Abstract
Abstract
As the Internet continues to grow and flourish, more and more consumers like to buy things on the Internet. Online auction has become the biggest marketplace, because of the convenience and speed of online transactions. Ministry of Economic Affairs of Taiwan reported C2 C transaction rise by 20% annually, and its market is expected to grow continuously, meaning C2 C marketplace is receiving much increased attention. Nonetheless, there are many risks that come with online shopping. The lack of ability to inspect merchandise before purchase will put consumers a trick of fraud (identify theft). Therefore, effectively increasing consumers’ trust in online sellers will make consumers more willing to buy goods from them. This study reported some characters that might influence consumers’ trust in online sellers, and investigated the influence of trust on purchase intention. The focus of this study is to examine the consumers’ perceived effect of sellers’ transaction assurance seal, reputation and ability to protect trade information. With Yahoo! Auction as the research subject, 260 questionnaire surveys were conducted online, and 226 were considered effective. The research hypotheses were developed by using structural equation modeling (SEM), and the results supported the model.
Introduction
The convenience of Internet has enabled E-commerce to online auctions allow buyers to bid online on the items they want, and find a variety of products in a worldwide marketplace from their home. The impressive growth of online auctions in recent years also forced online retailing corporations to change game and business strategies. Ministry of Economic Affairs of Taiwan reported in “The New Age of the Internet E-Commerce Development Plan” that the online shopping market place in year 2010 rise to US$154 million, and a 22.35% increase from last year, with C2 C market (Yahoo! Auctions) at US$67.4 million (43.78% ), and B2 C market (e-malls, such as PCHome) at US$8.6 billion (56.22% ). The number of users buying items on onlineauctions shows increase as well. In 2010, buyers spent an average of US$235 on online auctions, which is lower than in 2009 (US$275), due to the downturn of economy, while the overall market showed clear signs of growth due to the increasing number of online auction users. Customers are attracted to online shopping not only because of high levels of convenience, but also because of broader selections of products, competitive pricing, and greater access to information, all of which save their time and money. Unlike traditional methods of buying things at physical stores, buyers can find an immense amount of products along with the flexibility of naming their own price online, and get more merchandise information and services of sellers [35].
The Internet is also a focal point for social networking. People share interests, activities, backgrounds, or real-life connections among friends, and even influence others’ choice through the word of mouth [78]. The Internet allows consumers to search for the information from an immense list of product items and compare prices [48]. Not just online shoppers so, but those who like to go to a real store will shop online before making purchase at physical stores. Consumers consult others’ opinions before making purchases to enhance their confidence in the product they want [9]. Trust and word of mouth play an important role in their purchase behavior [53, 70]. Previous studies also pointed out the best way for consumers to reduce perceived risks is to trust in a group of individuals with a common characteristic instead of “retailers”[83].
In order to validate the effect of sellers’ transaction assurance seal, reputation and ability to protect trade information, it is important to understand how consumers develop these perceptions. This paper proceeds as follows. Section 2 provides the conceptual development of consumer perceptions of trust, security and privacy, reputation, risks as well as purchase intentions in the online environment through literature review, and hypothesizes the relationship between the constructs. In section 3, we test the model with one empirical study, Yahoo! Auctions. Section 4 provides the test results and outlines the theoretical and managerial implications. Finally, recommendations are also made for future research and practice.
Reasoned behavior theory
Fishbein’s behavioral intention model [52] states that “a person’s performance of a specified behavior is reasonable and systematic [51]. The reasoned intention is the motivator of a specified behavior, which is determined by his or her subjective norm and attitude. Subjective norm refers to the influence of people in one’s social environment on his behavioral intentions. Attitude refers to his or her beliefs about the consequences toward the behavior. Nonetheless, it is difficult to predict a person’s behavioral intention merely by his or her attitude or subjective norm. Ajzen and Fishbein [37] revised the theory and included other variables, namely demographics, attitudes toward targets, and personality traits to form the Theory of Reasoned Action (TRA), shown in Fig. 1.
Concept of trust
The concept of trust refers to the inherent belief in others [18]. It reflects people’s subjective perception of people’s reliability. Trust is a multi-dimensional concept. Trust comes from social interaction, and helps people reduce social uncertainty and risks. The common phenomenon of trust helps us to understand and control other social environment, and predict the behavior of others [20]. Shapiro et al. [28] thought high levels of trust carry distinct benefits for business. Trust is at the heart of all kinds of social relationships [39]. The evidence explaining the relationships between trust and relationships are usually mixed.
In the context of online shopping, consumers can only see item description (containing text and, optionally, photos) instead of real objects that are accessible at the physical stores. As a result, trust becomes an important factor which influences their loyalty [34].
Technology Reasoned Action (TRA) is the theoretical foundation of the Technology Acceptance Model (TAM). TAM is a very important conceptual construct in e-commerce research. The model is the professional representation of the theory of reasoned action. In the TAM, perceived usefulness and ease of use are considered two major factors affecting attitude toward accepting a technology, which affects behavioral intention to use that technology and its actual use. Nevertheless, attitude construct was omitted in later proposed a revised TAM because it is not significant [33].
Online trust
While visiting a new web site, consumers usually ask themselves: shall I trust this vendor whom I haven’t never contacted directly and worked with? Among different forms of trust being discussed among researchers, interpersonal trust and institutional trust are the two types of concepts that are frequently applied to analyze the behavior in e-commerce.
According to McKnight et al. [21], interpersonal trust is the perception you have that other person will not do anything that harm your interest. This trust is based on experiences embedded in personal roots and relationships. In e-commerce environment, interpersonal trust will influence an individual’s willingness to trust a new website. Those who have high levels of interpersonal trust feel secured when cooperating with well-known companies. Therefore, institutional-based trust in e-commerce is regarded as trust in the laws and technologies for successful transaction [22, 31]. Institutional trust reflects an issue-oriented perspective, whereby citizens trust or distrust government or institutions because they are satisfied or dissatisfied with the current policies [59]. It can also refer an individual’s trust in government or institutions [59].
Effect of transaction assurance seal
The transaction assurance seals are provided to an online vendor once it meets a set of principles and criteria including system availability, system security, system integrity, and system maintenance ability. Take the subject of this study Yahoo! Auctions’ “Safe Seller” as an instance. Vendors applying for security seal need to go through ID verification by authenticating accounts. The ID verification consists of five stages: verifying bank account (free of charge), Credit Card Fast Authentication, Natural Citizen Digital Person Certificate (free of charge), Financial Certificate, and successful registration of “PayEasy Member”. The account verification mechanism is divided into: Yahoo! Security Credentials (free), Yahoo! Account Lock (members having a natural person certificate is free of charge), and Citizen Digital Certificate (those owning a HiNet dynamic password have to pay for the registration). Credit card authentication is free of charge for credit card members of China Trust Commercial Bank and CitiBank, while financial certificate is free for those owning a TWCA valid certificate.
Perceived effect of sellers’ reputation score
Reputation mechanisms serve as a pivotal part of market-driven evaluation system. It enables buyers to share their experience with a specific seller. Such mechanisms use “word of mouth” by accumulating and disseminating transaction records as a means of communication in online context [14]. Despite of its convenience, online reputation mechanisms are informal and self-regulating, since the available market and transaction information may be misleading and inaccurate. Then, the perceived effect of online reputation mechanisms will depend on whether consumers believe in the information provided by reputation mechanisms.
Take eBay as an instance.eBay.com, the leading e-commerce site, totaled forty-five billion dollars in trade in 2005. eBay’s rating system allows sellers and buyers to rate each other. More than half of completed transactions result in feedback provided by buyers and sellers (51.7% from the sellers, and 60.7% from the buyers). Meanwhile, reputation and price are closely related. The price difference offered by a new seller without any rating records and a good seller is 8.1% [63, 64]. Some researchers consider eBay’s rating system is an important factor to influence users’ level of trust [43].
At a macro level, institution-based trust and trust in community vendors are formed from trust transfer [45]. Take online auction providers as an instance. The third party authentication services can lend credibility to buyers as a result of transfer of trust [62]. Empirical studies also demonstrated that consumers trust the buyers they deal with because they trust in the third party who provides the security commitment [46]. Based on the above study, we hypothesize the following:
The study focused on the accumulated reputation score of the seller. If sellers have a high reputation score, it implies that they: (1) are highly experienced in the auction market, and (2) have a positive reputation, in other words, a positive “word-of-mouth”. In other words, sellers with a high reputation score have good word-of-mouth. According to Chiles and McMackin [72], reputation is a valued asset, because it requires long-term investment of resources and effort. A good reputation also signals past forbearance from opportunism [29]. Hence, a seller’s reputation has a positive effect on his or her trust in the buyer [30, 67]. If the seller’s reputation score is high, they earn trust from the buyers. Therefore, the following hypothesis is proposed:
Perceived effect of sellers’ protection of transaction data
Unquestionably, e-commerce has made shopping easier and more convenient, and provided an income for others. But e-commerce can be quite dangerous. Fraud and identity theft continue to be a challenge for e-Commerce. Buyers use fake or stolen credit cards and online payment accounts or they outright lie about delivery. Therefore, protecting buyers’ private information is the most essential task in enhancing mutual trust [32]. Disclosing privacy means that individuals arewilling and forthcoming towards the sharing of personal and intimate information with others. Nonetheless, most buyers are worried about the disclosure ofprivacy in the Internet market. A Business Week/Harris poll of 999 consumers in 1998 revealed that privacy is the biggest obstacle preventing them from using Websites. According to the survey of 1999 IBM Consumer Privacy Survey conducted by Harris Interactive, 78% of consumers are concerned about disclosing their personal information on a Web, while 54% of the consumers are unwilling to purchase on a Web, because they are afraid their private information will be stolen. Branscum [16] also mentioned two-thirds of consumers are worried about disclosing personal information online. Privacy is the “control over personal data”, and the right endowed to others for use [85].
When buyers know their transaction information is kept confidential, the perceived transaction risks and concern will be greatly reduced [3]. When the privacy mechanism is reliable, buyers will trust the sellers. Consequently, increased online trust will influence buyers’ behavioral intention positively. We, therefore, propose the following hypothesis:
Perceived risk
In the research realm of online shopping, buyers’ perceived privacy and security risk strongly affect their online purchase behaviors [49]. Privacy and security risk are the two major types of risks for interviewees in online transaction contexts [4, 84]. Privacy risk associated with e-commerce includes the leakage of personal information, such as the mailbox, age, gender, income, and etc. without his or her consent, and online retailers sending advertisements, or unsolicited marketing for product information. Security risk usually concerns the disclosure of vendor’s financial data and credit card information. Their study shows that privacy and security exerts considerable influence in Internet shopping. The perceived risk in online transaction is immense, is higher than that at brick-and-mortar stores[1, 69].
Salisbury et al. [79] used the Technology Acceptance Model (TAM) to measure determinants of network behaviors. They found that the perceived security on the Web affects purchase intentions online. They define the perceived security on the Web as “the extent to which one believes that the World Wide Web is secure for transmitting sensitive information.” A higher degree of the perceived risk reduces buyers’ purchase intention [55]. Perceived risk is considered the biggest obstacle of e-commerce transactions. Therefore, if online service providers understand consumers’ uncertainties, they shall provide suitable solutions to help consumers alleviate the level of risks [70]. A lower level of perceived risk can enhance the purchase intentions and actual sales [75]. We suggest the followinghypothesis:
Purchase intention
In the original theory of reasoned action (TRA), intentions affect an individual’s behavior. They are indications of how an individual plans to perform a given behavior [37]. Purchase intention refers to an individual’s willingness or inclination to purchase a product or services in the future [40]. It is “in a certain time period, the mental state of planning to consume a certain amount of products of particular brand” [26, 50]. Consumers’ favorable attitude towards a product or brand fosters their intention to purchase it. In a word, purchase intention is the subjective norm of an individual to purchase a specific product or brand.
Purchase intention is the result of a cognitive evaluation of outcomes associated with performing the behavior. Price can be an indicator of a level of quality [47]. Generally speaking; the perception of quality has a positive impact on price. When the perceived quality is higher than price, consumers’ perception quality will be greater, and their purchase intention will be higher. Dodds et al. [76] referred purchase intention as “the possibility that consumers will purchase a certain product”, and comes from perceived value. The perceived value will make consumers buy things, and their purchase intention is driven by acquired benefits and value [10, 76].
A consumer’s purchase behavior is formed based on his or her perceived need and the information available to them [12, 66]. Gefen et al. [19] defined purchase intention as “buyers’ intention to participate in the formation of online transaction relationship with the seller”. We propose trust and perceived risks affect consumers’ beliefs in behavioral intention asfollows:
Research model
This study aims to explore whether buyers online transaction will be influenced by transaction assurance seals, and the perceived effect of reputation score, and protection of transaction data. Then, we further investigate if these constructs affect buyers’ trust and purchase transaction. Miyazaki and Krishnamruthy [5] found that buyers feel secure when trading with stores when the seal was present, and they are willing to provide personal information. Stores owning a security seal operate at a higher standard than those without, since they need to compensate buyers for breach of privacy and safety policies. In turn, buyers’ purchase intention is increased due to additional warranty services from these stores. In this study, we take “Safe Seller” Seal on Yahoo! Auction, as the target. “Safe Seller Seal” mechanism provides two stages authentication: bank account number and auction account verification. We investigate the effect for the presence of “Safe Seller” seal, reputation score and perceived effect of protection of transaction data” on buyers’ trust, and the effect of trust on the reduction of the perceived risk, and increased purchase intention. The research model is shown in Fig. 2.
To further analyze the data, this study adopted the structural equation modeling (SEM). The measurement method was built upon measurement model analysis and structural model analysis. Analysis was undertaken with statistical software packs of SPSS 18 and Amos 20. To ensure the reliability as well as the validity of the questionnaire, Composite Reliability (CR) and Cronbach Alpha were applied to test the hypothesis model. The examination of construct validity consists of convergent validity and discriminant validity. In this study, the confirmatory factor analysis (CFA) and the average amount of variation extracted (AVE) were employed to test the convergent, and the average amount of variation extracted (AVE). To assess discriminant validity, the square root of AVE of each construct was compared with the correlation between that construct and other constructs.
Online survey length started from May 3rd, 2013 to May 24th, 2013. A total of 260 questionnaires were collected, while 31 questionnaires were respondents without making purchase on Yahoo! Auctions. A total of 229 questionnaires were considered valid. By gender, the respondents were 34% female, and 66% male. The majority of respondents (61% ) indicated that their age fell between 20 and 24 years, 24% fell between 29 and 29 years, and while 10% below 19 years. In summary, 95% of the respondents were less than 30 years of age. It means young people had a propensity of shopping online. By education, 58% of the respondents were college graduates, while 35% were post graduates. In a word, 93% of respondents were highly educated. By occupation, 74% of the respondents were students, while 13% reported work experience, with 7% working in IT industry and 6% in the service sector. By income, most respondents were on a low income. 53% of the respondents had incomes below US$350, 28% between US$351 and US$699, and 10% between US$700 to US$1,049. By Internet experience, up to 91% of the respondents had at least five years of experience, while only about 3% less than three years. It shows the subjects this study had a considerable degree of knowledge in the use of the Internet.
All shoppers surveyed had to meet the criteria of making at least two online purchases in a typical 6-month period. Among those surveys, 65% indicated they made 1– 4 online purchases in six months, and 24% made 5 to 8 online purchases. In sum, 89% of the respondents made less than 8 online purchases. With regards to purchased items, 57% of the respondents purchased clothes and shoes, and 31% shopped for books and magazines, while 32% for computer peripherals. The results reveal that there are factors attracting consumers to purchase online. By purchase amount in the recent six months, about 40% of the respondents spent US$33 to US$100, followed by 25% for US$33, because the majority of the respondents were students. Nevertheless, there are 10% of respondents who spent more than US$233. It represents high-ability consumers also made online purchases. By mode of payment, about 45% of the respondents picked up purchased items and paid at the convenience store, and about 15% of the respondents paid in cash upon delivery. In a word, 60% of the respondents preferred to settle the payment on delivery, which means that most of the consumers still have concerns over online payment. By collection of goods, 56% of the respondents picked up items and pay at the convenience store, 16% paid on delivery, and 5% via face-to-face payment. About 77% of the respondents hoped to pay on the day they received ordered items. Finally, 96% of the respondents saw the “Safe Seller” on Yahoo! Auctions; about 90% had a “slight degree of understanding” about the e-shop, while 91% of the respondents purchased items from “Safe Seller”. In this study, composite reliability was presented to test the consistency of content construct indicators. High CR (composite reliability) indicates that potential variables are highly related; i.e., in internally consistent. All CR values are greater than 0.7 and the latent variables (dimensions) appear to have good composite reliabilities [41]. In this study, composite reliability for all constructs exceeded 0.8, indicating a good internal consistency. There are two subtypes in the assessment of construct validity: convergent validity the validity and discriminant validity. Convergent validity indicates the factor loadings of measured variables on potential variables, and each factor loading is statistically manifest. All the factor loadings were significant, since they exceeded 0.5, and the AVE values exceeded the threshold of 0.5 as well. Thus, to ensure the variance explained above 0.5, some of the manifest variables need to be eliminated. Table 1 presents the inter-construct correlations and square root of AVE.
In addition, the square root of the AVE was tested against the inter-correlations of the construct with the other constructs in the model to ensure discriminant validity [17]. If the correlation of one construct ismuch higher than the square root of the AVE, it means the construct is able to account for measured dimensions of another construct. As shown in Table 2, all the square root of the AVE exceeded the correlationswith other variables. For example, the square root ofthe AVE value for the “transaction security seals” is 0.861, while the AVEs for other constructs are less than 0.861 (maximum is 0.835), and so are other constructs. The result supports the discriminant validity of the model, shown in Table 2.
The CFA results verified all the observed variables are related to latent variables, representing the uni-dimensional model has good convergent validity and discriminant validity. Next we implemented goodness-of-fit statistics and tested for the correlation between these variables. Previous studies suggested that the expected value in every category should be less than 5. The recommended threshold value is: GFI is greater than 0.8, the adjusted goodness of fit index of the model is greater than 0.8, NFI greater than 0.9, NNFI greater than 0.9, the CFI is greater than 0.9, RMSEA less than 0.08 [41, 61]. In this study, the chi-square value divided by the value of the degree of freedom is 1.944, GFI = 0.863, AGFI = 0.824, NFI = 0.916, NNFI = 0.95, CFI = 0.957, RMSEA = 0.065, which meet the recommended values by previous researchers. This result indicates a good fit to the model.
The measurement model has met the criteria of reliability, validity and goodness of fit model. Next we started to verify the hypothesis. The verification of our analysis of buyers’ trust and purchase intention which contain the construct of sellers’ trading site is divided into three subtypes: “sellers’ transaction assurance seal”, “sellers’ reputation score”, and “sellers’ protection of transaction data”. The test results are as follows.
An examination of the t value suggested that the path was significant (path coefficients = 0.64, t = 11.829, P < 0.001). The result shows a significant association between perceived effect of buyers’ transaction security seals and trust in buyers.
The data revealed that perceived effect of buyers’ reputation score is strongly related to their trust in buyers (path coefficients = 0.27, t = 8.212, P < 0.001).
The correlation parameter estimates show that perceived effect of buyers’ protection of transaction data is positively related to their trust in sellers (path coefficients = 0.12, t = 3.757, P < 0.001).
The data shows buyers’ trust and their perceived risk is negatively correlated (path coefficients=– 0.23, t=– 2.464, P < 0.01).
The test result shows that buyers’ trust and their purchase intention are positively related (path coefficients = 0.78, t = 9.353, P < 0.001).
The survey data shows that buyers’ perceived risk and purchase intention are negatively related (path coefficients=– 0.01, t=– 0.085, P < 0.001), representing sellers’ perceived risk is not significant in determining buyers’ purchase intention. Table 3 presents the results of hypothesis verification.
The cluster analysis, based on the considered variables, has split “sellers’ transaction assurance seal” into two clusters: low-level v.s. high-level perception of transaction assurance seal. The median value calculated among all the samples is 32 points. Clusters with the median value higher than 32 points represent sellers’ high-level perception of the transaction assurance seal, and vice versa for those below 32 points. The two clusters of samples are 117 and 109, shown in Figs 3 and 4 respectively.
Above diagram shows that only three paths are significant for high-level perception of sellers’ transaction assurance seal: the impact of perceived effect of sellers’ transaction assurance seal on trust, the impact of perceived effect of sellers’ protection of transaction data on trust, and the impact of trust on purchase intention. On the other hand, in the cluster of low-level perception of sellers’ transaction assurance seal”, there are four paths presenting significance: the impact of perceived effect of buyers’ transaction assurance seal on trust, perceived effect of reputation score on trust, the impact of trust on purchase intention, and perceived risk on purchase intention.
These results show that buyers who have high-level perception of sellers’ transaction assurance seal consider it important for them to trust in buyers. In other words, buyers care about the security of transaction data and worry about the privacy problem over the Internet. Nonetheless, their trust does not influence their perceived risk. It implies that buyers still have concerns over the potential risks, no matter how trustworthy the buyer is. In addition, their perceived risk does not influence their purchase intention, but their trust in sellers fosters their purchase intention.
However, for those who have low-level perception of buyers’ transaction assurance seal, their perceived effect of sellers’ reputation score influences their trust in sellers. They feel comfortable and reliable when transacting with experienced sellers; i.e. web sites which carry security seals. Their trust will not be affected by their perceived risk. However, their trust and perceived risk affect their purchase intention, implying risks are a concern.
Conclusions and implications
Managerial implications
The perceived effect of sellers’ transaction assurance seal, seller reputation score, and protection of transaction data significantly influence buyers’ trust
The perceived effect of sellers’ transaction assurance seal strengthens buyers’ trust. Based on the clustering comparison shown in Fig. 3, the perceived effect of sellers’ transaction assurance seal signals a positive effect on buyers’ trust, regardless of the extent of the effect. The result is especially important to “Safe Seller” on Yahoo! Auctions, since the “Safe Seller” increases their purchase intention. It is recommended third-party’s verification is present in order for sellers to increase the number of customers. In addition, sellers’ reputation score positively influences trust, implying sellers with high reputation is experienced and have successful transactions records and can attract buyers to buy their products. We suggest sellers accumulate their reputation scores to boost their business. The perceived effect of sellers’ protection of transaction data has a positive effect on buyers’ trust, indicating sellers’ ability of protecting buyers’ personal information strongly affects buyers’ trust in sellers. Hence, it becomes an important task for sellers to convince buyers that transactions can be done safely, and that their security system can avoid the disclosure of personal account and privacy.
Trust in the seller’s will be increased and perceived risk will be reduced
Risks arise from information asymmetry and consumers’ unfamiliarity with the sellers. To alleviate consumers’ uncertainty and perceived risk, sellers should take other approaches to build up consumers’ trust and reduce their perception of risks. Empirical studies also supported the concept of trust. Trust makes online users less concerned and reduces their perceived risk [25, 82]. Once users’ trust is strengthened, they perceived risk will be reduced accordingly, and their purchase intention will be increased as well. Pavlou and Gefen [57] also indicated buyers’ perceived risk will be significantly influenced by buyers’ trust in online stores and their willingness to buy. As such, it is necessary that sellers should strive to obtain buyers’ trust. After a certain level of trust is reached, risk will become less important, and buyers’ purchase intention will be increased significantly.
Trust in sellers significantly influences buyers’ purchase intention, but buyers’ perceived risk will not significantly affect their purchase intention
Past research indicated trust and perceived risk significantly affect purchase intention. Bhattacherjee [2] and Gefen [18] found that there is a direct relationship between trust and the purchase intention. Past test result is inconsistent with the test result here. From clustering comparison shown in Figs. 3 & 4, buyers who have low-level perception of sellers’ transaction assurance seal present a high level of perceived risk. Such perception influences their purchase intention. The reason is mainly because consumers are unfamiliar with Yahoo! Auctions. Their trust in Yahoo! Auctions makes them believe that Safe Seller strictly supervises every seller in online transactions, and they will be more willing to trust the seller as a result of trust transfer.
Limitations and suggestions for future research
As with other survey research, interpretation of our results is subject to certain limitations. The major purpose of this study is to examine whether internet seals foster sellers’ trust in the transaction. Among theauction platforms in Taiwan, only Yahoo! Auctions has such security mechanism. In addition, the research subject only focused on users in Taiwan. Due to the failure to obtain the Auctions platform’s agreement for further investigation, questionnaires and distribution of paper questionnaires can only be collected online. A majority of respondents ranged from 21 to 30 years of age. Most of them have a relative low income, a condition that may potentially limit the applicability of our research findings in other settings or populations. Hence, the study fails to represent the complete auction markets.
Lack of time and fund unavailability constrains the research scope, so only users in Taiwan were investigated in this study. In addition, the operating style for online business varies by country and by culture and the test result will be different as well. Therefore, we suggest that culture can be taken into consideration if time and cost permits, so that a diversified culture research can be presented. In online auction markets, buyers’ need varies as well. We think product involvement can also be an indicator. If buyers are highly involved in the product items they want to purchase, it can be assumed their purchase intention and trust propensity will be strong. Furthermore, other potential factors may also affect buyers’ purchase intention, such as credit card [27] and Escrow Service [57]. Hence, it is suggested that other service mechanisms provided by other online auction stores can be taken as independent variables to test the buyers’ purchase intention and trust propensity for future research. More research is needed to test these assumptions.