Analysis of Chinese millennial consumers' preference for online advertising channel in China

Analysis of Chinese millennial

consumers' preference for online

advertising channel in China

BACHELOR DEGREE PROJECT THESIS WITHIN: Business Administration NUMBER OF CREDITS: 15 ECTS Points PROGRAMME OF STUDY: Marketing Management AUTHORS: Liwen Zhang,Jinjing Song, Xinyu Li TUTOR: Jenny Balkow JÖNKÖPING May 2020

i

Acknowledgements

We would like to acknowledge and express our heartfelt gratitude to those who contributed and supported the development of this thesis.

First of all, we would like to thank our thesis tutor Jenny Balkow for her tireless support. This thesis lasted for half a year, during this period, she has been giving us valuable opinions and careful guidance with her professional knowledge. In addition, her serious scientific attitude and rigorous academic spirit also deeply infected and inspired us.

Meanwhile, we would also like to thank our peers in the seminar for giving us valuable opinions and different perspectives during each seminar. Then, we would like to thank every respondent who participated in our online questionnaire. Their participation makes this thesis accessible to important data for analysis.

Also, we would also like to acknowledge Anders Melander for providing a clear guideline and structure for the thesis.

Finally, we would also like to thank our parents, families, and friends. During our bachelor program, the great encouragement and support they have given to us is our precious wealth in our life.

______________ ______________ ______________

Liwen Zhang Jinjing Song Xinyu Li

ii

Bachelor Thesis

Title: Analysis of Chinese millennial consumers' preference for Chinese e-commerce

advertising channel

Authors: Liwen Zhang, Jinjing Song & Xinyu Li Tutor: Jenny Balkow

Date: 2020-05

Keywords: Online advertising channel, preference, preference difference, Chinese

millennial, Taobao user

——————————————————————————————————

Abstract

An increasing number of foreign companies want to enter the Chinese market. Online advertising is one of the indispensable means of the company's operation and marketing, and foreign companies poorly understood the online advertising channels in China. Therefore, this study uses the AISAS model to study the preferences of Chinese millennial generation Taobao users for online advertising channels. And research through three dimensions are gender, city level and different product categories. Specifically, this research through online questionnaires and finally collected 150 samples, it was found that Chinese millennial generation Taobao users favored the promotion of WeChat public accounts, followed by Pre-movie adverts. Further, the study confirms that consumers do have preferences for online advertising choices and that they will change as they are placed on different e-commerce platforms. Based on the previous study's interpretation of preferences, the preferences are defined as five exact post behaviors to better evaluate consumers' choices for online advertising. Moreover, through data analysis, it is found that there are differences between male and female online advertising channels among Chinese millennial generation Taobao users. Females prefer online advertising channels that are interactive than males such as live streaming. Males prefer online advertising channels that are dominated by pictures/images for instance mini banners in App and Web. In addition, live streaming has an advantage in the household goods category.

iii

Table of content

2.1.1 Traditional advertising vs. Online advertising ... 7

2.1.2 The development of online advertisement and status quo ... 7

2.2 ONLINE ADVERTISING CHANNELS ... 11

2.3 CONSUMER (MILLENNIALS) PERCEPTION OF ONLINE ADVERTISING CHANNEL ... 14

2.3.1 Generation Cohort: Millennials ... 14

2.3.2 Consumers' perception of the transition from offline to online advertising channels ... 15 2.4 THEORETICAL FRAMEWORK ... 17 2.4.1 Preference ... 17 2.4.2 AISAS model ... 20 3. METHODOLOGY ... 22 3.1 RESEARCH APPROACH ... 22 3.2 LITERATURE REVIEW ... 25 3.3 SAMPLING ... 26 3.4 DATA COLLECTION ... 27 3.5 PROBLEM FORMULATION ... 30 3.6 METHOD OF ANALYSIS ... 31 3.7 TRUSTWORTHY CONSIDERATION ... 31 3.8 ETHIC ... 33 4. RESULT ... 34

4.1 OVERVIEW OF THE DATA ... 34

4.2 MINI BANNER IN APP AND WEB ... 37

4.3 LIVE VIDEO STREAMING ... 38

4.4 WECHAT PUBLIC ACCOUNT ... 39

4.5 INFLUENCER PROMOTION ON SOCIAL MEDIA ... 40

4.6 INFLUENCER PROMOTION ON VIDEO PLATFORM ... 41

4.7 OPEN PAGE OF THE APP ... 42

4.8 PRE-MOVIE ADVERTS ... 43

4.9 INFLUENCER PROMOTION ON SHORT VIDEO PLATFORM ... 44

5. DATA ANALYSIS ... 45

5.1 THE PREFERENCE OF CHINESE MILLENNIAL TAOBAO USERS ... 45

iv

6. CONCLUSION & DISCUSSION ... 52

6.1 SUMMARY OF FINDINGS ... 52 6.2 THEORETICAL CONTRIBUTION ... 55 6.3 PRACTICAL CONTRIBUTION ... 57 6.4 LIMITATION ... 58 6.5 FURTHER RESEARCH ... 59 REFERENCE LIST ... 60 APPENDIX A ... 73 APPENDIX B ... 81

1

1.Introduction

This chapter introduces the background of e-commerce and its advertising channels. Furthermore, it discusses the existing problems and gaps in the current research status. Finally, the purpose of this study is stated.

1.1 Background

E-commerce age by the rapid development of internet technology alters the mainstream trend of global consumption, which formed a popular consumption pattern that tends to rely on online purchasing. According to statistics on global e-commerce retail sales (Statista, 2020), sales have increased from 1,336 billion dollars in 2014 to 4,206 billion dollars, and sales are expected to surge to 6,542 billion dollars in the next 2023. Those figures presented a profitable selling situation that enables the majority of industries to reflect the common tendency and devote themselves to catch potential business opportunities that the e-commerce age creates for them to bring positive benefits and attractive future prospects compared with the previous consumption.Indeed, a positive development of online consumption provides more benefits such as sales stimulation, cost savings and entering obstacles elimination. At the same time, online shopping develops a consumption pattern with price transparency and shopping convenience that promotes the transformation of consumption concepts among modern customers (Yang, Zhao, & Wan, 2010). Furthermore, consumers begin to adopt and enjoy the new model because of those advantages, and eager to better keep up with the modern consumption pattern. Therefore, those positive achievements of the transformation in the digital age, the new thoughts and habits of consumption for customers have a great influence on the trend of global consumption which accelerates the development of online

2

purchasing. As a derivative of online consumption, the e-commerce platforms are regarded as effective and essential channels for enterprises to choose for advertising (Lee, 2001). E-commerce enterprises need to tailor the most appropriate platform and develop effective marketing strategies through close cooperation with platforms in order to gain more profits and market share.

Under the lead of the global consumption trend, there is no doubt that China as a large population density with high speed of enterprises updating has resulted in a large online sales volume. On July 16, 2019, the 24th China Internet Development Statistics Report released by China Internet Network Information Center (CNNIC) showed that due to the financial crisis, the number of online shopping users increased from 74 million to 88.88 million, an increase of nearly 14 Million (Yang, Zhao, &Wan, 2010). Thus, we can see from that China has become a major force in the global model of reliance on online consumption since Yang, Zhao, and Wan (2010) shows there is one of four Chinese customers is online consumers.

In order to meet better satisfaction of Chinese customers, Chinese retail industries pay more attention to foster online sales strategies which also helps the development of e-commerce platforms such as Alibaba. Ali statistics (2019) showed that the overall sales reached up to 268.4 billion on double eleven shopping festival that is held on November 11 every year and is equivalent to the black Friday. Therefore, these increased consumption data evident that active cooperation with e-commerce platforms can indeed win more value. Nowadays, with the continuous innovation of science and technology, many novel advertising modes are derived from e-commerce platforms which need online enterprises centered on in-depth research among lead consumers in terms of attitudes and preferences. Due to the vast land and abundant resources in China, different living environments and cultural habits have created different preferences among different cities. Second, there are gender differences in preferences. Therefore,

3

enterprises are supposed to implement targeted delivery strategies to achieve the maximum feedback effect due to the differences in representativeness.

1.2 Problem discussion

The research on online advertising channels is mainly carried out in terms of its effectiveness and influence. However, most of the existing research is aimed at studying online advertising in a certain category or several categories for comparison rather than studying specific online advertising channels. The existing study pointed out that there are differences in the effectiveness of different types of online advertising channels. Xu, Duan, and Whinston (2014) pointed out that display ads have a relatively low direct impact on purchase conversion compared to search ads, but they are more likely to stimulate subsequent visits through other ad formats. However, because online advertising channels are complex and diversified today, and they contain multiple elements and features, it is impossible to clearly classify online advertising channels. Therefore, there is a gap in this field. And in terms of actual contribution, the actual help for the enterprise is not direct.

In addition, existing research emphasized the differences in preferences of consumers in online advertising channels. Related results indicate that there are different variables in preference differences, such as gender (Weiser, 2000). However, none of the research in this area has involved or studied Chinese consumers or Chinese millennial consumers. More specifically, most of the existing research just set the study object as the majority of consumers (Alalwan, 2018), or just studied the millennials' perception of online advertising (Agarwal & Arora, 2019). Meanwhile, most of the existing studies were mainly aimed at consumers in Europe, the United States, or South Asia, and there were few studies on the status of Chinese online advertising or online advertising channels. Therefore, it is uncertain whether Chinese millennial consumers have differences in preferences in the aspect of online advertising channels. Simultaneously, there has

4

reason to doubt whether the previous conclusions about preference differences can be applied to Chinese millennial consumers.

1.3 Purpose

In order to explore the gaps in the previous literature, the purpose of this study is to investigate the preferences of Chinese millennial consumers for online advertising channels on the basis of clarifying the current mainstream online advertising channels. Also, the research will focus on discovering whether Chinese millennial consumers in different genders, different regions, and face different product categories have different preferences for online advertising channels. This lead up to the following research questions:

Research Question 1

Which online advertising channel do Chinese millennial generation Taobao users prefer?

Research Question 2

Is there any difference in the preferences of Chinese millennial generation Taobao users of different genders for online advertising channels?

Research Question 3

Is there any difference in the preferences of Chinese millennial generation Taobao users of different living areas for online advertising channels?

Research Question 4

Is there any differences in the preferences of Chinese millennial generation Taobao users for different product types of online advertising channels?

In terms of the contribution, this research aims to provide practical assistance to help companies better understand the millennial consumers in China and their preferences.

5

In addition, this study also attempts to provide some advice on online advertising channels for foreign companies eager to enter the Chinese market.

6

2.literature review

This chapter describes the important literature and concepts related to this research theme and uses online advertising, online advertising channels, millennial consumers, and preferences as keywords to conduct literature searches to collect previous research in related fields. Simultaneously, the above keywords are also used to build the entire logical foundation of this chapter to organize and evaluate previous research. This chapter covers the definition and development of online advertising and online advertising channels and millennial consumers. Meanwhile, it also includes millennial consumers' perception of online advertising channels. Further, related theories of preference are also summarized with the theoretical basis and theoretical framework of this study and the application methods in this study are expounded.

2.1 Online advertising

The pieces of literature indicate the definition of online advertising respectively, among them, Li & Lo (2015) state that online ads are transmitted through online video systems that enable users to watch and even share it. However, the early article provides a more precise definition as deliberate messages placed on third party websites including search engines and directories available through Internet access (Ha, 2008). Meanwhile, the author emphasized that deliberate messages mean that the advertiser intends to place the message on the online medium (Ha, 2008). In the research of Hwang, McMillan and Lee (2003), they interpret that Internet advertising is more than just banners, buttons, and pop-ups. At the same time, company websites are broadly defined as Internet advertising (Hwang et al., 2003). Therefore, the definition of online advertising or Internet advertising is based on the research direction and field of the author, and it can be specific or widely defined. In addition, over time and the development of online

7

advertising, the understanding of online advertising has become diversified. There is an obvious large time span between the above literature, so this is another reason why they have various definitions of online advertising.

2.1.1 Traditional advertising vs. Online advertising

In the earlier literature, Hanekom and Scriven (2002) pointed out the status quo and the development trend of traditional advertising as well as online advertising while giving a detailed division and interpretation of traditional advertising. The authors discussed the characteristics of traditional advertising such as controllability, easy preservation, and high credibility in three categories: print media (magazines and newspapers), live media (radio, movies, etc.) and outdoor media (Hanekom & Scriven, 2002). In another literature, the authors based on the above clear classification of traditional advertising, by showing traditional advertising, F2C (firm to consumer) impressions and C2C (consumer to consumer) social messages on brand building and customer acquisition (Vries, Gensler & Leeflang, 2017). The authors pointed out that traditional advertising is still effective nowadays, more specifically, traditional advertising is the most effective way to affect consumer awareness, consideration, and customer acquisition (Vries et al., 2017). Furthermore, Vries et al. (2017) and Fulgoni (2015) also mentioned that the spread of C2C or F2C on social media can complement traditional advertising. Although the authors affirm that traditional advertising has an undeniable positive impact on brand reputation and brand recognition, it also points out that traditional advertising lacks interactivity.

2.1.2 The development of online advertisement and status quo

Online advertising includes countless delivery platforms, including search engine results, email ads, AD breaks, brand blogs, organic and paid social media ads, mobile ads, and display ads (including banner ads) that appear on desktops and laptop devices (Namin, Hamilton & Rohm, 2017).

8

Banner advertising is currently the most widely used online advertising of all types, which is easy to develop and inexpensive to launch (Papatla & Bhatnagar, 2002; Constine, 2017). The forms mostly come from text and images which are not dependent on the continuous information updates for advertisement companies (Suckkee & Leckenby,1999; Telang, Boatwright &Mukhopadhyay, 2004). However, some studies have shown that banner blindness is one of the key issues to reduce the effectiveness of banner ads, which significance is to expose the existence of such fatal problems as low recognition rate and short recall time in banner ads (Brajnik & Gabrielli, 2010). And it has been regarded by consumers as the online equivalent of direct mail (Breuer & Brettel, 2012). Specific research represents that the click-through rate of banner ads attracting the corresponding frequency between outside has dropped to as low as 0.36 percent ("Home", 2020). Stimulating and clearly targeted banner advertising can reduce the emergence and influence of blind spots to some extent (Brajnik & Gabrielli, 2010). As a result, the advertising market has gradually developed new types of counter ads such as affiliate ads, which review as ‘smart banners’ by (Dou, Linn & Yang, 2001).Compared with ordinary banner ads, smart banners usually have more accurate orientation that reduce the difficulty of addressing the right target group, and most advertising forms do not need high upfront investment cost or operation cost (Dou et al., 2001).

Search engine advertising is based on active autonomous search of consumers to produce visual advertisements that match facts or consumer cognitive information (Breuer & Brettel, 2012). Search-engine ads currently account for 47% of online advertising revenue (Silverman, 2010). Dou et al. (2001) believe that it can be classified as one of the most frequent and mature online advertising methods in ‘smart banner’, since there are specific figures shows that 90% online users tend to use it in their daily life to meet various needs when surfing the Internet (IAB, 2009). Due to the user's own participation for such ads bring excellent positioning results (Nabi & Oliver, 2009), fully extend the service life of advertising (Keller, 1987) and create online advertising

9

interesting to enhance recall effects (Hyun, Gentry, Park & Jun, 2006). Search engine advertising can be divided to two species: search engine optimization (also known as natural rank) and paid search engine advertising (called the keyword advertising), advertisers generally choose the required advertising strategies according to the purpose of delivery (Eisinger, Rabe, & Thomas, 2006; Malaga, 2008). However, according to Dou et al. (2001) studies, search engine advertising can bring more than twice the cost compared to ordinary banner advertising, since it requires constant information maintenance and updating to respond to timely search demand. Moreover, search engine advertisements cannot guarantee to create durable attention, because the impression of advertisements has two sides, which could result in the self-judgment reaction made by Internet consumers when they click (Breuer & Brettel, 2012).

Price comparison advertising focuses on price comparison on internet platforms rather than giving rich information guidance, and its effectiveness is based on the click-through rate of advertising links (Montgomery, Hosanagar, Krishnan & Clay, 2004). Retailers can pay a premium to improve the ranking of advertisements on online platforms to increase online attention, so price comparison ads are more representative of advertising channels than objective tools for market selection (Breuer & Brettel, 2012). Higher price means the higher position of other sponsorship, so leads to advertisers getting more clicks, but also increases the advertisers advertising costs (Nabout, 2015; Küçükaydin, Selçuk & Özlük, 2019). Therefore, advertisers need to ensure that bid for the keyword in an effective way, so that they won't be over budget, at the same time maximize the whole advertising series of potential revenue (Küçükaydin et al., 2019).

Coupons and loyalty advertisements mainly aim to stimulate customers' purchasing behavior that can cultivate customer loyalty through the rebate mechanism of repeat customers and the form of program signing rewards (Breuer & Brettel, 2012). Its design focuses on the attractiveness of the content, which is usually executed in conjunction

10

with the marketing plans of advertising companies (Kim, Shi & Srinivasan, 2001; Liu & Yang, 2009). Both of these two types of ads have obvious functional effects, more suitable to be put in the relative demand of the consumer group, (Breuer & Brettel, 2012) shows that implementing the correct combination of the two types of ads and banner ads from content and structure can improve the final advertising effect. Breuer & Brettel (2012) conducts specific experimental study on the carryover effects of four common online advertising patterns shows that SEM (search engine marketing) has the longest effect, followed by banner advertising, CLA (coupons loyal advertisements) and PCA (price comparison advertisements). The longer carryover effect can suggest that the lengthier the advertising effect and memory cycle, which affect customer handling behavior such as purchase (Nabi & Oliver, 2009; Massaro, Petty & Cacioppo, 1988). In addition, in terms of four online advertising models, Breuer & Brettel (2012) has studied the different influence of previous and new customers that can create the direct sales numbers, which find that CLA works better for existing customers than for new customers, whereas PCA works better to attract new customers.

At present, in order to better segment consumers, the advertising market also divides online advertising types into on-demand ads and push ads according to consumers' prominent characteristics (Rappaport, 2007). Spilker-Attig and Brettel (2010) analyzed that on-demand advertising can make better use of the Internet to effectively interact between consumers and advertisers, and with the development of the online advertising market, consumers are generally inclined to switch from traditional banner advertising to on-demand advertising. Search engine advertising and price comparison advertising are generated by Internet users' own search requests, Rappaport (2007) classified them as a typical type of on-demand advertising. According to the study represented by Spilker-Attig and Brettel (2010), on-demand advertising has a high requirement on the right delivery channel, which determines the final carryover effect of its delivery and avoids some blindness. In other words, the type of push ads focuses on the provision of services, reducing the dependence on the platform (Spilker-Attig & Brettel, 2010). Push

11

advertising diminishes users' participation in the content and, to a certain extent, decreases the credibility of the content spread by advertisements, thus reducing the effectiveness of such advertising (Wai Lai & Liu, 2020).

2.2 Online advertising channels

Categories/classification of online advertising channels

Pretty, Cacioppo and Schumann (1983) developed the Elaboration likelihood model (ELM), showing that there are two "routes to persuade consumers”: a Central Line and a peripheral line. When consumers are motivated to deal with the advertisement, they will be involved in the "central" route, which represents a positive cognitive processing of advertisement input, and led to a more permanent change and can predict behavior (Tam & Ho, 2005). And the "periphery" processing of advertising is under the situation of low motivation, leading to a temporary attitude change, thus we cannot predict behavior (Tam & Ho, 2005). Rappaport (2007) believed that the main characteristic of the ELM, the differentiation between the central and peripheral route, can be adapted to the main differences in online channels that are on-demand channels and push channels. At the same time, push channels focus on dealing with the one-way communication, its characteristic is when browsing the web to interrupt consumers such as E-mail and affiliate banners, while the characteristics of on-demand channels are interactivity (Rappaport, 2007).

The effective promotion of these two online advertising channels is inseparable from the participation of social media platforms. Kaplan and Haenlein (2010) broadly defines social media as "a group of internet-based applications that build on the school and technological foundations of Web 2.0 and allow the creation and exchange of user-generated content”.

Previous conceptual articles on the phenomenon of social media have categorized social media platforms in a number of ways (Kaplan & Haenlein 2010; Kietzmann et

12

al., 2011; Zhu & Chen 2015). Zhu and Chen (2015) developed a typology based on two characteristics of social media: the nature of the connection (profile-based and content-based) and the level of message customization (the extent to which services are tailored to meet individual specific preferences). Voorveld, van Noort, Muntinga, & Bronner (2018) defines two characteristics together and lists four categories of social media:

1. Relationship represents a social media platform based on personal data, mainly composed of customized messages; This category includes platforms such as Facebook and LinkedIn (Voorveld et al., 2018).

2. Self-media media platforms are also personal data based but offer people the opportunity to manage their own social media communication channels. A classic example is Twitter (Voorveld et al., 2018).

3. Creative outlet platforms, such as YouTube and Instagram, allow users to share their interests and creativity based on content (Voorveld et al., 2018).

4. Collaboration platforms are also content based, but allow people to ask questions, get advice, or find the most interesting news and content of the day (Voorveld et al., 2018).

Based on the above classification of social platforms, the corresponding online advertising channels are analyzed to obtain the following categories:

13

Categories of the current mainstream online advertising channels

Categories Definition Example

Mini Banner in App and Web

Advertisements that appear automatically after opening a website or application. Whether the user clicks or not, the ads will appear.

Weibo

Live video

streaming/live broadcast

Absorbing and continuing the advantages of the Internet, using video to live broadcast on the network. The relevant content of the product display can be published on the Internet. It has the characteristics of excellent performance, full content, strong interactivity, unlimitied geographical location, etc. TikTok webcast; Taobao webcast; etc. WeChat public account

Enterprises can realize interaction with users through text, pictures, voice and video on WeChat. It is a mainstream online interactive advertising method.

Influencer

promotion on

social media

The increasingly extended industry chain through product promotion through the celebrities who have a large number of followers and powerful topical on Weibo.

Weibo; Instagram; Xiaohongshu; etc. Influencer promotion on video platform

The increasingly extended industry chain through product promotion through the celebrities who have a large number of

Youtube ；

14

followers and powerful topical on video platforms.

Influencer

promotion on

short video clips platform

The increasingly extended industry chain through product promotion through the celebrities who have a large number of followers and powerful topical on short video clips platforms.

TikTok

Open page of the mobile app

Advertisements that appear automatically when opening the application. Whether the user clicks or not, the ads will appear, but they can click the “skip” button to skip the ads.

Weibo open page

Pre-movie adverts

Video form ads appearing on the video platform before or during the video is played.

YouTube; iQiyi; etc.

2.3 Consumer (millennials) perception of online advertising channel 2.3.1 Generation Cohort: Millennials

Before discussing millennials, a broader concept is the Generational Cohort Theory. It proposes that members of each individual generation have similar values, beliefs, attitudes, and behaviors (Chung, Chen, & Lin, 2015). According to research, there is indeed a significant difference in consumption levels between generations, but such significant levels of difference only occur between different generations and do not involve each individual generation (Eastman & Liu, 2012). They believe that when enterprises plan to segment the consumer market, the generation cohort is used as a variable, which is more meaningful and effective than other variables, such as gender, income, or education. And this research will focus on exploring millennials.

15

According to the Merriam-Webster and Cambridge Dictionary’s definition of millennial, Millennials refer to people born in the early 1980s to the late 1990s. Since they grew up in the Internet age and tend to use online media in their daily lives, this generation knows how to use the Internet effectively as well as social media. Also, they are the largest group of people who use the Internet for online shopping (Muda, Mohd, & Hassan, 2016). At the same time, according to the research summary of Howe and Strauss (2008), millennials have seven important core characteristics: special, sheltered, very confident, team-oriented, conventional, pressured, and achieving. In terms of team-orientation, research has shown that the trend of millennials receiving social influence or word-of-mouth of social media to obtain corporate information will increase significantly in the coming years (Jacobsen & Barnes, 2018).

2.3.2 Consumers' perception of the transition from offline to online advertising channels

This section summarizes that consumers’ perception of the change in advertising channels rather than millennials’, is due to the fact that during the reading of previous literature, most of which did not limit the research population to millennials, but just set it for the majority of consumers (Alalwan, 2018), or just studied about the millennials' perception of online advertising (Agarwal & Arora, 2019). So, this study will conduct research based on existing literature knowledge.

First, digitalization has gradually become one of the main trends that change society and business (Parviainen, Tihinen, Kääriäinen & Teppola, 2017). As the basic element of the digital transformation of society, the development of the Internet has naturally become one of the mainstream trends. In addition, because millennials are hedonistic and highly connected with other people or things (Moreno, Lafuente, Carreón, & Moreno, 2017). For example, in terms of the online travel industry, two-thirds of online users in the United States in 2013 relied on online channels to obtain travel information (Pelsmacker, Dorai-Raj, & Zigmond, 2018). And this generation is more likely to be

16

attracted by electronic discount coupons, and also prefer interactive two-way communication with brand companies (Moreno et al., 2017). So, these factors have also driven the trend of digital business from millennials onwards.

There is no doubt that many online advertising channels have their own advantages. Compared with traditional one-way advertising tools such as newspapers or magazines, the interactive features of online channels can give consumers the opportunity to make a voluntary choice as to whether they are interested in this advertisement. For example, during YouTube advertising time, users do have to watch the pop-up ads, but it only lasts for a few seconds, after which they can choose to click the "Skip Ads" button to give up watching the ads (Belanche, Flavián, & Pérez-Rueda, 2017). In addition, consumers can also use the comments left by other people for reference. A survey shows that about 74% of consumers participate in online media reviews, while 46% of users rely on social platforms to make purchasing decisions (Rust & Oliver, 1994).

However, at the same time, in the process of literature review, it could be found that while consumers have a positive perception of online channels, it also has some problems. One of the most representative is consumer privacy. In order to get rid of the chaotic and complicated Internet advertising environment, advertisers gradually tend to target consumers' social media platform information to accurately deliver highly relevant advertisements (Jung, 2017). However, the results show that although highly relevant ads and consumers ’attention to ads are positively correlated, at the same time, some consumers will also negatively process such highly relevant messages because they think their privacy is being tracked. It is worth mentioning that a survey in 2007 showed that respondents over the age of 20, that is, pay more attention to online privacy than other age groups (Painea et al., 2007). So next we research how the relationship between advertising content and its intermediary improves millennials' perception of online advertising channels.

17

2.4 Theoretical framework 2.4.1 Preference

In 1991, Ajzen and Fishbein proposed the theory of planned behavior and found that preferences can accurately predict and analyze consumers' purchasing behavior. Studies have shown that consumers do have different preferences in different online advertising channels (Schreiner, Rese & Baier, 2019; Anderl, Becker, Wangenheim & Schumann, 2015; Ahn, Lee, Lee & Paik, 2012). However, although the relevant literature confirms that the differentiation is directed to the field of general online advertising, it ignores the specific research of various countries and groups. Therefore, based on this theoretical fact, we will discuss consumer preferences for Chinese online advertising channels and analyze its specific preferences, thus obtaining the main research questions of this study:

Which online advertising channel do Chinese millennial generation Taobao users prefer?

At the same time, the researches prove that consumers' preference choices will have obvious differences according to gender, and preferences will also change according to different commodity types and industry sectors (Schreiner, Rese & Baier, 2019; Ahn, Lee, Lee & Paik, 2012). For example, according to Schlosser, Shavitt and Kanfer (1999), male and female Internet users have different perceptions of Web advertising, meanwhile, Weiser (2000) also indicated that there have different preferences for Internet use patterns among males and females. difference. In detail, the male responds more strongly to graphic advertising stimuli than women. However, the female ’s attention is evenly distributed on graphics and text (Goodrich, 2014). In addition, this study divides the product types into two categories during the literature search process, namely utilitarian products and hedonic products. Compared with the latter, the former is a product that relies more on objective factors (Verhagen, Boter, and Adelaar, 2010), because consumers of utilitarian products are more concerned about how to achieve

18

their ultimate shopping goals and needs in an efficient manner (Childers, Carr, Peck & Carson, 2001), so they will pay more attention to the recommended content in the process of obtaining product information (Verhagen et al., 2010). In contrast, the channel elements of more personalized content will, to a certain extent, influence consumers' impressions of hedonic products (Verhagen et al., 2010). Therefore, based on the theoretical fact that “consumers' preference choices will have obvious differences according to gender, and preferences will also change according to different commodity types and industry sectors”, this study proposes relevant hypothesis regarding gender and product category:

Hypothesis 1: Chinese millennials Taobao users have different preferences for online advertising channels of different categories of products.

Hypothesis 3: Chinese millennial Taobao users have different preferences for online advertising among male and female.

In addition, based on the above theoretical facts and considerations of China's national conditions and actual environment, that is, due to a large number of cities in China, China's urban hierarchy (appendix) has emerged. In detail, the “City Business Charm Ranking” published by CBN Weekly evaluates 337 cities in China through the five-dimensional index of agglomeration of commercial resources, urban hub, urban activity, lifestyle diversity, and future plasticity (CBN, 2019) [Original article is in Chinese, the translation made by researchers]. And the data source is the commercial store data of 170 mainstream consumer brands, the user behavior data of 18 Internet companies in various fields, and the big data of the city form the data agency (CBN, 2019) [Original article is in Chinese, the translation made by researchers]. Therefore, the classification of Chinese cities is inseparable from business, economy, and infrastructure. Among them, it is worth noting that consumer activity indexes such as online shopping activity, social activity indexes such as TikTok activity and consumption potential indexes such as online consumption growth index, are all important indicators for city ranking (CBN,

19

2019) [Original article is in Chinese, the translation made by researchers]. Therefore, the close relationship between different levels of cities and online consumption and social media applications is obvious. The report contains five levels, namely first-tier cities, new first-tier cities, second-tier cities, third-tier cities, fourth-tier cities, and fifth-tier cities (CBN, 2019) [Original article is in Chinese, the translation made by researchers]. According to the 2019 City Business Charm Ranking, the specific city levels are divided as follows (Appendix B).

As mentioned above, the hierarchical mechanism contains diversified data, not just single data such as GDP or per capita income. Therefore, the connection between city level and the enterprise or business is inseparable. As an important part of business, the online advertising channel maybe has the link with city levels. However, for now, there is no research to show whether consumers in different city levels have different preferences for online advertising channels.

The specific ranking is inseparable from the business, economy and infrastructure of each city. Therefore, this study puts forward the corresponding hypothesis, that is, different preferences may come from different influences of the residential area, the resulting customs, differences in consumption capacity, etc., so the following hypotheses are proposed:

Hypothesis 2: Chinese millennial Taobao users living in first-tier cities, new first-tier

cities, second-tier cities, third-tier cities, and other regions have different preferences

for online advertising channels.

Finally, it is worth mentioning that Zeithaml (1988) believes that consumer preference is a kind of psychological evaluation. This process includes a series of psychological processes such as perceived risk, perceived satisfaction, and product evaluation caused by a product's perceptual attitude. Therefore, preference can be understood as a

20

subjective emotional attitude, which cannot be directly explained by objective experimental data. Therefore, this study will also apply the AISAS model as an auxiliary model in order to reflect consumption preference from the perspective of consumer behavior, so as to avoid cognitive errors caused by the subjective assumption of studying preference directly.

2.4.2 AISAS model

AISAS model is a consumer behavior analysis model proposed by Dentsu Group (Kuang, 2012). It is a new theory based on the development of today's Internet and wireless applications. Specifically, it refers to attention, interest, search, action, and sharing respectively. Surveys conducted by Sugiyama and The Dentsu Cross Switch Team (2010) show that consumers who are in the stage of product awareness are mainly sources of product information through television, magazine or online media advertisements. Therefore, as an effective online platform for enterprises to reach consumers, the online advertising channel can be used to specify preferences by using consumer behaviors, thereby helping this research to explore consumer preferences for online advertising channels.

In this study, this theoretical framework will mainly help researchers design the questionnaires. As previously mentioned, because preference is a subjective emotional attitude and cannot be directly explained by objective experimental data. Therefore, this study uses the AISAS model, that is, to specify preferences through the perspective of consumer behavior, thereby designing a questionnaire, finally, it provides a more convincing reference for the data analysis of this research.

Specifically, four of the elements in the model, namely attention, interest, search and action, will be expanded and defined in this study as five follow-up actions for consumers’ perception of online advertising channels: ignoring the advertise, interested in and search for relevant product information, add products to shopping lists, and direct

21

purchases, which can help this research to analyze data to determine the different preferences of Chinese millennial consumers for online advertising channels.

22

3. Methodology

This chapter introduces the research method and design to be used in this study. And the sampling method, data collection method and data analysis method are also explained. Finally, the quality reliability of the research method in this paper is discussed.

3.1 Research approach

Research philosophy is the belief about the way in which data about phenomena should be collected, analyzed and used, which involves the source, nature and development of knowledge (Bajpai, 2011). Research philosophy often reflects the author's important assumptions that can serve as the basis for a research strategy ("Research Philosophy - Research Methodology", 2020). Within the scope of business research, there are four main types of research philosophy: pragmatism, positivism, realism, interpretivist. The correct choice of research philosophy depends mainly on whether the research focuses on the difference between facts or numbers, in other words from the choice of quantity and quality.

The study will conduct from a positivism-paradigm, which originated in the natural sciences. As a philosophy, positivism is in line with the empiricist view that knowledge comes from human experience (Collins, 2010). It has an atomic, ontological perspective of the world, including discrete, observable elements and events that interact in an observable, deterministic, and regular manner (Collins, 2010). Positivism paradigm rests on the assumption that social reality is singular and objective and is not affected by the act of investigating it that is therefore highly applicable to the research since its

23

analysis objective social reality and the knowledge comes from objective evidence about observable and measurable facts (Collis & Hussey, 2009). Which means this study will relate to the viewpoint that researchers need to concentrate on facts instead of the meaning and has provision for human interest that are explained by phenomenology ("Positivism - Research Methodology", 2020).

Crowther and Lancaster (2008) argue that as a general rule, positivist studies usually adopt a deductive approach. A deductive approach involves "developing hypotheses (or hypotheses) based on existing theories, and then designing research strategies to test hypotheses" (Wilson, 2010). Gulati (2009) believes that deductive means from special to general reasoning. If causality or connection seems to be implied by a particular theory or case example, it may be correct in many cases. The deductive design may test to see if this relationship or link did obtain on more general context (Gulati, 2009). Deductive methods can be explained by hypothetical methods, which can be derived from theoretical propositions. In other words, deductive methods involve conclusions inferred from premises or propositions ("Deductive Approach (Deductive Reasoning) - Research-Methodology", 2020). Deductive methods can explain the possibility of causality between concepts and variables, the possibility of quantitatively measuring concepts, and the possibility of generalizing research results to a certain extent ("Deductive Approach (Deductive Reasoning) - Research-Methodology", 2020). Further, the research involves a deductive process with a view to providing explanatory theories to understand social phenomena which is appropriate for the research since it regards some models and theories as the research base measurements for results (Gorton, 2020).

Conducting research based on a positivism-framework implies the use of a quantitative method of analysis. Quantitative research can be described by Bryman and Bell (2015) as "an opinion that needs to collect numerical data and exhibiting the relationship between theory and research, is a deduction, a predilection for natural science methods,

24

and a concept of social reality with objectivism”. In other words, quantitative research mainly uses statistical techniques to examine the relationship between numerically measured variables.

Quantitative data collection methods are based on random sampling and structured data collection instruments ("Quantitative Data Collection Methods - Research-Methodology", 2020). The results of quantitative research are usually easy to present, summarize, compare and generalize ("Quantitative Data Collection Methods - Research-Methodology", 2020). The study will use the post behaviors of consumers through designing appropriate surveys, then imply the outcomes to reflect the consumer preference further. Hence, a quantitative approach will guide through which methods to use and how the data is collected in addressing the research question. In addition, it will set a logical structure which will ensure credible research.

In order to approach the chosen research question as effectively as possible, a quantitative design with a survey study methodology for the study is appropriate. The essence of the survey methodology can be interpreted as "inquiring individuals on one or more topics and then describing their responses (Jackson, 2011). In business research, survey methodology for major data collection is used to test concepts, reflect people's attitudes, establish customer satisfaction levels, conduct segmented research, and a set of other purposes ("Survey Method - Research-Methodology", 2020). Survey can be broadly divided into three categories: mail survey, telephone survey and personal interview (Trochim, 2020).

In other words, a survey is designed to collect primary or secondary data from a sample with a view to generalizing the results to the chosen population, with the personal interview which allows the study to have direct observation of consumers' reflections in terms of various online advertising channels that the study focuses on ("Primary Data & Secondary Data: Definition & Example - Statistics How To", 2020). Finally, the

25

study will analyze the final results through the graphic illustration with the combination of various data analysis strategies such as inferential statistics and descriptive statistics that help the study encode the data from methodology.

3.2 literature review

The literature review serves as an important theoretical basis for this research. The main resource of this literature review is periodicals in related fields and relevant pieces of literature in this field are collected as comprehensively as possible for review. In order to better collect references, online advertising, online advertising channels, Chinese millennial consumers, and preferences are used as keywords for literature retrieval. The existing literature search is mainly carried out through the databases of Google Scholar, Scopus, and Jönköping University’s library.

Through the review of the existing research, two problems/conclusions regarding preference were found. First, preference itself is subjective. Therefore, the direct application in research will lead to misunderstandings and deviations and make the research fail. Therefore, in order to solve this problem, the researchers introduced the AISAS Model as the basic tool for research and developed four follow-up actions based on this, namely ignore, interest & search, adding to the shopping cart, and purchasing. These four follow-up actions are used as the main basis for constructing online questionnaires to evaluate consumers' preferences for different online advertising channels. The four follow-up actions established by introducing this framework can help this study promote the collection of statistical data and increase the credibility of the research.

Second, through the literature review, it is found that there are differences in preferences, namely gender, different living areas, and different product types. However, the existing research takes the entire consumer group or consumers in Europe and America as the research object. Therefore, it is impossible to determine whether

26

there are differences in preferences among Chinese millennial consumers. If so, whether the differences are the same as the conclusion of previous literature. Therefore, three hypotheses of this study are proposed.

Furthermore, because the recent foreign journals and articles in this field are very limited, so to ensure the real-time nature of theory and information, some Chinese literature is introduced for research in the literature review stage. The Chinese references are translated into English by researchers for research. In order to avoid expression differences between different languages, all translations will be verified and checked by a linguistic major student. Meanwhile, the note will be placed behind the translation to inform the original language of the source.

3.3 Sampling

This research will select convenient sampling and combine it with snowball sampling to obtain more comprehensive data. Snowball sampling refers to that the investigator distributes the questionnaire to people who meet the research object and encourage the respondent to pass the questionnaire to people with similar characteristics (Collis & Hussey, 2013). The estimated sample size for this study is 120-160. However, due to snowball sampling, around 200 samples may eventually be collected. Since the research object is Chinese millennial consumers, the study will follow Merriam-Webster and Cambridge Dictionary’s definition of millennial, that is, people born in the 1980s and 1990s. The questionnaire will be sent to the researchers' Chinese friends of the millennial generation as well as posted it on social media platforms such as Weibo and WeChat. In order to ensure the validity of the sample, the keywords of the questionnaire, namely "millennials" and "Chinese consumers" will be clearly marked in the front title of the questionnaire. At the same time, the questionnaire will be asked about the age range of the respondents to screen out samples that do not meet the subject of this study. In addition, in order to reduce misunderstandings and differences within language, the questionnaire will be conducted in Chinese. After the data collection is

27

completed, the questionnaire will be translated into English and verified by a linguistic major student. However, the original questionnaire will also be kept referring to.

3.4 Data collection

This research will use online questionnaires as a method for collecting primary data. Hence, this study will select the Tencent survey as the questionnaire production platform. The platform can enable respondents in China or abroad to access smoothly and the Chinese are allowed. In addition, its data visualization and filtering functions could help to monitor data collection in real-time. The questionnaire will focus on multiple-choice questions to help subsequent data analysis be more intuitive. Respondents will answer questions about online advertising, online advertising channels, and their preferences and perceptions. Meanwhile, the respondents are asked about their age range, gender, area of life, and whether the user of TaoBao. These problems will bring a more intuitive data foundation for the subsequent analysis of data and could help to filter the invalid sample.

The questionnaire will be divided into two main parts. The first part will focus on basic questions. Respondents will first be asked if they are Chinese citizens. This question is used to screen respondents who are not Chinese since it cannot be excluded that the situation that foreign people can correctly recognize mandarin. Secondly, the respondents will be asked whether it is a Taobao user and the year of birth. It is worth noting that the age in the questionnaire is avoided in order to more accurately determine whether it is a millennial. In addition, this part also contains the gender question. However, in the questionnaires of western countries, most of the options include female and male, polygender, and transgender, etc. However, in most cases, there are only two options for female and male in the questionnaire in China, so this questionnaire will only involve these two options. Also, the gender in the later analysis stage will be used as a research dimension to study whether male and female Chinese millennial Taobao users have different online advertising channel preferences, therefore, the setting of this

28

option will be more conducive to data analysis and conclusion inferred. Furthermore, the frequency of using Taobao shopping and the frequency of contacting online advertising will also be asked. In order to reduce the personal bias and misunderstandings of frequency by respondents, the frequency will be marked in detail. More specifically, the frequency involves different levels such as never, rarely, sometimes, and so on. However, the respondent has a personal understanding of the definition of each frequency, which may lead to errors in the study. Therefore, it is marked in detail behind each frequency to assist gather accurate data, for instance, "Always" - more than eight times a week.

The second part of the questionnaire will mainly design questions related to various online advertising channels and linked to different product categories and follow-up actions. At the beginning of this part, a brief description of each online advertising channel and practical cases will be provided in order to help respondents understand each online advertising channel correctly. The follow-up actions of receiving online advertising in different ways from different channels will be the focus of this part. In simple terms, it aims to investigate actions after watching online ads, such as buying immediately, favorite, adding to a shopping cart, etc. At the same time, the question will introduce different product categories as a research dimension, in order to study whether Chinese millennial generation Taobao users have different preferences for online advertising channels when facing different product categories.

Based on the research questions and the classification of online advertising channels, relevant survey questions will be designed. And after the formation of the preliminary questionnaire, it will first check whether there contain ambiguities and negative questions. Then it will be distributed to a small sample for testing. The small sample is expected to be 5 people. These five people will be selected from friends/family around the researchers and have to meet the research objects of this study, that is, Chinese, millennials, Taobao users. The personal feedback and suggestions will be collected

29

after they answer the questionnaire. Also, they will be asked whether it clearly understands the purpose of this research and the key information contained in it and whether there are misunderstandings or incomprehensible wording in the questionnaire. In addition, the answers to the questionnaire and the length of time it takes will also be collected to use to adjust and modify the length of the questionnaire to avoid failed collect data and invalid results. To test this questionnaire before fully distributing it could help to avoid the misleading, jargon and ambiguity question and improve the success rate of data collection.

After completing the modification and adjustment, the final version of the questionnaire will be formed and distribute it through the Tencent survey platform. The questionnaire will form a link, which will be sent to the researcher's friends and social media platforms that meet the research object. During the distribution process, the data collection situation and the respondent feedback will be monitored in real-time. The questionnaire will be distributed for about one week, but the specific time will depend on the questionnaire collected and will stop distribution when the returned survey reaches around 190. Due to the consideration that there may be cases that the questions are not answered as required. Therefore, after the collection is completed, invalid questionnaires will be screened according to the average number of their response time, more specifically, those whose response time is below the average will be considered invalid. Then, a second screening is conducted based on the age of the respondent to select respondents who do not meet the millennial generation as well as Chinese nationality. At the same time, according to the answer of whether the user of Taobao to filter the respondents who are not the user of it. Finally, the final data for the next analysis will be formed.

30

3.5 Problem Formulation

As mentioned earlier, this article has four research questions, and the first research question will directly draw conclusions and discussions through data analysis, without making assumptions. In terms of the rest of the three research questions, which are:

Research Question 2

Is there any difference in the preferences of Chinese millennial generation Taobao users of different genders for online advertising channels?

Research Question 3

Is there any difference in the preferences of Chinese millennial generation Taobao users of different living areas for online advertising channels?

Research Question 4

Is there any differences in the preferences of Chinese millennial generation Taobao users for different product types of online advertising channels?

will set up three hypotheses. The hypothesis for these research questions are as follows:

H1: Chinese millennials Taobao users have different preferences for online advertising channels of different categories of products.

H2: Chinese millennial Taobao users living in first-tier cities, new first-tier cities,

second-tier cities, third-tier cities, and other regions have different preferences for

online advertising channels.

H3: Chinese millennial Taobao users have different preferences for online advertising among male and female.

31

3.6 Method of analysis

After the sample screening, the data for the final analysis will be obtained. Based on the four research questions of this study, the analysis will be divided into two main parts. The first, the proportion will be used, which is the proportion of all respondents ’preference for each channel in the total preference, then compare each other to determine which is the most popular online advertising channel for Chinese millennial Taobao users.

The second part will apply Chi-square test to test the three hypotheses. Due to the variables involved in the hypothesis proposed by this study, such as gender, product category and living area belong to disordered nominal variables, also, this study tested whether these variables have differences between two groups or multiple groups. Therefore, Chi-square test is considered to be the appropriate analysis method to test the hypothesis in this research.

3.7 Trustworthy consideration

The trustworthiness of quantitative research reflects on the validity and reliability of findings to ensure research quality. There are four main criteria for Validity: Face validity shows whether the problem is related to the purpose of the research on the surface; content validity focuses on whether the concept can be completely covered; criterion-related validity lies in whether the measurement method can be effectively judged or as a correct criterion for measuring; constructing validity concentrates on whether the research question is consistent with the research topic. While reliability focuses on the stability and consistency of response measurement results. Internal validity, external validity, reliability and objectivity are the four specific criteria in pursuit of a trustworthy study (Trochim, 2020).

How well the threats to internal validity have been controlled, and the validity of the instruments and measurements used in a study is the primary method to establish trustworthiness (Malakoff, 2012). Internal validity is supported when the dependent

32

variable changes only on the independent variable and not on other confounding variables (Malakoff, 2012). Keep following possible threats to internal validity: history, selection, maturation, pretesting, instrumentation, treatment replications, subject attrition, statistical regression, diffusion of treatment, experimenter effects, and subject effects (Malakoff, 2012).

External validity was used to generalize the study sample to a larger population. Use external forms of validity, such as statistical confidence limits, to make reasonably accurate claims (Malakoff, 2012). Quantitative researchers must examine the following factors that may affect external effectiveness and generality: subjects, context, time, interventions, and measures (Malakoff, 2012).

Prove reliability by checking the consistency of a set of measurements or the measurement instruments used in the study. Retest also known as the stability method to demonstrate reliability, in which a measure is performed on a group of individuals, waited for a period of time, and then read the same measure again to the same group of individuals (Malakoff, 2012). Equivalence is a measure that can be used to apply two forms of the same test to a group of individuals and then correlate the scores of the two groups of individuals (Malakoff, 2012). Equivalence and stability assessments are another way to test reliability by providing one form of instrument to the same group of people over time (Malakoff, 2012). Consistency is another way to measure reliability by having the evaluator observe the same behavior and check whether they have similar and consistent results. Reliability can be reviewed as the basis of validity in quantitative research since it is a measure of whether a study gets the same result every time to achieve the universality of research results.

Quantitative research focuses on the facts from an objective perspective, and it also needs to ensure that the participants do not interfere in the research process to eliminate the emotional bias and other interfering factors to reduce the accuracy of the test results.

33

Objectivity is performed through methodological procedures such as instrumentation and randomization (Malakoff, 2012).

Strive to show trustworthiness in the research by following the methodology presented. In approaching individuals for the survey, provide them with rights such as participation rejection, anonymity, and confidentiality. Anonymity focuses on protecting the personal identity of participants, while confidentiality provides protection for the collected information related to researchers (Bell & Bryman, 2007).

3.8 Ethic

In order to enable participants to provide data related to the research without any sense of restraint, anonymity is regarded as an important means of safeguarding rights. At the same time, the existence of confidentiality solves the participants' problems related to the possible disclosure of personal privacy in the process of participation. The provision of these rights can make participants understand the degree of trust in this study and maximize the provision of meaningful experimental data to obtain more accurate research results. I.e., Give a chance to those people who like to participate instead of offering financial or other rewards to attract participants. At the same time, respondents have the right to withdraw halfway. In addition, the personal information contained in it will only be used for this study.

34

4. Result

This chapter introduces the collection results of the online questionnaire. The specific content of this chapter is connected with the questionnaire. In order to more intuitively display the data distribution and survey situation, this section adopts data visualization that is presented in the form of the pie chart. (The questionnaire is available in appendix A.)

4.1 Overview of the data

According to the statistics of the total number of questionnaires, we collected a total of 227 questionnaires. In order to be consistent with our research on demographic characteristics (millennials, Chinese nationality, Taobao users) and research questions, we selected 150 valid responses for data analysis. Of the 150 responses, 71 (47.3%) were men and 79 (52.7%) were women, according to the distribution of the male-to-female ratio. According to the pie chart 1, it could be intuitively seen that the proportion of male and female respondents to this survey is almost equal. As a result, single sex has less impact on the final data of the entire sample.

Secondly, among the 150 effective responses, 44 people in the new first-tier cities accounted for 29.3%, and 40 people in the first-tier cities accounted for 26.7%. In the second - and third-tier cities, 50 and 16 are respectively, accounting for 33.3% and 10.7% respectively. Therefore, according to the pie chart 2, the respondents living in the third-tier cities are the least, while the population in the other three regions is almost the same.

35

Pie chart 1 Pie chart 2

According to respondents' monthly shopping frequency of Taobao, the questionnaire defines 1 as never used (0 time) and 5 as very often used (8 times or more). As can be seen from thepie chart 3, the majority of respondents use Taobao to make purchases about four times a month.

Pie chart 3

In addition, the frequency of daily exposure to online advertising was defined as 1 point - never (0 times) and 5 points - very frequent (8 times or more). Among the 150 respondents, the vast majority of the respondents (94.7%) are exposed to online

36

advertising. And nearly half of them “viewed” online advertising 4-5 times per day. Therefore, the conclusion can prove the universality of online advertising and the necessity of this research question.

Pie chart 4

Also, This questionnaire surveys eight different online advertising channels, and defines four possible follow-up actions for consumers：ignoring ads, feeling interested and searching for relative product information, adding products to shopping carts and direct purchase . At the same time, the questionnaire divides the advertising products involved into food products, clothing and luggage products, electronic products, household products and others (e.g.cars).