Whether and how the use of different social media affects consumer attitudes and purchase intentions for different product categories?

Master Thesis for the Master of Science in Business Administration

“Whether and how the use of different social media affects

consumer attitudes and purchase intentions for different

product categories?”

Student: Han Xie, 10887628

Study: Master of Science in Business Administration, Marketing Track

Email: shaniahan0126@126.com

Supervisor: Dr. Umut Konuş Word count: 11,052

Faculty: Faculty of Economics and Business Pages:69

This document is written by Student Han Xie who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Signed: Han Xie  

Date:  26/06/2015    

In this thesis I investigate whether and how the use of different Chinese social media Sina Weibo (micro-blog) and WeChat (social media messaging app) affects consumer attitudes and purchase intentions for different product categories, namely clothing, electronic devices, and flight tickets. Besides, I also examine how the purchase intention is affected by demographic traits and psychographic traits. The results find that generally consumers have a higher purchase intention on WeChat platform rather than Sina Weibo plaftorm. Besides, on both platforms, consumers exhibit higher purchase intention for the clothing category than in the electronic device and flight ticket category. I also find that, on both platforms, consumers with age above 30 have higher purchase intention. In addition, studies find that innovativeness and E-WOM have strongly positive impact on consumer purchase intention on both platforms, but the economic magnitudes on two platforms are very close. In my sample, impulsiveness seems to have no impact on both platforms.

Keywords: Purchase intention, Social Media, Sina Weibo, WeChat

2.1 Overview of social media  ...  6  

2.1.1  Introduction  of  Social  media  ...  6  

2.1.2  Social  media  marketing  and  purchase  intention  ...  6  

2.1.3  The  emergence  of  Chinese  social  media  platforms  ...  8  

2.1.3.1  Sina  Weibo:  A  Chinese  micro-­‐blog  ...  8  

2.1.3.2  WeChat:  A  Chinese  social  messaging  app  ...  9  

2.1.3.3  A  comparison  of  Sina  Weibo  and  WeChat  ...  11  

2.2 Demographic and psychographic variables  ...  13  

2.3 Electronic word-of-mouth variables  ...  14  

2.4 Formulate the research question  ...  16  

3.   CONCEPTUAL FRAMEWORK  ...  16  

3.1 The Conceptual Map  ...  16  

3.2 Hypothesis Development  ...  17  

3.2.1  Hypotheses  of  different  social  media  platforms  and  product  categories  ...  18  

3.2.2  Hypotheses  of  demographic  traits  ...  19  

3.2.3  Hypotheses  of  psychographic  traits  ...  20  

3.2.4  Hypotheses  of  Electronic  word-­‐of-­‐mouth  ...  21  

4. METHODOLOGY  ...  23  

4.1 The sample collection  ...  23  

4.2 Survey design  ...  23  

5. EMPIRICAL RESULTS  ...  25  

5.1 Missing value and recording  ...  25  

5.2 Reliability test  ...  25  

5.3 Scale Means and correlation check  ...  26  

5.4 Hypotheses testing  ...  28  

5.4.1  Paired  sample  t-­‐test  of  H1  ...  28  

5.4.2  Independent  sample  t-­‐test  of  H2  ...  30  

5.4.3  Regression  analysis  of  H3  ...  34  

5.4.4  Regression  analysis  of  H4  ...  35  

6. DISCUSSION  ...  38  

6.1 Discussion of Hypotheses  ...  38  

6.1.1  Discussion  of  Hypotheses  1  ...  38  

6.1.2  Discussion  of  Hypotheses  2  ...  40  

6.1.3  Discussion  of  Hypotheses  3  ...  42  

6.1.4  Discussion  of  Hypotheses  4  ...  44  

6.2Managerial Implication  ...  46  

6.3 Limitation  ...  48  

7. CONCLUSION  ...  50  

References ... 53

1. INTRODUCTION

Social media has emerged following the innovation and development of the Internet and mobile app technology. In recent years, the social media buying behavior of consumers have led to businesses identifying the enormous marketing potential of social media platforms (Lun 2012). In contrast to traditional face-to-face sales, e-commerce on social media platforms is conducted through SNS, blogging, micro-blogging and social mobile technology (Li 2012). As enterprises utilise this new form of marketing, social media is assuming a central role in the online activities of an increasing proportion of businesses [Mengni 2012]. Arguably, social media presence is now firmly established as a key factor in shaping consumer purchase intention (Hsu, Yen, Chiu 2006). Numerous studies have focused on the issue of purchase intention. (Brown, Pope, Voges, 2003). They assert that customer behavioural intentions can be regarded as signals of actual purchasing choice and therefore require analysis (Bughin et al., 2007). As such, investigating consumer purchase intention on social media platforms may help to accurately predict consumer preferences, which would enable marketers to devise an effective product promotion strategy (Lenhart, Purcell, Smith, Zickuhr, 2010).

The shift to a socialist market economy and development of digital and online technology has culminated in widespread changes in China’s social media marketing environment (Lu & Weber, 2007). With the world’s largest online population, the

country has also become one of the largest online retailing markets in the world (Lee, 2009; Kan, 2012). By 2009, those with experience of shopping online accounted for approximately a quarter of the country’s population (Lee 2009). By the end of 2011, there were 308 million social media users in the country while the internet penetration rate had reached 38.4 percent, thereby exceeding the global average (Internet World Stats 2012). Furthermore, the Boston Consulting Group predicted that, by 2015, the Chinese online market will achieve sales volumes of $364 billion. Such statistics hint that China breed a great potential business opportunities in the social media marketing environment now, and to exploit the great potential opportunities it is important for enterprise to analyze consumer purchase intention on Chinese social media platforms.  

However, as international social media platforms such as Facebook, YouTube, and Twitter are blocked in China, enterprises cannot afford to overlook consumer purchase intention with regards to the social media marketing environment or risk losing competitive advantage. In addition, to fully understand the purchase intention and attitude of Chinese consumers and to apply appropriate promotion strategies on Chinese social media platforms seems difficult because of the culture difference and language barrier (Li et al. 2008). Thus, this thesis Thus, this thesis will present research on the two largest Chinese social media platforms: the Chinese micro-blog platform, Sina Weibo, and the Chinese social messaging app, WeChat. In doing so, the objective is to identify their main features and differences, and investigate whether and how the use of different social media platforms affects consumer purchase

intention and attitudes, hoping to draw some insights and managerial implications from the study of Chinese social platform.

At present, Chinese social media platforms are not as widely used for company promotion strategies as they should be. This can be attributed to an absence of research on this topic which would help to guide managers and decision-makers (Haejung Kim, Jiyoung Kim, Ran Huang 2014). Existing studies have a tendency to focus on Chinese social media platforms, particularly Sina Weibo and WeChat. Yet, the differences between international and Chinese social media platforms are rarely acknowledged. While some studies have suggested that Sina Weibo is essentially the Chinese Twitter, this assessment fails to recognize the richer multimedia functionalities of the former. More convincing is that Sina Weibo is the Chinese equivalent of Twitter incorporating aspects of both Tumblr and Instagram (Kaplan & Haenlein 2010). Moreover, Sina Weibo’s ‘expanded blog feature’ enables users to post far more information than is possible in a 140-character tweet. Meanwhile, WeChat can be considered a combination of Whatsapp and Facebook. Nevertheless, differences still exist, such as WeChat’s heightened privacy options, while the primary function is communications between friends rather than a public broadcasting platform.

This thesis use survey approach to study whether and how different Chinese social media platforms impact consumer buying attitude and intention in China. The sample

is obtained from sending online questionnaires to both Sina Weibo users and Wechat users. The questionnaire consists of three parts, 37 questions in total. In the first part, 22 questions are included to ask respondents about their purchase attitude on Wechat and Sina Weibo respectively. The second part measures the moderators such as psychographic traits, and electronic word-of-mouth and the last part measures the respondents’ demographics characteristics such as income, age, gender, and etc.

In addition to the study of overall purchase intention, I break down the consumer purchases intention into three different product categories, namely clothing, flight tickets and electronics, to investigate which product category attracts consumer purchase intention the most. I use demographic traits such as age and gender, and psychographic traits such as impulsiveness, innovativeness, and electronic word of mouth to explain the variation of consumer purchase intention.

Related to existing literature, my findings show that consumers indeed exhibit systematically different purchase intentions on Wechat and Weibo platforms. Liao and Cheung (2001) argue that different functions of online shopping platforms affect consumer purchases behavior differently. I find that consumers in my sample have significant higher purchase intention on Wechat platform, a finding which might be explained by the ease of payment function embedded in Wechat. Furthermore, consistent with Slama & Tashchian (1985) who stated that consumer decision process,

results show that ages do affect the consumer purchase intention, but, except for the clothing category, I do not find any significant difference of purchase intention in genders. Last, psychographic traits such as innovativeness and E-WOM have strong positive impact on the consumer purchase intention, results which are consistent with Donthu & Garcia (1999) and Wells & Chen (1999), but impulsiveness seems to have no effect on the purchase intention.

Overall, the contribution of this paper is to provide insights for marketers to understand how consumer purchase intention varies on Chinese social media platforms, and how demographic traits, psychographic traits and electronic word-of-mouth influence the consumer purchase intention on both of them. The results of this study will help companies to understand consumer purchase intention on different Chinese social media platforms and help marketers find effective marketing strategies

The rest of my thesis is organized as following: section 2 reviews the relevant literature in detail, section 3 presents the conceptual framework and proposes the hypotheses, in section 4 and section 5 I explain the methodology and exhibit the empirical results, respectively, discussion of the empirical results, managerial implication, and limitation are reported in the section 6, and finally section 7 concludes.

2. LITERATURE REREVIEW

2.1 Overview of social media

2.1.1 Introduction of Social media

Social media has been defined as ‘a group of internet-based applications that build on the ideological and technological foundations of Web 2.0 which allowed the creation and exchange of User Generated Content (Kaplan & Haenlein, 2010). Thus, social media enables consumers to engage in the production, consumption and exchange of information online (Pergolino, Rothman & Miller, 2012). Meanwhile, tools facilitating social media marketing include micro-blogs, blogs, social network sites (SNS) and social mobile apps. Using different channels, consumers can ‘post’, ‘tag’, ‘digg’ and ‘blog’ to share their opinions and preferences. Accordingly, social media provides a platform for the creation and circulation of content that can be used by consumers to inform others about products, brands, services and issues (Blackshaw & Nazzaro, 2006). As an increasing number of users are utilizing this‘ collective intelligence’ enabled by the Internet, it represents a serious challenge to traditional marketing practices (Litvin, Goldsmith & Pan, 2008). At present, social media serves an array of different functions, while its global popularity continues to rise. Nevertheless, in China, several mainstream social media platforms are blocked by the Chinese government, including Facebook and YouTube. Consequently, Chinese equivalents of banned western social media platforms have emerged.

Social media marketing (SMM) is an online marketing model which aims to realise marketing objectives through activity on social media networks (D’Silva et al., 2011, p. 756). The importance of using social media for product marketing has increased dramatically with the development of social media platforms. As social media has a considerable influence on online purchasing, promotions and advertisements are usually featured on social media websites. Meanwhile, one of the fundamental benefits of social media advertising is that companies can access users’ demographic information to achieve their desired promotion strategy. Moreover, companies are able to communicate with their existing and potential customers at all times and learn about their preferences. Thus, this data can be used to make upgrades to products and services in order to enhance customer satisfaction (D’Silva, et al., 2011, p. 756).

In order to determine what makes a valuable promotion to consumers, it is necessary to discover the factors underlying user preferences in terms of social media. Yet, the business press and existing research offers limited guidance for marketing managers seeking to introduce social media into their marketing strategies. Therefore, many managers are unaware of the potential benefits of using social media in promotional strategies. Although social media enhances the impact consumer-to-consumer communications have on the market, this areas requires further research (Mangold & Faulds, 2009)

response behavior (Li Daugherty & Biocca, 2002). In order to understand consumer buying preference on social media platforms it is essential to understand consumer purchase behavior and the decision processes (Solomon, 2008). For instance, social media users are often attracted by the ease of accessing information, convenient purchasing process and payment methods, warranties and the option to purchase later (D’Silva et al., 2011, p. 756). Consequently, if there are two social media platforms with different features, the consumer purchasing decision will differ. Nonetheless, there is an absence of research exploring whether buying intention varies on different social media platforms. Therefore, in this research, the two Chinese largest social media platforms will be analyzed to determine whether disparities exist in the purchase intention.

2.1.3 The emergence of Chinese social media platforms 2.1.3.1  Sina  Weibo:  A  Chinese  micro-­‐blog  

Weibo is a hugely popular Chinese micro-blog and broadcast medium. Like its western competitors, such as Twitter, users are able to publish posts of up to 140 characters with attachments, including images and video hyper links (Gao, Abel et al., 2012). In recent years, its popularity has risen sharply and user numbers have increased dramatically. At present, China’s three most popular micro-blog service providers are Sina, Sohu and Tencent (Website of China Daily, 2014; Eley & Tilley 2009, p. 82).

In August 2009, one of Chinese largest Weibo portals, Sina Corporation, launched Sina Weibo. Statistics released by Sina Corporation indicate that daily active users rose by 2.5 million in the last quarter of 2013, whilst numbers increased by 4.2 percent from 58.9 million to 61.4 million in December (Gao, Abel et al., 2012). Estimates suggest that there are over 500 million to registered users (Website of Seeking Alpha, 2014; Website of Weibo, 2014.) User profiles feature the individual’s name, a brief description, entails about the number of followers and fans that a user has and the number of posts that the user has made. Furthermore, there are two types of accounts: regular and verified user accounts. Verified user accounts, which are subject to censorship by the website service group, are mainly held by celebrities, athletes, famous and other public figures. Tweets contain text, videos, images and hyper links. There is an interactive element, as users can ‘like’, comment, save or repost tweets. Moreover, there is also a dialogue mode that enables users to have private conversations. (Website of Seeking Alpha, 2014;Website of Weibo,2014).

2.1.3.2  WeChat:  A  Chinese  social  messaging  app  

WeChat is a Chinese instant messaging app available for smart phones. Its primary function is to enable users to communicate with friends and groups effectively. By 2014, WeChat had over 600 million active users and 100 million international registered users, despite focusing its marketing efforts on mainland China (Website of

Tencent, 2014; Websiteof Chinese Social Media101,2014).

Some unique features distinguish WeChat from its western equivalent, WhatsApp. Nevertheless, many similar features are shared. For example, the ‘video call’ function in WeChat is similar to the FaceTime function offered by Apple. Another feature available to WeChat users, ‘moments’, is also referred to as ‘friend circle’ in Chinese. This enables users to share text, images and hyper links or comment on each post. However, all comments are only view able by reciprocal friends in order to protect users’ privacy. (WebsiteofTencent2014; Website of Chinese Social Media101, 2014).

Three other features of WeChat, ‘shake’, ‘look around’ and ‘drift bottle’, enable users to expand their friend network. By shaking their phone, users can make new friends if other users are shaking their phone at the same time. This feature bridges gaps in terms of the location of users. Users are able to search for friends located nearby by using the ‘look around’ function. Meanwhile, the ‘drift bottle’ function enables users to throw a virtual bottle containing a message into the imaginary sea which another user can catch and either respond or throw it back into the sea. Thus, it represents an anonymous exchange between strangers. Importantly, WeChat has its own WeChat shop (micro-shop), consumers could buy the product on the shop through the feature of WeChat payment (wallet pay), which means consumer could pay their order directly through WeChat (Website of Tencent, 2014; Website of Chinese Social Media101, 2014).

2.1.3.3  A  comparison  of  Sina  Weibo  and  WeChat  

It is apparent that ‘precision’ is the most significant difference between Sina Weibo and WeChat. Specifically, when a Sina Weibo user updates a post, the post is broadcasted to all his followers when they refresh the news feeds. However, when Weibo user refreshes, she receives a stream of newsfeeds which are sorted on the updating time. Thus, she might not necessarily read every newsfeeds posted by others whom she follows, because the outdated news feeds might be pushed to the bottom if many users update lately. In contrast, if a Wechat user follows a public account1, the public account can push notification periodically to remind Wechat user that there is an unread update.

WeChat users have a conversational relationship, as two users must first add each other as friends on the app in order to make the relationship reciprocal. However, regular Weibo users do not need to add each other as friends. This represents a multi way relationship rather than a reciprocal one. Therefore, WeChat offers ‘close-loop communication’ in a private space that differs from the ‘open transmission’ offered by Weibo (Gao, Abel et al., 2012; Lien, Cao et al., 2014). WeChat can be described as inward, private and communicative whereas Weibo is considered outward, open and

1   _{A  public  account  is  similar  to  private  account,  but  it  has  more  customizable  features  such  as  pushing  articles,}   sending  coupon,  offering  seasonal  discount.  

transmissive (Lien, Cao et al., 2014). Figure1 highlights the main differences between Sina Weibo and WeChat.

Figure 1. The difference between Sina Weibo and Wechat

Platform Sina Weibo Wechat

Type Micro-blog Social message app

Parent company Sina corporation Tencent Ltd.

URL http://www.weibo.com/ http://weixin.qq.com/

Users 500 million (2014) 600 million (2014)

Functions 140 characters per post; Verified account; Medal reward system; Customized template; Inline media

(photo/video/music etc.); Portal-like page (Weibo Square);

Trends-highlights Hot topics;

"Open" platform; Third party apps

Hold to talk; Games; Photo/Video/Location etc; QR code; Voice/video call; Group chat; Official account; App integration; Timeline (moment); Shake; Drift bottles; Wallet pay Brand promotion “Quality content creation +

KOL transmission” mode

“ Unique content creation + advocator push” mode Interactive Marketing Diverse formats; Open to

creative implication

Relatively simple formats, which may be further developed in future Sales promotion Strong dissemination

power for product promotion

Suitable for precision marketing to clearly defined audiences Products innovation Could be a fantastic tool,

just like an online focus group, to collect insight from advocates

Marketing can get valuable insight by social data mining and analytics

2.2 Demographic and psychographic variables

The primary factor motivating marketers to utilize social media is the ability to communicate instantly with a global audience (McLaughlin, Gerdes, Westfall & Nutter, 2010, p. 16). However, the level of participation varied according to demographic traits. Thus, ‘demographic segmentation’ divides the market into smaller groups using variables such as age, gender, income, occupation, civil state, religion, race and nationality (Heath, 1996). Brocke, Richter and Riemer (2009) assert that the use of social media networks predominantly varies by gender, age, nationality and civil state. Pew Research Center for the People and the Press (2007) determined the difference between the uses of choosing social media based on gender. The findings reiterated that women are the main users of social media. Meanwhile, social media is most popular amongst teenagers and young adults (Lenhart & Madden, 2007, p. 3).Despite some studies proposing that demographic factors influence consumer purchase intention on social media platforms, further research comparing how it varies on different social media platforms is required. Similarly, studies focusing on Chinese social media platforms would enhance understanding in a global context.

Although some studies have explored the demographic correlates of the construct, limited research has focused on the psychographic correlates of social media use (Lenhart & Madden, 2007, p. 3). Yet, psychographic studies’ escalating interest in the concept of lifestyle has led to the latter being incorporated into lifestyle research (Vyncke, 2002, p. 448). Psychographic variables in commercial communications have

been widely used to acquire an enhanced understanding of the target audience and the mechanism of media selection (Hindman, 2000). Meanwhile, innovativeness refers to the degree to which an individual makes innovative decisions regardless of the feedback provided by others (Midgley & Dowling, 1978, p. 235). Individuals with this trait will be more likely to explore various channels and potentially accept and use them again in the future. As such, this psychological consumer trait may determine the use of social media platforms. Similarly, impulsiveness refers to individuals who are prone to taking spontaneous actions without prior planning (Puri, 1996). Accordingly, innovativeness and impulsiveness are two significant psychographic factors which may impact consumers’ purchase intention on different social media platforms.

2.3 Electronic word-of-mouth variables

In recent years, many studies have explored the effect of word-of-mouth (WOM). Accordingly, WOM is considered capable of influencing people’s views on products and services, which can ultimately determine whether individuals make purchases (Chevalier & Mayzlin, 2006; Herr, Kardes & Kim, 1991; Bansal&Voyer, 2000; Kim, Han&Lee, 2001). Nevertheless, the manner in which WOM influences people’s attitudes and behaviors has been shaped by the continued development of the Internet. Consequently, the term ‘electronic word-of-mouth’ (E-WOM) has been used to describe this new phenomenon.

E-WOM can be defined as, ‘any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet (Hennig-Thurau et al., 2004, p.39). Existing research has examined various forms of E-WOM, such as discussion forums, product reviews, blogs and social networking sites (Dwyer 2007).

With the prevalence of online social communications, E-WOM has become established as a key topic for business and marketing researchers (Chu & Choi, 2011). Social media’s widespread popularity has led to the establishment of a dual-core dumbbell structure of online information dissemination that includes mainstream forums, micro-blogs and portals (Li, 2011). Study asserts that peer-to-peer influence has been transformed by these two core sources of influence. Meanwhile, rather than being passive recipient of information, consumer is in fact active participants in peer-to-peer product recommendations through E-WOM (Chu & Kim, 2011).

A growing body of research on E-WOM in marketing literature concludes that E-WOM plays a pivotal role in social media. However, empirical studies examining how E-WOM impacts purchase intention on Chinese social media platforms are lacking. Furthermore, no studies explore whether the impact of E-WOM on consumer purchase decision-making varies social media platforms. As such, the primary objectives of this research is to examine whether and how E-WOM affects consumer attitudes and the intention to make purchases on the Chinese micro-blog, Sina Weibo,

and the Chinese social networking app, WeChat.  

2.4 Formulate the research question

Because little studies have explored the impact for customer purchase intention on different social media platforms, particularly there is no studies research on various Chinese social media platforms. Therefore, the objectives of this study are to examine the consumer purchase intentions and attitudes on Chinese social messaging app (WeChat) and micro-blog platform (Sina Weibo). Besides, base on the foregoing discussions, my research aim to test whether there are some differences for consumer purchase intentions on different product categories, such as clothing, electronics and flight tickets. Considering the gaps identified through the literature review on the researched fields, we developed a research question as “Whether and how the use of different social media affects consumer attitudes and purchase intentions for different product categories?”

3. CONCEPTUAL FRAMEWORK

3.1 The Conceptual Map

Based on the previews literature discussions, the use of Wechat and the use of Sina Weibo are the independent variables. Demographics, psychographics and E-WOM are determined to be three moderators. Consumer purchase intention will be the dependent variables that are influenced by different use of Chinese social media

platforms (Wechat and Sina Weibo). Demographics, psychographics and E-WOM will be tested as a moderator existed on the relationship between the use of platforms and purchase intention. Upon the critical evaluation conducted on previous literature, and the identified gaps, there are ten hypotheses will be formulated and proposed. The model with all tested hypotheses is shown in the Figure 2.

Figure 2

3.2 Hypothesis Development

We present the conceptual framework that illustrates the different social media platforms to purchase intention relationship and all the constructs of the research

The  use  of           Chinese  Social   media  platforms   Wechat   Sina  Weibo   Consumer   purchase  intention   Mo de ra to rs   Demographic     E-­‐WOM   Psychographic   Gender   Age   Impulsiveness   Innovativenes s   Clothing   Electronics   Fight  tickets  

question. The framework seeks to examine whether and how the use of different Chinese social media platforms (Wechat and Sina Weibo) impacts consumer purchase intention between different product categories (clothing, flight tickets and electronics). Besides, we investigate how the main variables (demographic traits, psychographic traits and EWOM) moderate the relationship. By explaining the building blocks that characterize our model, this section formulates ten hypotheses about the effects.  

3.2.1 Hypotheses of different social media platforms and product categories

Consumer purchase behavior can vary based on different functions of online shopping platforms (Liao & Cheung, 2001). As discussed, there are numerous differences between Sina Weibo and WeChat. Sina Weibo is a micro-blog style social media platform whereas WeChat is a social messaging app. Accordingly, each hasits own consumer base, functions, brand promotion strategies and sales promotion techniques. These types of disparities between Sina Weibo and WeChat could be significant factors that influence the consumer purchase intention in various ways. For example, WeChat operates its own online shop (‘micro-shop’), which enables users to purchase products from the store directly through the WeChat payment system (‘wallet pay’). In contrast, Sina Weibo users can only make purchases through a third party.

Researchers have also claimed that different product categories influence consumers’ online purchase intention and attitudes (Bhatnager et al., 2000; Liao & Cheung, 2001). Based on the existing literature, there is a marked difference in people’s mental

processes when purchasing different products. Studies have also shown that this phenomenon can be impacted by social media platforms (Atsmon et al., 2010). In this study, three products categories (clothing, plane tickets and electronics) will be used to assess which product categories individuals are more likely to purchase on Sina Weibo and WeChat. Consequently, it is necessary to formulate hypotheses.

3.2.2 Hypotheses of demographic traits

The existing literature identifies that demographics influence the consumer decision-making process, purchase intention and channel behavior (Slama & Tashchian, 1985; Konuş et al., 2008). As such, age and gender are factors requiring consideration. It is apparent that men and women purchase different products, while they also have distinct shopping habits (Schmitz, 2012).In terms of purchasing preferences, male consumers have a greater interest in technological products (Kwong et al., 2003;Ang et al., 2001). In contrast, female consumers are more likely to purchase clothing (Cheung & Prendergast, 2006).

Besides gender, the age of consumers is also an important factor. For example, it determines outlook on technological innovations. The younger generations were born

H1a:  Consumers  are  expected  to  show  a  stronger  purchase  intention  on  WeChat   (social  messaging  app)  rather  than  on  Sina  Weibo  (micro-­‐blog).  

H1b:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   consumers  are  expected  to  show  a  stronger  purchase  intention  on  clothing  rather   than  electronics  and  flight  tickets.  

into the digital age and are thus more comfortable with this technology. Conversely, the older generations are not as familiar with this technology, despite living through this technological revolution ((Schmitz, 2012). Consequently, the younger generations are the main users of social media and they prefer to use the social media platforms with various innovative functions. Moreover, this also means that the younger generations are more inclined to use innovative channels when making purchases.

In order to enhance understanding, gender and age will be used as the demographic factors to test the impact on the purchase intention. Accordingly, the following hypotheses have been proposed:  

3.2.3 Hypotheses of psychographic traits

Psychographic factors have a direct impact on consumer behaviour (Ailawadi, Neslin& Gedenk, 2001). For this research, innovativeness and impulsiveness must be considered. Impulsive individuals are prone to taking sudden actions without prior planning (add reference). Similarly, impulsive behavior can be defined as choosing

H2a:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   women   are   expected   to   show   a   stronger   purchase   intention   (such   as   to   buy   clothing,  electronics  and  flight  tickets)  rather  than  men.  

.  

H2b:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   consumers   aged   under   30   are   expected   to   show   a   stronger   purchase   intention   (such   as   to   buy   clothing,   electronics   and   flight   tickets)   rather   than   people   aged   over  30.  

option which provide immediate hedonic benefits with long-term consequences (Puri, 1996). For instance, many impulsive app users may be attracted to purchasing through this channel because of its ease and convenience. In this scenario, impulsive consumers may prefer to purchase items through WeChat rather than Sina Weibo, as WeChat operates its own shop (micro-shop). Therefore, consumers are able to purchase items through the WeChat shop using the platform’s ‘wallet pay’, instead of through third parties on Sina Weibo.

In addition, another important consideration is innovativeness. Innovativeness has been explored by existing studies on internet use (Donthu & Garcia, 1999; Wells & Chen, 1999). Individuals with a preference for innovation are prepared to adopt new technology and adjust their lifestyle accordingly. Therefore, it is reasonable to suggest that the use of WeChat has a positive correlation with innovativeness since it has a wider range of fashionable and technical features than Sina Weibo. As such, the following hypotheses are proposed.  

3.2.4 Hypotheses of Electronic word-of-mouth

H3a:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   impulsiveness   and   innovativeness   are   expected   to   have   a   positive   effect   on   consumer  purchase  intention.  

H3b:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   impulsive   and   innovative   consumers   are   expected   to   show   a   stronger   purchase   intention  on  WeChat  than  on  Sina  Weibo.  

As the previous literature reveals, there is a unique phenomenon of product-focused E-WOM on social media which involved the important management implications (Chu and Kim 2011). Therefore, keeping a watchful eye on E-WOM on social media is necessary, especially for the perspective of different platform structure, and even the different product categories. (Chu & Kim, 2011). Therefore, in my paper, I assume that E-WOM will affect the consumer purchase decision-making differently on Sina Weibo (micro-blog) and WeChat (social messaging app) because of the various features. As discussed before, WeChat users should first add each other as friends on the app in order to make the relationship reciprocal. However, Sina Weibo users do not need to add each other as friends. Therefore, WeChat offers ‘close-loop communication’ in a private space and it can be described as the inward platform, whereas WeChat is considered an outward, open platform. I formulate the hypothesis to test the impact of E-WOM on three different product categories.

H4a:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   E-­‐WOM  is  expected  to  have  a  positive  effect  on  consumer  purchase  intention.    

H4b:   E-­‐WOM   is   expected   to   have   a   stronger   effect   on   consumer   purchase   intention  for  Wechat  rather  than  for  Sina  Weibo.  

H4c:   On   both   Sina   Weibo   (micro-­‐blog)   and   WeChat   (social   messaging   app),   E-­‐WOM  is  expected  to  have  a  stronger  effect  for  consumer  to  buy  clothing  rather   than  to  buy  electronics  and  flight  tickets.  

4. METHODOLOGY

4.1 The sample collection

Questionnaire distribution is the research method for this study of consumers’ purchase intention and attitude on different social media platforms (Sina Weibo and Wechat). The sample for this study is obtained from sending questionnaires through Internet to both Sina Weibo users and Wechat users in China. Because of the large size of population, no sampling frame is considered. Mostly, a self-selection sample was adopted and it allowed identifying people willing to fill in the questionnaires. The sample is collected as large as possible to reduce the sampling error. And a non-probability sampling method is applied to collect data. The minimum sample size of this research is 200 users. And according to previous research, the expected response rate is 55%-65% such that we need send out at least 300 questionnaires for data gathering. The data collection begins at the date of 25th April 2015. On the 5th May 2015, 201 questioners had been filled in. 168 were used of analysis because there are 33 questionnaires missing too many data.

4.2 Survey design

In order to test the hypotheses a quantitative research approach is adopted. In particular, a one-time cross-sectional online survey is used to properly conduct the research and collect data about the consumer purchase intention towards different

Chinese social media platforms (Sina Weibo and Wechat), and the impact of moderators on their relationship. The survey strategy is widespread in business and management research. Indeed, surveys allow the collection of data through standardized questions from a large number of people in a cost-effective way.

The questionnaire consists of three parts, which includes 37 questions in total. The first part measures the independent variables, the use of Wechat and Sina Weibo. In this part, 22 questions are included and respondents are asked to answer questions about the purchase attitude on Wechat and Sina Weibo respectively. Besides, we used three different product categories (clothing, electronics and flight tickets) in the first part to investigate the different purchase intention among them. The second part mainly measures the moderators such as psychographic traits, and electronic word-of-mouth. To be able to test whether moderators impact the different social media platforms and purchase intention relationship, the second part of the survey consisted of six psychographic questions and four electronic word of mouth-related questions. The last part measures the respondents’ demographics characteristics. For instance, respondents are asked some basic information such as gender, education, income and age, five questions included in the third part. Measures for independent variables and moderators are composed of multi-statements in which participants were asked to answer using a five-point Likert scale ranging from (1) Strongly Disagree to (5) Strongly Agree.

5. EMPIRICAL RESULTS

In order to answer the research question:” Whether and how the use of different social media affects consumer attitudes and purchase intentions for different categories”, the collected data will be analyzed in this chapter. In the first section I will study the results in a descriptive manner based on statistical analysis. In the second part I will test my hypotheses by subjecting the data to t-test, correlation analysis and regression analysis.

5.1 Missing value and recording

Before data analysis, the data will be examined on missing values. In statistics, missing data occur when no data value is stored for the variable in an observation. In our data analysis, the frequency test was used to check all variables of missing data. The missing value is not detected in my data. Recording can be applied to indicate if there are counter-indicative items in the scales. Counter-indicative items are items that are indicative or the opposite of what you want to measure in a scale. Furthermore, in our analysis, in order to enable to use gender as the moderator, the items need to be adjusted from 1(male), 2(female) to 1(male), 0(female).

5.2 Reliability test

were run for electronic word of mouth, innovativeness, impulsiveness, gender as well as age for buying intention and attitude of different social media platforms (Sina Weibo and Wechat). The Cronbach’s alpha, which represents the estimator of the internal consistency, has been tested to verify if all the items in one scale measure the same. In the analysis, if the Cronbach’s alpha> 0.7 that means the data could be considered acceptable. The table 1 shows innovativeness and impulsiveness consisted of 3 items respectively (α=0.848; α=0.739), the electronic word of mouth subscales consist of 4 items (α=0.934). Both of the Sina Weibo and Wechat for purchase intention subscales consisted of 4 items (α=0.839; α=0.874). The Cronbach’s alpha data shows the high level of internal consistency. In Table 1 the Cronbach’s alpha of the variables that were used in the final analysis can be observed.

Table 1: Reliability of scales

Variable N of Items Cronbach’s Alpha*

Purchase intention Sina Weibo 4 0,839

Purchase intention Wechat 4 0.874

Innovativeness 3 0.848

Impulsiveness 3 0.739

Electronics word of mouth 4 0.934

*Cronbach’s Alpha should > 0.70

5.3 Scale Means and correlation check

created for hypothesis testing. The mean of all items was calculated and used to describe a variable. Means and standard deviations of all variables are exhibited in table 2. Correlation matrix could be provided by STATA with the table of correlation coefficients. The significance value of the correlation and the sample size (N) on which it is based are displayed. Pearson’s correlation coefficient was calculated in order to determine whether the variables were related or not. Clearly, each variable is perfectly correlated with itself and r=1 along the diagonal, as showed in Table 2 (Field, 2013).

Table 2: Descriptive Statistics

Means, Standard Deviations, Correlations

Mean Std 1 2 3 4 5 6 7 1. Sina Weibo 2.23 1.19 1 2. WeChat 3.49 1.19 0.19* 1 3. Impulsiveness 3.32 0.70 0.05 0.15* 1 4. Innovativeness 3.17 0.84 0.21** 0.32** 0.36** 1 5. E-WOM 3.60 0.97 0.17* 0.16* 0.26** 0.12 1 6. Gender 0.37 0.48 -0.05 0.03 -0.15* 0.21** -0.18* 1 7. Age 0.76 0.43 -0.17* -0.04 0.07 0.01 -0.05 0.08 1 ** Correlation is significant at the 0.05 level (2-tailed)

5.4 Hypotheses testing

Statistical test in STATA were conducted in order to test whether the expected associations and relationships in the four hypotheses were significant in our study.  

5.4.1 Paired sample t-test of H1

In order to compare the buying score between the two independent variables (Sina Weibo and Wechat) in our study, we use the paired sample T test to test the hypothesis 1a. As shown in the Table 3, we can see the purchase intention of Sina Weibo average at 2.2 with standard deviation 0.09, while the purchase intention of Wechat is around 3.5, and the difference -1.26 is highly significant—the t-statistics is -10.80 (p<0.01). Hypothesis 1a is accepted.

Table 3 Paired sample t-test result of H1a

H1. Testing of the purchase intention between Sina Weibo and Wechat

Variables Mean Std. Err. t-Value (p-value)

Purchase Intention (Sina Weibo) 2.23 0.09 Purchase Intention (WeChat) 3.49 0.09

Difference -1.26 0.12 -10.80 (<0.01)

Hypotheses 1b aim to test consumer purchase intention and attitude towards different product categories (clothing, electronics and flight tickets) on different Chinese social media platforms (Sina Weibo and Wechat). To further investigate hypothesis 2, Paired sample T test was conducted. In the Table 4, we could see the purchase

intention on Sina Weibo and Wechat towards three different categories (clothing, electronics and flight tickets). The average values of the three categories on Sina Weibo are 2.50, 2.69 and 2.52, while the standard errors for the three categories are the same—0.08. The mean buying intentions on Wechat for the three categories are higher than that on Sina Weibo, which are 2.50, 2.69 and 2.52 respectively. The standard errors of the three categories on Wechat are 0.08, 0.09 and 0.09. The t-value of purchase intention is -5.01 (p<0.01) for clothing, -1.71 (p=0.09) for electronics and -0.63 (p=0.53) for flight ticket.

Table 4 Paired sample t-test result of H1b

Panel A: purchase intention towards clothing on Sina Weibo and Wechat

Variables Mean Std. Err. t-Value (p-value)

Clothing (Sina Weibo) 2.50 0.08

Clothing (WeChat) 2.94 0.08

Difference -0.44 0.09 -5.01 (<0.01)

Panel B: purchase intention towards electronics on Sina Weibo and Wechat

VVariable Mean Std. Err. tt-Value (p-value)

Electronics (Sina Weibo) 2.69 0.08

Electronics (Wechat) 2.85 0.09

Panel C: Purchase intention towards flight tickets on Sina Weibo and Wechat

Variables Mean Std. Err. tt-Value (p-value)

Flight tickets (Sina Weibo) 2.52 0.08

Flight tickets (Wechat) 2.58 0.09

Difference -0.06 0.09 -0.63 (0.53)

5.4.2 Independent sample t-test of H2

Hypothesis 2 investigates whether and how the demographic traits, such as gender and age, influence the relationship between different social media platforms and purchase intention. Besides, we have checked the buying attitude on clothing, electronics and flight ticket based on gender and age.

Table 5 uses the independent sample t-test to investigate whether gender would affect the purchase intention. Panel A and B reports the t-test results of Weibo platform and Wechat platform respectively. As we can see in the panel A of Table 5, the purchase intention on Sina Weibo average at 2.28 with standard error 0.11 for female against average of 2.16 and standard error 0.16 for male. The difference between genders is 0.11 and the t-value is 0.6 (p=0.55). Comparing with Sina Weibo, the purchase intention on Wechat is higher on Wechat for both man and woman. The mean value of purchase intention for female is 3.47, while it is 3.53 for male. The standard error of the intention is 0.10 for woman and it is 0.07 higher for man. The t-value of this test is -0.34 (p=0.85). Therefore, the hypothesis 2a is rejected.

Table 5 Independent sample t-test result of H2a

Panel A: Purchase intention on Sina Weibo by gender

Group Mean Std. Err. t-Value (p-value)

Female 2.28 0.11

Male 2.16 0.16

Combined 2.23 0.09

Difference 0.11 0.19 0.6 (0.55)

Panel B: Purchase intention on Wechat by gender

Group Mean Std. Err. t-Value (p-value)

Female 3.47 0.10

Male 3.53 0.17

Combined 3.49 0.09

Difference -0.07 0.19 -0.34 (0.85)

Table 6 breaks down the purchase intention into difference product categories, and uses paired t test to investigate that whether the gender affect the purchase intention on different product categories. First, for the clothing category, the women’s attitude for buying shows a mean value of 2.84 with a 0.07 standard error. It comes to be 2.51 of the average value from men with a 0.14 standard error. For electronic products, male show a higher purchase attitude by a 2.78 mean value against 2.76 from female. The attitude of female for buying flight ticket on social network is lowest, which has an average of 2.57. However, the mean value of purchase intention of male keeps the

same as clothing category. The t-value of the three categories are 2.36 (p=0.02), -0.18 (p=0.86) and 0.43 (p=0.67) respectively.

Table 6 Independent sample t-test result of H2a (gender)

Panel A: buying attitude for clothing by gender

Group Mean Std. Err. t-Value (p-value)

Female 2.84 0.07

Male 2.51 0.14

Combined 2.72 0.07

Difference 0.33 0.14 2.36 (0.02)

Panel B: buying attitude for electronics by gender

Group Mean Std. Err. t-Value (p-value)

Female 2.76 0.07

Male 2.78 0.14

Combined 2.77 0.07

Difference -0.03 0.15 -0.18 (0.86)

Panel C: buying attitude for flight tickets by gender

Group Mean Std. Err. t-Value (p-value)

Female 2.57 0.08

Male 2.51 0.14

Combined 2.55 0.07

To further investigate whether and how ages affect the consumer buying attitude on Sina Weibo and Wechat separately, independent sample t-test is conducted to check this relationship. Table 7 investigates the impact of consumer ages on purchase intention in difference platforms Sina Weibo and Wechat.

Table 7 Independent sample t-test result of H2b (age)

Panel A: Purchase intention on Sina Weibo based on age

Group Mean Std. Err. t-Value (p-value)

Age over 30 2.6 0.20

Age under 30 2.12 0.10

Combined 2.23 0.09

Difference 0.48 0.21 2.26 (0.02)

Panel B: Purchase intention on WeChat by age

Group Mean Std. Err. t-Value (p-value)

Age over 30 3.58 0.18

Age under 30 3.46 0.11

Combined 3.49 0.09

Difference 0.11 0.22 0.51 (0.61)

The mean of purchase intention on Sina Weibo is 2.6 for the group that age over 30, and it is 2.12 for group age under 30. The standard errors for these two groups are 0.20 and 0.10 respectively, and t-value is 2.26 (p=0.02). On Wechat platform, the purchase intention for consumers over 30 has a mean of 3.58 with standard error 0.18.

The mean and standard error of purchase intention for group under 30 are 3.46 and 0.11. The t-value of this test is 0.51 (p=0.61). Interestingly, respondents with age above 30 show higher purchase intention, and such difference is significant at 5% level for Weibo platform.

5.4.3 Regression analysis of H3

In order to identify how innovativeness and impulsiveness affect consumer purchase intention on Sina Weibo and Wechat, we run the regression to test hypothesis 3. We run two regressions using the purchase intention on Wechat platform and the purchase intention on Weibo platform as dependent variables respectively, including Impulsiveness and Innovativeness and controlling some other variables. Table 9 reports the results of OLS regression, where the dependent variables are the purchase intention for Wechat (column 1) and for Weibo (column 2). The regression coefficients loaded on innovation are 0.381 and 0.394 for Wechat and Sina Weibo, respectively, and the statistical significances are significant at 1% level. The impact impulsiveness is insignificant for both Wechat and Weibo platform. For other control variables, I find that income has large impact on the purchase intention in Wechat platform, the coefficient of income is 0.206, which is significant at 5% level. In contrast, the income seems does not affect the purchases intention in Weibo platform, if it is, the impact is negative (-0.069). In addition, the coefficients of spending time, which is the total time spent in social platform per week, are significantly affect the purchase intention on both platforms (0.204 vs -0.212), but interestingly the spending

time affect the purchase intention in opposite directions, i.e., the coefficient of spending time is positive in Wechat platform while is negative in Weibo platform. For other control variables, we do not find any statistical significant impact on purchase intention.  

Table 9 Regression analysis result of H3

VARIABLES Purchase intention

(Wechat) Purchase intention (Sina Weibo) Impulsiveness 0.065 -0.046 (0.140) (0.143) Innovativeness 0.381*** 0.394*** (0.118) (0.121) Gender -0.095 -0.283 (0.204) (0.209) Age -0.237 -0.317 (0.218) (0.224) Education 0.113 -0.114 (0.144) (0.147) Income 0.206** -0.069 (0.096) (0.098) Spending time 0.204* -0.212* (0.116) (0.119) Constant 0.762 2.722*** (0.735) (0.753) Observations 167 167 R-squared 0.153 0.112

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1  

5.4.4 Regression analysis of H4

Hypothesis 4a investigates whether and how electronic word of mouth (E-WOM) impacts consumer purchase intention on Sina Weibo and Wechat. In Table 7, we separate the EWOM for two platforms.

Table 10 Regression analysis result of H4a

VARIABLES Purchase intention

(Weibo) Purchase intention (Wechat) Impulsiveness -0.104 -0.021 (0.140) (0.138) Innovativeness 0.355*** 0.372*** (0.118) (0.114)

E-WOM (Sina Weibo) 0.264***

(0.081) E-WOM (Wechat) 0.289*** (0.088) Gender -0.240 -0.032 (0.203) (0.199) Age -0.276 -0.156 (0.218) (0.213) Education -0.123 0.074 (0.143) (0.140) Income -0.026 0.239** (0.0962) (0.0936) Spending time -0.239** 0.169 (0.115) (0.113) Constant 2.080*** 0.102 (0.757) (0.741) Observations 167 167 R-squared 0.168 0.207

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

In both platforms, E-WOMs are highly significant at 1% level, and the coefficients are 0.264 and 0.289 for Wechat and Sina Weibo, respectively. Thus, we can see that the hypothesis H4a is accepted that E-WOM is expected to show positive effect on consumer purchase intention. The innovation still has the largest impact on purchase intention even we introduce the E-WOM in our regression. The coefficients of

innovation are 0.355 and 0.372 for Wechat and Sina Weibo, respectively, and both of the coefficients are significant at 1% level.

Hypothesis 4b investigates whether the E-WOM works differently for purchase intention on social media platforms when consumer faces different product categories. We run the regression for different product categories and show the result in Table 11.

Table 11: regression analysis result of H4b

VARIABLES Clothing Electronics Fight Tickets

E-WOM 0.409*** 0.390*** 0.251*** (0.032) (0.033) (0.036) Impulsiveness 0.057 0.082 0.091 (0.097) (0.099) (0.106) Innovation 0.218*** 0.274*** 0.282*** (0.081) (0.082) (0.088) Gender -0.236* 0.036 -0.061 (0.140) (0.143) (0.153) Age -0.061 -0.160 -0.355** (-0.150) (0.153) (0.164) Education 0.061 0.019 0.079 (0.099) (0.101) (0.108) Income -0.015 0.077 0.003 (0.066) (0.068) (0.072) Spending time -0.019 -0.058 -0.157* (0.079) (0.081) (0.087) Constant 0.391 0.281 0.984* (0.525) (0.537) (0.575) Observations 167 167 167 R-squared 0.313 0.296 0.221

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

categories. The coefficients of E-WOM are 0.409, 0.390, and 0.251 for clothing, electronics, and flight tickets categories, respectively, and all of them are significant at 1% level. In addition to the E-WOM, the innovation is still highly significant and economic meaningful. The impacts of innovation are similar for electronics and flight ticket categories, in which the coefficients are around 0.275, and the impact of innovation is lower in clothing category, in which the coefficient is approximate 0.22. For other control variables, I only find statistically significant impacts of age (-0.355 ) and spending time (-0.157) in flight tickets category, and the others are not significant.

6. DISCUSSION

In the discussion three aspects are covered. Firstly, I discuss the hypothesis testing with the results derived from previous section. Secondly, I explore the managerial implications the findings demand. Thirdly, I present some limitations and area for further research of this research paper.

6.1 Discussion of Hypotheses

6.1.1 Discussion of Hypotheses 1

H1a: People are expected to show stronger purchase intention on Wechat rather than

on Sina Weibo.

expected to show stronger purchase intention on clothing rather than electronics and flight tickets.

Our study present the purchase intention of Sina Weibo average at 2.2 with standard deviation 0.092 (Table 3,), while the purchase intention of Wechat is around 3.5, the difference -1.26 is highly significant—the t-statistics is -10.28. Therefore, at least in my sample, peoples are more likely to buy products on Chinese social media platform Wechat rather than on Sina Weibo. Table 2 (panel A, panel B and panel C) reports the buying intention in different product categories, namely cloth, electronic devices and fly tickets, on different social platforms Weibo and Wechat. Generally, peoples show higher purchase intention on Wechat platform in all categories. The difference of buying intention of cloth between two platforms is significant at 1% level (t-value -5.01), which shows that, in both Sina Weibo and Wechat platforms, people exhibit higher purchase intention on clothing. However, we do not find any statistically significant differences in buying electronic devices and fly tickets. Thus, we can conclude that hypothesis 2 is accepted.

Overall, my results of hypotheses 1 are in general consistent with expectation and previous literature that consumer purchase intention could be affect by different functionalities of different social platforms (Liao and Cheung 2011) and by different product categories (Bhatnager et al 2000). As mentioned before, one possible underlying mechanism why consumers exhibit higher purchase intention on Wechat

platform could be that Wechat has its own method of payment, which to large extent facilitate the purchase, while, in contrast, purchase in Weibo platform has to interact with third parties.

6.1.2 Discussion of Hypotheses 2

H2a: On both Chinese social media platforms (Sina Weibo and Wechat), women are

expected to show stronger purchase intention (e.g. buy clothing, electronics and fights tickets) rather than men.

H2b: On both Chinese social media platforms (Sina Weibo and Wechat), people aged

under 30 are expected to show stronger purchase intention (e.g. buy clothing, electronics and fights

As the result, female exhibit slightly higher average purchase intention than male do in Weibo platform, a finding which is consistent with our expectation, but the difference is not statistically significant. However, in Wechat platform, the story seems reversed—male show higher purchase intention, but the difference is still not statistically significant. Therefore, we would say that female do not significantly exhibit higher purchases than male do, and, therefore, the hypotheses 2a is rejected.

Although the difference of overall purchase intention between female and male is not statistically significant, it is still interesting to investigate whether there is systematic difference between female and male over different sub-categories. Table 6 breaks

down the purchase intention into different product categories, namely buying cloth, electronics, and fights. Overall, females show higher purchase intention for cloth and fly tickets, and males exhibit higher purchase intention for electronic devices. However, we only find statistically significant difference of buying cloth, for other two product categories I do not find any significant difference. Thus, the conclusion draw from Table 6 seems to be consistent with that of Table 5, i.e., women do not show higher purchase intention than men do.

Hypotheses 2b investigates the impact of consumer ages on the purchase intention in Sina Weibo and Wechat. Interestingly, respondents with age above 30 show higher purchase intention, and such difference is significant at 5% level in Weibo platform (table 7). Intuitively, we expect to observe that consumer with age under 30 shows higher purchases intention, since they might have better knowledge in accessing the mobile internet, or considering that youth are usually more likely pursuing devices with the latest technology, they might have more convenient methods to make mobile payment than older people do. However, note that Table 7 only display the group mean difference without controlling other factors that might affect the purchase intention, we cannot concretely say that people with age over 30 present higher purchase intention than people with age under 30 do. One possible explanation is that younger people would be busy with their prospect and aspiration, thus they have less time spent on social platform than older people do. Or it could also simply be that older people have more disposable income to purchase than younger people do.

Overall, my results of hypotheses 2 are partially consistent with the existing literature. On the one hand, my result shows that ages, as a demographic variable, do affect the consumer purchase intention, a finding that is consistent with (Slama & Tachchian 1985; Konus et al 2008). On the other hand, in contrast with literature which argue that men are more likely to buy technological products (Kwong et al, 2003) and women are more likely to buy cloth (Cheung and Prendegast 2006), it seems that in my sample gender does not matter too much. I only find that there is a significant difference of cloth purchase intention between male and female, but gender seems does not affect other categories such as electronic devices and flight tickets. The reason why my sample fails to reveal the difference of purchase intention between male and female could be that there are many large alternative online shopping platforms in China, such as Alibaba or Amazon, when consumers demand flight tickets or electronic devices, they just simply switch to other platforms which are supposed to be more professional in e-commence field, and thus reducing the statistical significance.

6.1.3 Discussion of Hypotheses 3

H3a: Impulsiveness and innovativeness are expected to show positive effect on

consumer purchase intention.

H3b: Impulsive and innovative people are expected to show stronger purchase

Controlling for age, education, income, and spending time, we find that innovation has statistically significant impact on the purchase intention in both platforms. The results are reported in Table 9. Specifically, in Wechat platform, one standard deviation increase in innovativeness (0.84) would results in an increase of purchase intention by 0.32 (0.84*0.381), which is about 27.1% (0.32/1.19) of the standard deviation of purchase intention in Wechat platform. Thus, we could interpret such results as that the innovativeness has strong economic and statistical impact on the purchase intention. In Sina Weibo platform, one standard deviation increase in innovativeness (0.84) would results in an increase in purchase intention by 0.33 (0.84*0.394), which is about 27.8% of the standard deviation of purchase intention in Sina Weibo platform. Therefore, the innovativeness also has strong economic and statistical impact on the purchase intention in Sina Weibo platform.

In comparison of the economic magnitude, the economic magnitude of innovativeness is slightly higher in Weibo platform than in Wechat platform (0.33 vs. 0.32). In addition, the impulsiveness seems to have no impact on the purchase intention on both Chinese social media platforms. Therefore, we would say that we partially confirm the hypothesis H3a that innovativeness have positive effect on consumer purchase intention, but the impulsiveness has no effect on purchase intention. As for H3b, by comparing the economic magnitude of innovativeness in both platforms, I find the economic magnitude of innovativeness in Sina Weibo platform seems higher than that