Research in Electronic Commerce Frontiers (RECF) Volume 1 Issue 3, November 2013

www.seipub.org/recf

The Key Factors Affecting Consumers' Trust to Online Group-buying Gao Xirong*1, Chen Yubao2, Chen Qiang3 School of Economics & Management, Chongqing University of Posts and Telecommunications Chongqing 40065

1,2,3 1,2,3

[email protected]; [email protected]; [email protected]

*1

Abstract A structural equation model (SEM) of factors influencing consumers' trust to online group-buying (CTOG) was proposed. The factors included website community, customer response, website security, seller competence, seller integrity, seller benevolence, Self-efficiency, and scenario normality. The model was fitted based on the data obtained from a questionnaire survey. The results showed that seller benevolence, scenario normality, website security, customer response and website community had a significant impact on consumer trust separately. Based on these results, suggestions were put forward for websites and sellers to improve consumer trust in online group-buying. First of all, sellers should build benevolent image; secondly, sellers and online group-buying websites should work together to ensure the scenario normality; thirdly, online group-buying websites should commit themselves to website security, customer response and website community operation. Keywords Online Group-buying; E-Commerce; Consumers' Trust

Introduction Online group-buying refers to a new type of shopping online that consumers trade with sellers in groupbuying websites through the information exchange platform-Internet (Anand & Aron, 2003). The core of the online group-buying is requirements for gathering and volume discount. According to China Group-buying Statistics for 2012 released by www. Lingtuan.com, the authoritative group-buying navigation website, there were 5,988 group-buying websites at the end of December in 2011 in China while the number decreased to 2,976 at the end of June in 2012 (China eBusiness Research Center, 2012) and over half of these websites were turned off, which indicated that the group-buying industry has been in a transitional period from the bubble to the integration. During the transitional period, the online groupbuying industry exposed serious credit problems. According to the Report for User Experience and

Complaints and Detection of Q3 China's E-commerce in 2012 released by China's E-business Research Center, in the third quarter of 2012, the center received 23,156 complaints from the national e-business users, 5,419 of which were online group-buying and took up 23.4% of the total, ranking in the second place (China e-Business Research Center, 2013). The 5,419 complaints were mainly about fraud, counterfeit products, service shrink and privacy information disclosure, etc. The related researches showed that lacking of trust was the primary factor influencing consumers' group-buying. Therefore, it is of great importance to study the influencing factors of consumer trust in online group-buying. In light of the existing researches, scholars mainly conducted their researches on online group-buying from the following three aspects. Researches in the first aspect focus on discussing whether the flexible group-buying price mechanism is superior to the fixed traditional price mechanism. Results showed that when the requirements of consumers are not clear (Chen et al., 2004), sellers are willing to take risks and hope to expand markets of new products, and the number of demanders is higher in low price than that in high price (Chen et al., 2007), the flexible groupbuying price mechanism is superior to the fixed traditional one. Researches in the second aspect are mainly about approaches adopted by group-buying websites to gather sufficient consumers, including: 1) products of the same category instead of a certain product (Chen at el., 2010), 2) reliance on trust to build a long-term group-buying formation mechanism (Yamamoto and Sycara, 2001), 3) reliance on trust to gather consumers (Breban and Vassileva, 2002), 4) constructing consumers union organization to allow consumers to announce acceptable prices (Yuan and Lin; 2004), 5) sellers' reservation price and setting up bulk discount allocation system (Li et al., 2010). Researches in the third aspect focus on probing factors of consumers' acceptance to group-buying system and

37

www.seipub.org/recf

Research in Electronic Commerce Frontiers (RECF) Volume 1 Issue 3, November 2013

influencing consumers' adoption intention and behavior, including: 1) information of purchase decision (purchase conditions and prompt messages) (Mastuo and Ito; 2004), 2) incentive mechanism (Tan and Sutherland; 2004), 3) text comments about sellers' previous group-buying activities and current orders for goods (Kauffman et al., 2010a, 2010b).

expectation to his self-confidence under risky online environment. The expectation means that the trustee is expected to not expose the weakness of the subject under the risky online environment (Corritore et al., 2003). Once consumers have trust in a certain website, they will accept the website's credit and believe that the website will not deceive their money.

Chinese scholars laid emphasis on business model, current developing situation and tendencies, chances and challenges and existing problems in their researches on online group-buying. So far, Qin and Wan (2011) analyzed why the online group-buying fraud was formed under the consumer herd behavior theory, and showed that the fact that group-buying websites only provide the number of consumers who have participated in group-buying would cause incomplete information and thus making other consumers blindly believe and follow group-buying, leading to online fraud. Ning and Zhang (2011) constructed the influencing factor model of consumers' impulse willingness to join online group-buying on the basis of traditional consumer impulse buying behavior theory, and indicated that product features, scene characteristics and consumers' personalities have positive correlation with consumers' impulse buying willingness in online group-buying and that product features have the most obvious impact effect.

According to the current literature, factors that influence consumers' trust can be classified into three kinds: subject, object and interactive factors. Shankar and other scholars (2002) summarized the three kinds of factors that influence the online trust. The subject factors refer to features of consumers, including shopping experience online, propensity to trust and technical capacity etc. The object factors refer to features of websites and sellers. Features of websites consist of history of the website, online community, credit of website, internet security and the third party certification marks etc., while features of sellers include scale and reputation etc. The interactive factors contain responsiveness of clients and communicational ability etc.

In general, the current literatures didn't make deep research in the key factors of restraining online groupbuying consumers' trust and lacked solid evidence. Therefore, this thesis constructed the influencing factor model of consumers' trust in online groupbuying based on previous researches, demonstrated and tested the constructed theoretical model and a series of relevant hypotheses and found out the key factors influencing online group-buying consumers' trust on the basis of questionnaire survey data. The result of this study will improve consumers' trust from websites and sellers. Model and Hypotheses Research Review of Trading Trust From the time immemorial, any trading activities can't be done without trust. In traditional trade, once trust was built, the consumer would be willing to take a risk, meaning that even though the consumer could expect that his/her own trust behavior would harm himself, he was still willing to expose his own weaknesses to the trustee (Mayer et al., 1995). Similarly, in online trading, trust refers to the attitude that a consumer has

38

According to the research made by Luhmann (1979), when society cannot get through rules and conventions decrease, individuals will take trust as the major way to reduce social complicacy. In contrast with traditional trading, online group-buying trading is short of regulations and conventions, and products or service of group-buying cannot be checked and touched immediately, so trust is much more complicated and important in online group-buying than in traditional business. Concept and Hypotheses 1) Website Features and Consumer Trust Website community refers to online communication space, including micro-blog, forum, customer evaluation system, online chat and personal space etc. The website community of the same theme attracts members who share the same interest. These members interact frequently and share their feelings, experiences and opinions. What's more, they are willing to adopt other members' suggestions. Urban et al. (2000) believed that consumers' perceived risk can be reduced and consumers' trust can be built by reflecting various users' feedback through online community. Customer response refers to consumers' perception to the promptness and interactivity of website response (Chen et al., 2008), as well reflects the quality and professionalization of websites' service.

Research in Electronic Commerce Frontiers (RECF) Volume 1 Issue 3, November 2013

It can help build consumers' trust, if websites quickly make responses to consumers' inquiry and properly deal with consumers' problems. Gefen and Straub (2003) proved that customer response has a positive effect on consumers' trust by means of empirical study. Website security means not only that a website does not expose consumers' privacy, not falsify and destroy trade information, but also means payment's not going wrong etc (Hoffman and Novak; 1996). Website security is a crucial factor that influences online consumers' trust. When online consumers make up their minds to buy, what they care most is whether the trade is safe or not and whether their privacy and money are protected effectively or not. Koufaris et al., (2004) proved that website security has positive influence on consumers' trust. Thus, the following hypotheses are put forward: H1: Website community has a positive impact on consumers' trust. H2: Customer response has a positive impact on consumers' trust. H3: Website security has a positive impact on consumers' trust. 2) Seller Features and Consumer Trust Seller competence refers to the consumers' sensation of sellers' ability and knowledge on accomplishing expectations. In group-buying, this kind of sensation depends on two points: one is whether sellers have the ability to accomplish the established behaviors; the other is whether sellers have approaches to guarantee established behaviors. If the two points are not satisfied, consumers' trust will be affected. Seller integrity, also called seller honesty, refers to the consumers' sensation of a series of conventional regulations during the process of trading. This sensation will make consumers build the trust to sellers little by little. In online group-buying, these conventional regulations include how sellers manage trading and sellers' service regulations for consumers' sensitive information etc. If sellers can well obey these regulations, it will help shape consumers' trust. Seller benevolence means that consumers can sense that sellers not only look out for their own interests but care about consumers' benefits. Generally speaking, well-meaning sellers would like to help consumers even without any rewards. Seller benevolence is a kind of altruism which lessens uncertainty and avoids opportunism. If

www.seipub.org/recf

consumers perceive sellers' kindness, sellers will get more trust from consumers. In light of the above analysis, the following hypotheses are put forward: H4: Seller competence has a positive impact on consumers' trust. H5: Seller integrity has a positive impact on consumers' trust. H6: Seller benevolence has a positive impact on consumers' trust. 3) Scenario Features and Consumer Trust The scenario features of this thesis mainly refers to scenario normality which means that whether the trading is successful or not depends on the customs and regulations shown in the process of trading (Baier; 1986). Scenario normality reflects that there is no abnormal or dangerous situations that affect normal trading. In online group-buying, if groupbuying websites and sellers can plan purchase rules in accordance with consumers' previous buying habits, group-buying websites and sellers will gain consumers' trust. It can be seen that scenario normality is related to online shopping websites as well as sellers. Therefore, the following hypothesis is put forward: H7: Scenario normality has a positive impact on consumers' trust. 4) Self-efficiency and Consumer Trust Self-efficiency refers to individual's sensation about his/her own relevant ability to be engaged in certain behavior and achieve the expected results in specific circumstances (Holsapple, et al.; 2005). Generally speaking, successful experience can strengthen self-efficiency while otherwise for repeating failure. In online group-buying, consumers are required to own definite computer skills, such as searching products or service, registering or logging in, ordering and purchasing according to the displayed information of websites. The stronger self-efficiency the consumers have, the more confidence they will have and the less problems they will meet and the more quickly trust will be set up. Holsapple et al., (2005) has proved that selfefficiency has a positive impact on consumers' trust. In light of the above, the following hypothesis is put forward: H8: Self-efficiency has a positive impact on consumers' trust.

39

www.seipub.org/recf

Research in Electronic Commerce Frontiers (RECF) Volume 1 Issue 3, November 2013

5) Consumer Trust and Usage Intention Usage intention refers to the intended use of a online group-buying website to perform any transactions. The current research showed that online trust obviously influences consumers' usage intention (Kim et al., 2004). Kim et al., (2004) has proved that online trust positively influences consumers' usage intention through an empirical research. Thus, the following hypothesis is put forward: H9: Consumers' trust has a positive impact on consumers' usage intention. According to the above hypotheses, this paper comes up with consumers' trust to online group-buying (CTOG) model shown in Figure 1:

students and young people included: firstly, young people's ability to accept new things is stronger than middle aged and elderly people who are over 40 years old. Secondly, they are familiar with computer and internet, so it is easy for them to communicate and trade through internet. Finally, they are the major groups of online group-buying. This study adopts paper questionnaire and electronic questionnaire methods. The number of paper questionnaires sent out is 400 and collected back is 362. The number of the electronic questionnaires sent out is 150 and collected back is 150. Therefore, the total number of collected questionnaires is 512 and the response rate is 93.09%. After excluding the invalid ones, 496 questionnaires are used for analyses in the study. The valid response rate is 96.87%. Descriptive Statistics

FIG. 1 RESEARCH MODEL AND HYPOTHESES

Questionnaire Design and Survey Questionnaire Design To test the research model (see Figure 1), a 37-item survey questionnaire was developed, 30 of which were designed to measure the latent variables as follows: the website community (WC), the customer response (CR), the website security (WS), the seller competence (SC), the seller integrity (SI), the seller benevolence (SB), the self-efficiency (SE), the scenario normality (SN), and the truest. To fit to these particular research subjects, the 30 items are adapted, compiled and modified from existing researches, and then measured on a 5-point Likert scale of ”1=strongly disagree, 2=disagree, 3=neutral, 4=agree, 5=strongly agree”. The remaining 7 items are participants' characteristics, including their gender, age, education background, online shopping experience, mean monthly expenditure on online shopping, online group-buying experience and whether they conduct or re-conduct group-buying. Before formal survey, fifty people were invited to participate in pre-survey, and the items were adjusted and modified based on participates' opinions. Questionnaire Survey The reasons for the selection of graduated college

40

There are 251 male respondents and 245 female respondents. 79.44% of the respondents are 21-30 years old. About 96% of the respondents are above bachelor education. 95.57% of the respondents are above bachelor and most of them are collage students. 95.76% of the respondents have shopped on the internet and 64.72% of the respondents have shopped on online group-buying websites, in which 71.17% of the respondents spend 1-199 RMB. 92.14% of the respondents will participate in online group-buying. Data Analysis Validity Analysis This study used exploratory factor analysis (EFA) to examine the questionnaire validity. SPSS17.0 software was run to analyze the paper. The KMO-value was 0.899 and Bartlett Test was significant (Sig = 0.000), indicating that the questionnaire is suitable for factor analysis. Ten constructs were extracted through principal component analysis method and Varimax orthogonal rotation method and their total variance explained was 66.33%. As it can be seen in Table 1, the loading of each item in its construct is above 0.5. So it can be concluded that the model has researched convergent validity and discriminant validity. Reliability Analysis Cronbach's α is used to test the questionnaire reliability and the results are shown in Table 2, from which it can be seen that all constructs' Cronbach's α values are higher than 0.6, indicating that the constructs are internally consistent.

Research in Electronic Commerce Frontiers (RECF) Volume 1 Issue 3, November 2013

www.seipub.org/recf

TABLE 1 ROTATED COMPONENT MATRIX OF CONSTRUCTS (N=496)

Construct

Item Q20 Q21 Q19 Q8 Q9 Q7 Q10 Q25 Q26 Q27 Q16 Q18 Q17 Q29 Q28 Q30 Q4 Q5 Q6 Q2 Q3 Q1 Q11 Q12 Q23 Q24 Q22 Q13 Q14 Q15

Seller integrity

Website security

Trust

Seller competence

Usage intention

Customer response

Website Community Self-efficiency Seller Benevolence

Scenario normality

F1 .780 .703 .687

F2

Construct

Loading F5 F6

F7

F8

F9

F10

.791 .714 .595 .753 .746 .714 .786 .747 .735 .784 .652 .585 .775 .705 .699 .866 .860 .678 .523 .519 .684 .675 .535

Number of items Cronbach's α 3 3 4 3 3 3 3 2 3 3

F4

.694 .669 .615 .608

TABLE 2 RELIABILITY STATISTICS OF CONSTRUCTS

Website community Consumer response Website security Seller competence Seller integrity Seller benevolence Scenario normality Self-efficiency Trust Usage intention

F3

0.625 0.640 0.704 0.748 0.784 0.713 0.656 0.724 0.772 0.787

Model Fit and Hypothesis Test

are showed in Figure 2. Fit indices show that the chi-square/degree of freedom ratio ( χ 2 /df) is 2.182 less than 3, the root mean square error of approximation (RMSEA) is 0.049 less than 0.05, the goodness of fit index (GFI), the comparative fit index (CFI) and non-normal fit index(NNFI) are more than or equal to 0.90 (see Table 3) indicating a good fit. TABLE 3 SELECTED GOODNESS OF FIT STATISTICS FOR CTOG MODEL

Fit Index

χ 2 /df GFI

Criteria/threshold value

0.90 >0.90 >0.90 0.90

0.97

0.97