 0.000 Accept H4 0.000 Accept D3 0.000 Accept

After the item analysis, it was also necessary to conduct a reliability analysis regarding the survey questions 8 and 9 to test if the survey questions were representative and reliable. In reliability statistics, the reliability of an item is defined by Cronbach α, higher value of α representing higher level of internal consistency for the survey items. As the Table 5 shows, the Cronbach α equals 0.962, which indicates that the data collected in the pilot study has a high level of reliability. Furthermore, regarding the Cronbach Alpha if Item Deleted, the Cronbach α will not increase if any item was deleted, therefore, all items in survey questions 8 and 9 should be retained.

(27)

22 Table 5: Results of Reliability Statistics (Cronbach Alpha)

Item Cronbach Alpha if Item Deleted Cronbach α Item Cronbach Alpha if Item Deleted Cronbach α A1 0.960 0.962 D4 0.959 0.962 A2 0.960 E1 0.960 A3 0.960 E2 0.960 B1 0.960 E3 0.960 B2 0.960 F1 0.960 B3 0.960 F2 0.960 B4 0.960 F3 0.961 B5 0.960 G1 0.960 C1 0.960 G2 0.961 C2 0.960 H1 0.961 C3 0.960 H2 0.961 D1 0.960 H3 0.961 D2 0.960 H4 0.961 D3 0.959 3.8 Survey conduction

The survey was conducted over the Tencent Survey platform, a professional Chinese online survey platform developed by the Tencent Customer Research & User Experience Design Center (CDC). This platform allows users to design, conduct, and monitor an entire survey project with user friendly interface, the platform has 516.79 million users in 2020 (Tencent Survey, 2020). Through the platform’s online survey distribution system, the survey was automatically and randomly distributed to the internet users in China. The respondents were allowed to answer the survey on multiple types of devices including iOS, Android, and PC. The invalid responses have been automatically rejected and excluded by the system to ensure the reliability of the survey results (Tencent Survey, 2020). In summary, the survey received 1648 views and 1089 completed responses excluding invalid responses, which representing a 66% response rate.

(28)

23

4. Data analysis and results

The quantitative data gathered from the survey were analyzed with SPSS 20.0. In this research, the data analysis can be divided into four section. The first section focusing on statistical analysis concerns respondents’ demographic backgrounds and their usage of social media. The second part focusing on descriptive analysis concerns respondents’ evaluation of mobile game company’s social media marketing activities and brand equity, it provides valuable data for the regression analysis. The third part is the Pearson analysis that aims to test the linear relationships between independent variables and dependent variables. The last part using regression analysis to test the research hypotheses and measure the effects of independent variables on dependent variables.

4.1 Statistical analysis

4.1.1 Respondents’ backgrounds

As Table 6 display, the proportion of men in the sample was almost equal to the women. Specifically, males accounted for 56.2% of respondents and females for 43.8%. Meanwhile, those respondents aged 19-29 accounted for the largest proportion, equal to 53.6%. The respondents aged between 30 and 39 accounted for 26.8%, meaning that the primary group of users of social media are those users between 19-39 years old. Furthermore, the educational backgrounds of most social media users are high school and undergraduate degrees.

Table 6: Respondent backgrounds

Element Frequency Percent

(%) Cumulative Percent (%) Gender Female 477 43.8 43.8 Male 612 56.2 100.0

(29)

24 Age Group 18 and below 122 11.2 11.2 19-29 584 53.6 64.8 30-39 292 26.8 91.6 40 and above 91 8.4 100.0 Educatio n

Middle School and Below 150 13.8 13.8

High School 439 40.3 54.1

Bachelor’s degree 473 43.4 97.5

Master’s degree and above 27 2.5 100.0

4.1.2 Respondents’ social media usage

As can be seen from Table 7, 47.5% of the respondents have used social media for 1-2 years, while second largest proportion was 21.5% for 2-3 years. Meanwhile, those who have used social media for less than a year and those who have used it for more than three years accounted for 16.3% and 14% respectively. Regarding the frequency of daily social media use, more than half of the respondents use social media 3-4 times a day on average, representing 51.6%. The proportion of 1-2 times was 22.6%, and only 8.5% of respondents use it more than 6 times a day. In addition, 44.6% of respondents spend 30 to 60 minutes a day on social media, and 27.7% of respondents spend about 1-2 hours. Only 4.4% of the respondents use social media for more than three hours a day.

(30)

25 Table 7: Social media usage

Element Frequency Percent (%) Cumulative

Percent (%) How long have you been using social media

Less than a year 178 16.3 16.3

1-2 517 47.5 63.8

2-3 234 21.5 85.3

3 years and above 160 14.7 100.0

The frequency of social media use per day 1-2 times 246 22.6 22.6 3-4 times 562 51.6 74.2 5-6 times 188 17.3 91.5 above 6 times 93 8.5 100.0 The amount of time spent on social media per day Less than 30 minutes 182 16.7 16.7 30- 60 minutes 486 44.6 61.3 1-2 hours 302 27.7 89.1 2-3 hours 71 6.5 95.6 above 3 hours 48 4.4 100.0

Table 8 displays that the most popular social media platform is TikTok, which accounts for 45.2% of the respondents. This reflects the rising developmental trend of the short video industry. Weibo, also known as a Chinese version of Twitter, accounted for 20.0%,

(31)

26 and Bilibili was mostly used by 13.8% of respondents. In addition, a small number of respondents mostly use Baidu Tieba and Zhihu.

Table 8: Popularity of social media platform Type of Social Media Frequency Percent (%) Cumulative Percent (%) Weibo 218 20.0 20.0 TikTok 492 45.2 65.2 Bilibili 150 13.8 79.0 ZhiHu 99 9.1 88.1 Baidu Tieba 75 6.9 94.9 Others 55 5.1 100.0

4.2 Descriptive statistical analysis

As shown in Appendix 4, the respondents’ evaluation of each variable item was measured based on the results of survey questions 8 and 9. The minimum and maximum values of each item are 1 and 5, which respectively represents "strongly disagree" and "strongly agree" to a specific statement.

The value of mean is a representative of the concentration measure, and it is a widely used statistical indicator that reflects the representation of data. It can be used to explain the concentration or representability of the data in an intuitive way. The mean value occupies an important position in most current statistical methods. Standard Deviation measures the dispersion of a set of data, and it is the square root of variance.

4.2.1 Brand popularity

The authors proposed three questions regarding the brand popularity variable. Table 9 shows the analysis results of the responses including the mean value, standard deviation and variance of each item, as well as the overall values of the variable. This provided

(32)

27 valuable data for the correlation and regression analysis. Since brand popularity is an independent variable in the research model, the mean and standard deviation of the overall composition have been analysed. We can see that the mean is close to 3.511 with a standard deviation of 0.803.

Table 9 Descriptive statistical analysis of Brand popularity

Variable Item N Min Max Mean Std. Dev. Variance

Brand popularity A1 1089 1 5 3.637 1.096 1.202 A2 1 5 3.432 0.988 0.975 A3 A 1 5 3.463 3.511 1.028 0.803 1.058 0.646 4.2.2 Information quality

The authors proposed five questions regarding the information quality variable. The analysis results including the value of mean, standard deviation, variance and the overall value of the variable are displayed in the Table 10. Since information quality is an independent variable in the research model, the mean and standard deviation of the overall composition have been analysed. We can see that the mean is close to 3.533 with a standard deviation of 0.743.

Table 10: Descriptive statistical analysis of Information quality

Information quality B1 1089 1 5 3.567 1.078 1.163 B2 1 5 3.552 1.067 1.139 B3 1 5 3.497 1.042 1.085 B4 1 5 3.529 1.048 1.099 B5 B 1 5 3.520 3.533 1.059 0.743 1.121 0.551 4.2.3 Involvement of influencer

The authors proposed three questions regarding the involvement of influencer variable. The analysis results including the value of mean, standard deviation, variance and the overall value of variable are displayed in the Table 11. Since involvement of influencer is an independent variable in the research model, the mean and standard deviation of the

(33)

28 overall composition have been analysed. We can see that the mean is close to 3.503 with a standard deviation of 0.831.

Table 11: Descriptive statistical analysis of involvement of influencer

Engagement of

opinion leader C1 1089 1 5 3.497 1.074 1.153

C2 1 5 3.510 1.086 1.180

C3 ₁ ₅ 3.503 1.062 1.129

C 3.503 0.831 0.690

4.2.4 Interaction with fans

The authors proposed four questions regarding the interaction with fans variable. The analysis results including the value of mean, standard deviation, variance and the overall value of the variable are displayed in the Table 12. Since interaction with fans is an independent variable in the research model, the mean and standard deviation of the overall composition have been analysed. We can see that the mean is close to 3.526 with a standard deviation of 0.763.

Table 12: Descriptive statistical analysis of Interactivity with fans

Interactivity with fans D1 1089 1 5 3.493 1.041 1.083

D2 1 5 3.560 1.084 1.175

D3 1 5 3.556 1.001 1.003

D4 ₁ ₅ 3.494 1.060 1.123

D 3.526 0.763 0.582

4.2.5 Intensity of Sales Promotion Activities

The authors proposed three questions regarding the intensity of sales promotion activities variable. The analysis results including the value of mean, standard deviation, variance and the overall value of variable are displayed in the Table 13. Since intensity of sales promotion activities is an independent variable in the research model, the mean and

(34)

29 standard deviation of the overall composition have been analysed. We can see that the mean is close to 3.554 with a standard deviation of 0.816.

Table 13 Descriptive statistical analysis of Intensity of Sales Promotion Activities

Intensity of sales promotion activities E1 1089 1 5 3.524 1.083 1.172 E2 1 5 3.543 1.081 1.168 E3 ₁ ₅ 3.596 0.981 0.962 E 3.554 0.816 0.666 4.2.6 Brand Loyalty

The authors proposed three questions regarding the brand loyalty variable. Table 14 shows the analysis results of responses including the mean, standard deviation and variance of each item, as well as the overall values of the variable. Since brand loyalty activities is a dependent variable in the research model, the mean and standard deviation of the overall composition have been analysed. We can see that the mean is close to 3.518 with a standard deviation of 0.865.

Table 14 Descriptive statistical analysis of Brand Loyalty

Brand loyalty F1 1089 1 5 3.589 1.137 1.292

F2 1 5 3.435 1.017 1.035

F3 ₁ ₅ 3.530 1.097 1.203

F 3.518 0.865 0.748

4.2.7 Perceived quality

The authors proposed two questions regarding the perceived quality variable. Table 15 shows the analysis results of responses including the mean, standard deviation and variance of each item, as well as the overall values of the variable. Since perceived quality activities is a dependent variable in the research model, the mean and standard deviation

(35)

30 of the overall composition have been analysed. We can see that the mean is close to 3.545 with a standard deviation of 0.884.

Table 15 Descriptive statistical analysis of Perceived quality

Perceived quality G1 1089 1 5 3.568 1.036 1.073

G2 ₁ ₅ 3.522 1.063 1.130

G 3.545 0.884 0.748

4.2.8 Brand Association

The authors proposed four questions regarding the brand association variable. Table 16 shows the analysis results of responses including the mean, standard deviation and variance of each item, as well as the overall values of the variable. Since brand association is a dependent variable in the research model, the mean and standard deviation of the overall composition have been analysed. We can see that the mean is close to 3.552 with a standard deviation of 0.784.

Table 16 Descriptive statistical analysis of Brand Association

Brand association H1 1089 1 5 3.507 1.069 1.142

H2 1 5 3.536 1.059 1.122

H3 1 5 3.574 1.032 1.065

H4 ₁ ₅ 3.592 1.039 0.080

(36)

31 4.3 Pearson correlation analysis

In this research, a Pearson correlation analysis was conducted to measure the degree of linear relationship (correlation) between the independent variables and dependent variables. As Table 17 shows, the results of Pearson correlation analysis indicate that brand loyalty is significantly correlated with the five dimensions of social media marketing, and the correlation coefficients all pass the significance test. Furthermore, perceived quality and brand association were also found to be significantly correlated with the five dimensions of social media marketing. Pearson correlation coefficient reflects the degree of linear correlation between two variables, and its value is between -1 and -1. When the linear relationship between two variables is enhanced, the correlation coefficient tends to be 1 or -1. If one variable increase and the other variable also increases, it means that they are positively correlated, and the correlation coefficient is greater than 0. If one variable increase and the other variable decreases, it means that they are negatively correlated, and the correlation coefficient is less than 0. If the correlation coefficient is equal to 0, there is no linear correlation between them. For example, the correlation coefficient between brand loyalty and brand popularity is 0.671**, which passes the significance level test. It reflects that brand loyalty is significantly positively correlated with brand popularity.

(37)

32 Table 17: Pearson correlation analysis results

Brand loyalty Perceived quality Brand Association Brand popularity Information quality Involvement of influencer Interaction with fans Intensity of Sales Promotion Activities Perceived quality Pearson Correlation .636** 1 Sig. (2-tailed) 0.000 Brand Association Pearson Correlation .721** .655** 1 Sig. (2-tailed) 0.000 0.000 Brand

popularity Correlation Pearson .671** .489** .585** 1 Sig. (2-tailed) 0.000 0.000 0.000 Information quality Pearson Correlation .639** .594** .649** .728** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 Involvement of influencer Pearson Correlation .631** .491** .577** .693** .671** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 0.000 Interaction

with fans Correlation Pearson .625** .570** .614** .706** .795** .683** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 0.000 0.000 Intensity of Sales Promotion Activities Pearson Correlation .588** .501** .598** .679** .706** .703** .702** 1 Sig. (2-tailed) 0.000 0.000 0.000 0.000 0.000 0.000 0.000

4.4 Regression analysis and findings

Regression analysis is the most important part in data analysis, it enables researchers to test hypotheses and answer research questions by providing statistical evidence. Therefore, the researchers have defined the regression model for each element of brand equity based on the formal multiple linear regression model below, where y represents

(38)

33 dependent variables and x represents independent variables. β0 is a constant term and βi

represents the regression coefficients, εi is the residual item.

𝑦 = 𝛽 + 𝛽 𝑥 + 𝛽 𝑥 + ⋯ 𝛽 𝑥 + 𝜀 The regression model can be defined as following:

𝐵𝑟𝑎𝑛𝑑 𝑒𝑞𝑢𝑖𝑡 = 𝛽 + 𝛽 ∗ 𝐵𝑟𝑎𝑛𝑑 𝑝𝑜𝑝𝑢𝑙𝑎𝑟𝑖𝑡𝑦 + 𝛽 ∗ 𝐼𝑛𝑓𝑜𝑟𝑚𝑎𝑡𝑖𝑜𝑛 𝑞𝑢𝑎𝑙𝑖𝑡𝑦 + 𝛽 ∗ 𝐼𝑛𝑣𝑜𝑙𝑣𝑒𝑚𝑒𝑛𝑡 𝑜𝑓 𝑖𝑛𝑓𝑙𝑢𝑒𝑛𝑐𝑒𝑟𝑠 + 𝛽 ∗ 𝐼𝑛𝑡𝑒𝑟𝑎𝑐𝑡𝑖𝑜𝑛 𝑤𝑖𝑡ℎ 𝑓𝑎𝑛𝑠 + 𝛽 ∗ 𝐼𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦 𝑜𝑓 𝑆𝑎𝑙𝑒𝑠 𝑃𝑟𝑜𝑚𝑜𝑡𝑖𝑜𝑛 𝐴𝑐𝑡𝑖𝑣𝑖𝑡𝑖𝑒𝑠 + 𝜀

4.4.1 Regression analysis concerning brand loyalty

The Table 18 shows the influence of social media marketing on brand loyalty. The regression results indicate that the R square is 0.531, which reflects that the research model can explain the 53.1% changes of brand loyalty. F (245.663) shows that the regression model passed the significance level test. Collinearity test results show that VIF values are all less than 10, indicating that there is no collinearity problem in the regression analysis.

The regression coefficient of brand popularity is 0.325 and the p-value is 0.000, which passed the significance level test. It reflects that if brand popularity increases by one unit then the brand loyalty will increase by 0.325. At the meantime, the regression coefficients of information quality, the involvement of influencer and the interaction with fans all showed that they passed the significance test.

4.4.1.1 Findings regarding brand loyalty

The regression analysis found that brand popularity, information quality, involvement of influencer and interaction with fans are positively affect brand loyalty. However, this study did not find that intensity of sales promotion activities significantly affects brand

(39)

34 loyalty. Since its regression coefficient is 0.053 and the p-value is 0.144, which is greater than 0.05 and failing the significance level test. Hence, we rejected the H13: Intensity of

sales promotion activities has positive impact on brand loyalty.

According to the standardized regression coefficient, among the five dimensions of social media marketing, brand popularity (0.302) has the greatest impact on brand loyalty, followed by the involvement of influencer (0.203), the interaction with fans had the least impact on brand loyalty (0.111).

Table 18: Regression analysis results (brand loyalty) Unstandardized Coefficients Standardized Coefficients _t _Sig. Collinearity Statistics

B Std. Error Beta Tolerance VIF

(Constant) 0.347 0.094 3.697 0.000 Brand popularity 0.325 0.037 0.302 8.820 0.000 0.368 2.715 Information quality 0.186 0.045 0.160 4.150 0.000 0.293 3.415 Involvement of influencer 0.212 0.035 0.203 6.128 0.000 0.393 2.544 Interaction with fans 0.126 0.043 0.111 2.936 0.003 0.304 3.294 Intensity of Sales Promotion Activities 0.053 0.036 0.050 1.463 0.144 0.377 2.655

a. Dependent Variable: Brand loyalty

R Square 0.531

(40)

35 4.4.2 Regression analysis concerning perceived quality

The Table 19 shows the influence if social media marketing on perceived quality. The value of R square indicate that the research model can explain 38.6% changes of brand perceived quality. F (136.075) shows that the regression model passed the significance level test.

The regression results show that information quality, involvement of influencer and interaction with fans are passed the significance test. For example, the regression coefficient of information quality is 0.386, and the p-value is 0.000, which is less than 0.05 and passed the significance level test. Thus, if the information quality increase by one unit, then the perceived quality will increase by 0.386.

4.4.2.1 Findings regarding perceived quality

The regression analysis shows that information quality, involvement of influencer, and interaction with fans are positively affect perceived quality. However, it was not found that brand popularity and Intensity of Sales Promotion Activities significantly affect perceived quality. Since none of the regression coefficients passed the significance level test. Hence, we rejected H2 and H14.

The standard regression coefficient reflects that information quality (0.325) has the greatest impact on perceived quality, while the variable with the second-most impact on perceived quality is interaction with fans (0.204). The involvement of influencer (0.082) has the least impact on perceived quality

(41)

36 Table 19: Regression analysis results (perceived quality)

Unstandardized

Coefficients Standardized Coefficients t Sig.

Collinearity Statistics

B _ErrorStd. Beta Tolerance VIF

(Constant) 0.759 0.110 6.908 0.000 Brand popularity 0.009 0.043 0.008 0.206 0.837 0.368 2.715 Information quality 0.386 0.052 0.325 7.380 0.000 0.293 3.415 Involvement of influencer 0.088 0.040 0.082 2.171 0.030 0.393 2.544 Interaction with fans 0.236 0.050 0.204 4.716 0.000 0.304 3.294 Intensity of Sales Promotion Activities 0.070 0.042 0.065 1.670 0.095 0.377 2.655

a. Dependent Variable: Perceived quality

R Square 0.386

F 136.075

4.4.3 Regression analysis concerning brand association

As shown in Table 20, the research model can explain 48.3% changes of brand association since R square equals 0.483. The F value is 202.717, which reflects that the regression model has passed the significance level test.

4.4.3.1 Findings regarding brand association

The regression results show that the five dimensions of social media marketing are positively related with brand association, since their regression coefficients all passed the significance test. The standard regression coefficient shows that the information quality (0.288) has the greatest influence on brand association, with the second-most being the

(42)

37 intensity of sales promotion activities (0.155). Brand popularity had the least impact on brand association (0.099).

Table 20: Regression analysis results (brand association) Unstandardized Coefficients Standardized Coefficients t Sig. Collinearity Statistics

B _ErrorStd. Beta Tolerance VIF

(Constant) 0.760 0.089 8.492 0.000

Brand popularity 0.097 0.035 0.099 2.753 0.006 0.368 2.715 Information quality 0.304 0.043 0.288 7.128 0.000 0.293 3.415 Involvement of

influencer 0.115 0.033 0.122 3.507 0.000 0.393 2.544

Interaction with fans 0.126 0.041 0.123 3.103 0.002 0.304 3.294 Intensity of Sales

Promotion

Activities 0.149 0.034 0.155 4.359 0.000 0.377 2.655

a. Dependent Variable: Brand Association

R Square 0.483

How does social media marketing affect mobile game companies’ brand equity in China?