Global e-Business Association
[ Article ]
The e-Business Studies - Vol. 17, No. 5, pp.53-68
ISSN: 1229-9936 (Print) 2466-1716 (Online)
Print publication date Oct 2016
Final publication date 30 Oct 2016
Received 05 Oct 2016 Revised 20 Oct 2016 Accepted 27 Oct 2016
DOI: https://doi.org/10.20462/tebs.2016.10.17.5.53

The Study on Differential Effects of Information in Online Auction Study

Min Jung Ko*
*Professor, Computer Science Institute, Dongguk University mjgo@dongguk.edu
온라인 경매에서 차별적 정보의 효과연구
고민정*
*동국대학교 전산원 교수 mjgo@dongguk.edu

Abstract

As the online auction market is growing rapidly, many studies have investigated the success factors for online auctions. The previous literature canbe grouped into the following areas: the optimization of auction bid price based on the economic principle and the utility maximization strategies of bidders and sellers. This study takesa holistic view by defining the online auction process as an integrated persuasion process. We assumed that the auction process is a persuasion process between bidders and sellers and used the dual process of Elaboration Likelihood Model (ELM). We investig ated how central or peripheral cuesaffect the persuasion process. We collected empirical data from eBay, an online auction site, using a crawling engine that we developed. To conduct this study, we selected the three most typical products traded in online auctions and analyzed the effects of cues or information on the auction success. The methodology and the theoretical and practical implications are further discussed.

초록

온라인 경매가 지속적으로 성장하고 있으며, 이와 관련된 연구가 활발히 이루어지고 있다. 이전에 경매와 관련해서는 경매의 낙차가격을 최적화하거나 경매자와 입찰자에게 최대 이윤을 제공하는 전략 제공이 대부분이었다. 그러나 이들은 경매 시작가에 영향을 미치는 요소들에 대한 연구, 즉 경매자수나 입찰자나 경매자들의 행동을 연구하는데 연구의 초점이 있었다. 이러한 연구들은 경매 프로세스의 일부분을 기반으로 하여 온라인 경매 전체적인 프레임에 대한 이해가 어렵고, 연구 품목이 제한적이어서 실제 경매 상황에 적용하기에 어려운 점이 많았다. 이를 해결하기 위하여 본 논문은 경매과정을 경매자와 입찰자 사이의 설득의 과정으로 정의하고, ELM을 기초로 하여 설득에 필요한 정보를 중심경로와 주변경로로 분류하고 정의한다. 여기에 eBay에서 실제 자료를 크롤링하여 경매성공에 영향을 미치는 요소들에 대한 실험을 하였다. 그 결과, 경매 물품에 따라서 경매 성공에 영향을 미치는 요소들이 차이가 있음을 알 수 있었다. 본 연구의 결과는 실제 웹사이트를 구성하는 설계자와 경매에 성공을 위하여 전략 시나리오 작성에 새로운 통찰을 제공하였다.

Keywords:

Auction Success, Internet Auction, ELM(Elaboration Likelihood Model), eBay, Winning Bid Price, Persuasion Process

키워드:

경매성과, 인터넷 경매, ELM, eBay, 낙찰가격, 설득 프로세스

Contents


Ⅰ. Introduction

Recently, online auction markets like eBay or Ali Express have grown rapidly (Agarwal et al. 2002; Bapna et al. 2003; Buschke et al. 1997). The rapid growth of the online auction market comes with changes to consumer purchasing patterns (Ba et al. 2002). Consumers tend to buy goods through online auctions because they can get much higher satisfaction through a successful online auction than through general Internet buying (Bapna et al. 2003; Stern et al. 2008). The importance and growth of online auction markets attracts many researchers (Ba et al. 2002); Buschke et al. 1997; Du et al. 2012). Previous literature can be grouped into two areas: the optimization of auction bid price based on the economic principle (Ba et al. 2003; Liu et al. 2000; Nelson and Phillip 1974) and the utility maximization strategies of bidders and sellers (Ba et al. 2003; Bapna et al. 2003, Halstead et al. 2003). These studies were not based on actual data and focused mostly on individual cases and specific factors including the starting price of a transaction, the bidder number, bidder’s and seller’s behavior in the auction market, and the online auction strategy (Chiou et al. 2009; Haruvy et al. 2013; Srinivasan et al. 2010). Most of these studies used conceptul approaches (Iivari et al. 1987).

Although previous studies helped to understand the mechanics and factors for the success of online auction markets, they examined limited parts and factors, making the results difficult to generalize (Chin et al. 1998; Strong et al. 1997). The data collected by researchers was also sporadic and the data collection process cannot be easily validated (Chiou et al. 2009; Du et al. 2012). Thus, the collected data have limitations in terms of reliability and transparency (Palmer et al. 2002; Paulo et al. 1999). In addition, the previous studies often use highly abstracted concepts, which are difficult to apply to real world situations.

In order to fill these gaps, we apply an Elaboration Likelihood Model (ELM) as the overarching theory. This offers a holistic view and assumes that the auction process is a persuasion process between bidders and sellers using central or peripheral information (Li et al. 2009; Strong et al. 1997). We collected the empirical data of three products typically traded in online auctions directly from the eBay website using a crawling engine that I developed. We then categorized the collected data into central or peripheral cues to test how these cues affect the outcome of the persuasion process in a differential manner.

We found that the central and peripheral cues affected online auction success in a differential manner according to the characteristics of each auction item, such as the level of personalization, formality, and complexity. This study contributes to the existing literature in two ways. Firstly, this study provides a holistic view of the online auction success by using the concept of persuasion with central and peripheral cues. Secondly, this study offers a way to collect online auction data that can be used to examine online auction success with actual transaction data. Practically, this study has implications for the design of online auction websites and the processes used to be successful.


Ⅱ. Literature Review

1. ELM (Elaboration Likelihood Model)

We defined an auction as the overall persuasion process. There are various cues in the persuasion process and the ELM assumes that messages and cues are processed through two paths (Chaiken et al. 1980; Petty et al. 1986). In this case, processing was determined based on the message quality and involvement. When a message has high complexity, the message is processed using the central path (Chaiken et al.1980; Petty et al. 1986).

This means that when buying a product with high complexity product, consumers need elaboration and tend to buy based on its performance and the detailed description of the product (Li et al. 2009). Otherwise, when buying a product with low complexity, consumers tend to buy based on impulse information and images (Li et al. 2009). Therefore, the ELM is useful to plan advertisements and sales strategies properly. However, each consumer has different criteria when deciding to purchase goods.

Looking at the processing path of the information in Figure 1, so that the structural (Systematic) processing occurs (Petty et al. 1986). In this case, enough rational, if provided by the sender specific information, consumers become immediately visible change of attitude. On the other hand, if it will occur heuristic processing during processing of the message, if the processing processes or rational later information, depending on the involvement of the relatively sensitive perception, change appearance in consumer attitudes (Li et al. 2009). In the former case, as in the above structural process, the results lead to a change in the instantaneous attitude which can be obtained, in the latter case, it becomes possible to go through the step of re-adjusting their own judgment. Potential model for refinement conclusion can be a theory proposed by whether taking a method of providing what information in any environment by dividing the process information in more detail. Moreover, it is a model which can be examined until should be presented to any situation what messages when the advertisement convince the mechanism.

[Figure 1]

Persuasion Process of ELM (Petty et al. 1986)

Our study classifies a variety of information provided by eBay into central and peripheral cues. We also hypothesize that these cues affect the success of an auction depending on the characteristics of the auction items. We define the direct information as the central cues, such as the number of images, the number of bidders, the starting price, the shipping fee, the delivery days, and the number of repeated upload. We defined the indirect information as the peripheral cues, such as the seller’s reputation and environmental factors.

2. Prior studies on online auction

Based on existing economic marketing, Gilkeson and Reynolds analyzed the success factors of Internet auctions according to the starting price, auction type, and number of unexpected bids. They found that these success factors can be used to determine the winning bid price on eBay (Gilkeson et al. 2003). Du and Yu found that the characteristics and performance of the seller affected auction success based on satisfaction, success, and effectiveness (Gilkeson et al. 2003). Du and Yu also used empirical research methods and price (cost) optimization instead of conventional economics-based quantitative analysis (Gilkeson et al. 2003). Zeithammer studied a sales strategy according to the seller’s commitment and the early prices (Zeithammer et al. 2007) and developed the optimal sales strategy of monopolistic competition markets in online auctions. Matsumoto and Fujita also proposed an agent that can maximize profits for a particular combination of products (Matsumoto et al. 2002). These studies used modeling methods and the price (cost) optimization approach.

By evaluating the asymmetry between the experience of a bidder and learned information, Srinivasan and Wang found a degradation phenomenon of reactivity that is generated from the status quo bias of an inexperienced bidder (Srinivasan et al. 2010). They also showed how a bidder’s experience can affect strategy evolution. Paulo and Oliverira studied the bidder’s strategy not in a typical auction, but in a sequential auction (Paulo et al. 1999). They also researched the importance of the design elements that affect auction strategies, such as bidder demand, participation experience, and auction design parameter. Akula and Mensace analyzed how the reputation of a bidder and seller affected each other (Akula et al. 2004). Bapna and Chang studied the results of successful auctions that used the overlapping auction strategy (Bapna et al. 2009). Park and Bradlow presented the key behavioral aspects that could predict a bidder’s behavior by modeling the empirical data and by dividing the bidding process into four stages (Park et al. 2005). Haruvy and Jap conducted a study of the advantages and disadvantages of bidding through the bidder quality process to predict the bidder behavior in anonymous situations (Haruvy et al. 2013). Bockstedt and Huat studied the differential strategy of successful sellers by evaluating their population and reputation (Bockstedt et al. 2011). These approaches were all based on empirical studies and peripheral cues for online auction success.

Halstead and Becherer studied the behavior of bidders according to auction market size (Halstead et al. 2003) and insisted that leadership and differential strategy are important. Yen and Lu analyzed the factors that lower consumer repurchase history through the Expected Rejection Theory (EDT) (Yen et al. 2008) and found that worry about the products lowered consumer repurchases because consumers cannot actually see the real products in an online auction. Matsumoto and Fujita studied the specific combination of products to maximize profits in a variety of online auctions (Matsumoto et al. 2002). Chiou and Wu proposed that a seller should take care when switching auction sites (Chiou et al. 2009) and researched the reaction of buyers when sellers change auction sites. Srinivasan and Wang studied ways to overcome uncertainty of the auction by using quality signals according to a buyer’s reaction (Srinivasan et al. 2010). Finally, Stern and Rone insisted that engineers should design auction sites based on closeness, impulse, and patience towards computer (Stern et al. 2008). These approaches were all based on modeling studies and peripheral cues for online auction success.

Table 1 classifies the previous studies into four types according to the approach and method.

Review of Prior Auction Success Studies


Ⅲ. Theoretical Background and Research Hypotheses

1. Auction Outcome

We studied the information available on the Internet auction site to determine whether independent variables impact the winning bid price and auction success. Auction success means whether the auction is a success or failure in the auction transaction (Gilkeson et al. 2003; Matsumoto et al. 2002; Zeithammer et al. 2007). The winning bid price is the final price after competition among bidders (Hulland et al. 1999). The winning bid price and auction success have been previously used to measure the success of auction performance (Bockstedt et al. 2011; Gilkeson et al. 2003). In our study, the variable of auction outcome was proposed with two dependent indicators as alternative measures.

Here, we present the basic premise that auction success is determined by two processes of information processing. With everything else being equal, central cues have a stronger impact than peripheral cues. It is assumed that central cues are more effective for products with high complexity and peripheral cues are more effective for products with low complexity.

Basic Assumption:

Auction success is determined by a dual information process depending on a buyer’s need for elaboration of an auction item. With everything else being equal, the effect of central cues on the auction success and the winning bid price will be stronger than the effect of peripheral cues.

2. Central Cues

When buying a product with high complexity, consumers tend to buy with based on performance and the detailed description of the product (Petty et al. 1986). We define the direct information as central cues, such as the number of images, the number of bidders, the starting price, the shipping fee, the delivery days, and the number of repeated unloads.

The number of images

The number of product images is the most basic association for auction goods (Li et al, 2009; Strong et al. 1997). The number of images reduces the uncertainty of the auction goods and consumers have more confidence during the auction. H1a: The number of images is positively associated with the auction success. H1b: The number of images is positively associated with the winning bid price.

The number of bidders

The number of bidders enhances the value of an auction item (Rafaeli et al. 2002). When there are more bidders, a higher bid price is set (Gilkeson et al. 2003). Therefore, the number of bidders will positively impact the auction success and winning bid price. H2a: The number of bidders is positively associated with the auction success. H2b: The number of bidders is positively associated with the winning bid price.

The shipping fee

If there is an increase in the shipping fee, the winning bid price will also be increased (Bockstedt et al. 2011). Therefore, higher shipping fees have lower auction success and increase the winning bid price. H3a: The shipping fee is positively associated with the auction success. H3b: The shipping fee is positively associated with the winning bid price.

The delivery days

If the delivery days increase, purchasers of auction goods will have to wait longer for their items (Bockstedt et al. 2011; Gilkeson et al. 2003). Therefore, delivery days are highly related to the auction success. H4a: The delivery days are positively associated with the auction success. H4b: The delivery days are positively associated with the winning bid price.

The auction duration

The auction duration involves the start and end time (Gilkeson et al. 2003; Liu et al. 2000) and, unlike a traditional auction, Internet auctions permit any potential bidder to electronically see the items at any time before the auction ends (Akula et al. 2004). In general, if the auction duration increases, the number of bidders increases. H5a: The auction duration is positively associated with the auction success. H5b: The auction duration is positively associated with the winning bid price.

The starting price

In any auction, the seller must prescribe the minimum price to the first bidder. Setting a lower starting price can be attractive to more bidders (Gilkeson et al. 2003). Conversely, if the starting price is very high, bidders might be discouraged to make a bid, which can be explained by the rational auction's model (Torkzadeh et al. 2002). H6a: The starting price is positively associated with the auction success. H6b: The starting price is positively associated with the winning bid price.

The numbers of repeated uploads

If the goods for auction are not sold within 7 days, the goods will disappear (Gilkeson et al. 2003). Therefore, sellers have to upload the items again. H7a: The number of repeated uploads is positively associated with the auction success. H7b: The number of repeated uploads is positively associated with the winning bid price.

3. Peripheral Cues

Under low elaboration conditions, many variables may act as peripheral cues (Chaiken et al., 1980; Petty et al. 1986). When buying a product with low complexity, consumers tend to buy based on impulse information and images (Chaiken et al., 1976). We defined indirect information as peripheral cues, such as the rating of the seller, the seller’s positive feedback ratio, the number of the seller’s other items on eBay, and the bidder’s rating. The seller’s rating, the seller’s positive feedback ratio, and the number of the seller’s other items on eBay. The seller’s credibility has a significant impact on the reliability of the product (Bockstedt et al. 2011; Srinivasan et al. 2010). In addition, since sellers cannot meet buyers, a seller’s credibility is an important criterion to a bidder. The seller's rating, shown as a star count, is a buyer’s feedback of the seller. The seller's positive feedback ratio is the ratio of positive feedback scores (Akula et al. 2004; Chiou et al. 2009). In addition, the number of the seller’s other items is important to evaluate the trust of auction goods.

H8a: The rating of a seller is positively associated with the auction success. H8b: The rating of a seller is positively associated with the winning bid price. H9a: The seller’s positive feedback ratio is positively associated with the auction success. H9b: The seller’s positive feedback ratio is positively associated with the winning bid price. H10a: The number of seller’s other item at eBay is positively associated with the auction success. H10b: The number of seller’s other item at eBay is positively associated with the winning bid price.

The bidder’s rating

The performance of a bidder increases with experience (Keeney et al. 1999; Rafaeli et al. 2002) and by participating in auctions repeatedly (Yen et al. 2008). A well trained seller is evaluated by many bidders, and the score assigned by the bidders is referred to as the feedback score. Therefore, a bidder’s rating positively impacts the success and winning bid price of the auction. H11a: The bidder’s rating is positively associated with the auction success. H11b: The bidder’s rating is positively associated with the winning bid price.


Ⅳ. Research Methodology

1. Data Collection Procedure

Figure 2 shows the crawling engine used to collect the actual data. This was designed to solve the transparency and temporal problems that can occur when collecting data manually (Nielsen et al. 1999; O'Brien et al. 1998). For the actual condition of the eBay auction goods, we used the collecting condition “used format,” which means auction. Here, the used format is classified into “Sold” and “Completed” auctions where sold means success and completed means failure. For our empirical study, we selected a Golf Driver, which have high complexity, Books, where condition is important, and iPods, which have standardized specifications.

The collection procedure used the following steps:

Step 1: Search “Golf Driver,” “Books,” and “iPod” using eBay’s search utility under “Sold” and “Completed.”

Step 2: Examine the titles of matching auctions, saving “Sold” and “Completed” records.

Step 3: Save the extracted records into a database, parsing each web page.

Step 4: Update the auction database after data cleaning.

Step 5: Extract the results from analyzing each item.

[Figure 2]

Auction Data Collection Engine

2. Operationalization of the Research Variables

In our study, we integrated the auction process as the persuasion process between bidders and sellers and used the dual process of the ELM. We investigated which cues have a strong effect on the persuasion process. By conducting research that was focused on empirical cues provided from eBay, we reduced the gaps between real online auction factors and our research cues. Table 2 shows the operationalization of research variables.

Operationalization of Research Variables

3. Data Description

We collected empirical data using online auction data from eBay. The sample was collected from July 17 to July 22, 2014. We retrieved pages that included general products like Golf Drivers, Books, and iPods (see Table 3). Golf Drivers accounted for the largest number of sold items. Books were based on a consumer’s experience (Palmer et al., 2002; Paulo et al., 1999), and it was relatively difficult to determine price as this is based on the condition of the books. Finally, iPods were highly personal as they are based on subjective tastes. However, the iPod was widely regarded as being popular and of standard value. Table 4 shows that approximately 86.6% of the products purchased were Golf Drivers, 6.1% were Books, and 7.4% were iPods.

Sample Characteristics and descriptive statistics

Number of Auctions

Table 4 shows the number of auctions and depicts the success and failure type of the 3,697 auctions, which are analyzed in the following section. The number of successful auctions was 3,105, or 84%, and the number of failures was 592, or 16%.

Table 5 shows the mean and range values for the collected auction data. There are differences between “all” (ALL) and “successful” (SUCC) auctions. In Table 6, ALL indicates all auction cases and SUCC indicates only successful auction cases. Many variables did not differ consistently or significantly between all and successful auctions.

Summary of Data Collected by item (all and successful auctions)

Logistic Regressions of Auction Success on Auction Features(All Auction Cases)


Ⅴ. Results and Analysis

Here, we provide the empirical results that focus on auction success and the winning bid price. Table 6 shows the results of the logistic regressions of the auction success. The results show that six factors were uniformly associated (across all types) with auction success. In most cases, the higher values of the variables yielded lower auction success. However, the number of bidders and the bidder rating were the exception. For iPods, the shipping fee, auction duration, starting price, and bidder rating affected the auction success. For Books, the model-fit value was valid but no variables affected auction success. For Golf Drivers, the number of bidders, shipping fee, starting price, number of repeated uploads, number of seller’s other items, and bidder rating affected the auction success. It is important to note that the multicollinearity for this sample data was examined using the Variance Inflation Factor (VIF) (Chin et al. 1998). All VIF values were between 1.044 and 1.354, which indicated there were no significant multicollinearity problems.

Table 7 shows the results of the multiple linear regression of the winning bid price in successful auction cases. Only five variables were significantly related to the winning bid price across all types: the number of images, number of bidders, starting price, number of repeated uploads, and seller’s rating. All were positively associated, except the number of repeated uploads. This means that the higher value of these variables yielded higher winning bid prices. In addition, the number of repeated uploads had significantly negative impacts on the winning bid price. For iPods, the delivery days and starting price impacted the winning bid price. For Books, the number of bidders and starting price impacted the winning bid price. For Golf Drivers, the number of bidders, starting price, seller’s rating, and bidder’s rating impacted the winning bid price. Interestingly, the adjusted r square for these regressions suggested relatively high explanatory power. We also checked the multicollinearity for this sample data by using the VIF (Chin et al., 1998). All VIF values were between 1.025 and 2.094, which indicated there were no significant multicollinearity problems.

Linear Regressions of Winning Price on Auction (Successful Auction Cases)

Auction success was statistically related to the number of bidders, shipping fee, starting price, number of repeated uploads, seller number of other items, and bidder’s rating. The winning bid price was associated with the number of images, number of bids, starting price, number of repeated uploads, and the seller’s rating. Therefore, hypotheses H1b, H2a, H2b, H3a, H6a, H6b, H7a, H7b, H8b, H10a, and H11a were supported at the 0.05 level of significance (see Table 8). Three variables were not associated with auction success and winning bid price: the delivery days, auction duration, and seller’s positive feedback ratio.

Hypothesis Testing Results


Ⅵ. Conclusion and Discussion

In this study, we complemented the limited previous research that focused on individual elements rather than the integrated persuasion process in online auctions. Using a persuasion model ELM, we classified the seller and bidder persuasion process based on central and peripheral cues and examined which cues were important to the persuasion process for each item. Based on the ELM model, we analyzed the effects between central and peripheral cues on successful online auctions according to different items. We chose three products and examined whether these cues affected bidders. As a result, five cues of the seven central variables, the number of images, number of bidders, shipping fee, starting price, and number of repeated uploads, affected the process of persuasion. In addition, three cues of the four peripheral variables, the seller's rating, number of seller's other items on eBay, and bidder's rating, affected the success of the auction. This means that the ELM model was related considerably to online auctions as a persuasion process. We found that not only the reliability of the seller but also the number of reliable bidders had important roles in online auctions. That is, since buyers have limited information when participating in an online auction they had some anxiety when buying goods. The number of bidders helped reduce this fear.

Additionally, we found that certain cues affect the success of an auction. The first item, a Golf Driver, had many specifications and designs. Generally, Golf Drivers have high complexity and buyers consider many conditions. As a result, the four central cues of the number of bidders, shipping fee, starting price, and number of repeated uploads affected the auction. The three peripheral cues of the seller's rating, number of seller's other items on eBay, and bidder's rating affected the auction. Compared to Books and iPods, Golf Drivers needed more cues for a successful auction transaction.

For Books, two of the central cues, the number of bidders and starting price, had an impact on auction success. The value of the books tended to be dependent on the condition of the books. Thus, the number of other bidders was an important cue. Finally, for iPods, the specifications of the products were standardized. Thus, this was not an important cue, but the factors of other services such as the shipping fee, delivery days, and auction duration were important. In addition, the bidder's rating was also an important cue.

In conclusion, we found that there were many different combinations of important cues according to the characteristics of an auction item. Although it was impossible to research all of the detailed cues for all items, products of higher complexity required more cues during the persuasion process. Products of lower complexity required environmental variables in addition to their own information for auction success.

Our study contributes to the existing literature about the success factors for online auctions in two ways. First, this study provides a holistic view of online auctions based on the persuasion process. Second, this study’s findings are based on actual auction data, which can be compared to the data from research that used the modeling approach. The results of our study can be used in the marketing and designing of websites to include characteristics of auction items, and can help sellers create an effective strategy. For future research, we will study the modeling persuasion path of Internet auctions in addition to studying how to include customized persuasive cues for each commodity. The methodology, theoretical, and practical implications are further discussed.

References

  • Agarwal, R., & Viswanath, V., (2002), “Assessing a firm's web presence: a heuristic evaluation procedure for the measurement of usability”, Information Systems Research, 13(2), p168-186. [https://doi.org/10.1287/isre.13.2.168.84]
  • Akula, V., & Daniel, A. Mé, (2004), “An analysis of bidding activity in online auctions”, E-Commerce and Web Technologies, Springer Berlin Heidelberg, p206-217.
  • Ba, S., & Paul, A. P., (2002), “Evidence of the effect of trust building technology in electronic markets: Price premiums and buyer behavior”, MIS Quarterly, p243-268. [https://doi.org/10.2307/4132332]
  • Bapna, R., Paulo, G., & Alok, G., (2003), “Analysis and design of business-to-consumer online auctions”, Management Science, 49(1), p85-101. [https://doi.org/10.1287/mnsc.49.1.85.12754]
  • Bapna, R., Paulo, G., & Alok, G., (2003), “Replicating online Yankee auctions to analyze auctioneers' and bidders' strategies”, Information Systems Research, 14(3), p244-268. [https://doi.org/10.1287/isre.14.3.244.16562]
  • Bapna, R., (2009), “Overlapping online auctions: Empirical characterization of bidder strategies and auction prices”, MIS quarterly, p763-783.
  • Bockstedt, J., & Kim, H.t G., (2011), “Seller strategies for differentiation in highly competitive online auction markets”, Journal of Management Information Systems, 28(3), p235-268. [https://doi.org/10.2753/MIS0742-1222280307]
  • Buschke, L., (1997), “The basics of building a great Web site”, Training & Development, 51(7), p46-48.
  • Chaiken, S., & Alice, H. E., (1976), “Communication modality as a determinant of message persuasiveness and message comprehensibility”, Journal of Personality and Social Psychology, 34(4), p605-614. [https://doi.org/10.1037/0022-3514.34.4.605]
  • Chin, W. W., (1998), “The partial least squares approach to structural equation modeling”, Modern methods for Business Research, 295(2), p295-336.
  • Chiou, J. S., Lei-Yu, W., & Yi-Ping, S, (2009), “Buyer satisfaction and loyalty intention in online auctions: Online auction web site versus online auction seller”, Journal of Service Management, 20(5), p521-543. [https://doi.org/10.1108/09564230910995125]
  • DeLone, W. H., & Ephraim, R. M., (1992), “Information systems success: The quest for the dependent variable”, Information systems research, 3(1), p60-95. [https://doi.org/10.1287/isre.3.1.60]
  • Dinerstein, E., (1991), Webster's New World Dictionary, New York, NY.
  • Donthu, N., & Adriana, G., (1999), “The Internet shopper”, Journal of Advertising Research, 39(3), p52-58.
  • Du, H. S., et al , (2012), “Empirical investigation of EachNet: the eBay model of C2C online auction in China”, Engineering Management, IEEE Transactions, 59(1), p160-175.
  • Fornell, C., & David, F. L., (1981), “Evaluating structural equation models with unobservable variables and measurement error”, Journal of marketing research, p39-50. [https://doi.org/10.2307/3151312]
  • Gilkeson, J. H., & Kristy, R., (2003), “Determinants of internet auction success and closing price: An exploratory study”, Psychology & Marketing, 20(6), p537-566. [https://doi.org/10.1002/mar.10086]
  • Goes, P. B., Gilbert, G. K., & Arvind, K. T., (2012), “Bidding Behavior Evolution in Sequential Auctions: Characterization and Analysis”, MIS Quarterly, 36(4), p1021-1042.
  • Halstead, D., & Richard, C. B., (2003), “Internet auction sellers: does size really matter?”, Internet Research, 13(3), p183-194. [https://doi.org/10.1108/10662240310478204]
  • Haruvy, E., & Sandy, D. J., (2013), “Differentiated Bidders and Bidding Behavior in Procurement Auctions”, Journal of Marketing Research, 50(2), p241-258. [https://doi.org/10.1509/jmr.10.0036]
  • Hoffman, T. P., Novak, DONNA, & LHTP, (1997), “A new marketing paradigm for electronic commerce”, The Information Society, 13(1), p43-54. [https://doi.org/10.1080/019722497129278]
  • Holbrook, M. B., & Elizabeth, C. H., (1982), “The experiential aspects of consumption: Consumer fantasies, feelings, and fun”, Journal of consumer research, p132-140. [https://doi.org/10.1086/208906]
  • Hulland, J., (1999), “Use of partial least squares (PLS) in strategic management research: A review of four recent studies”, Strategic Management Journal, 20(2), p195-204. [https://doi.org/10.1002/(SICI)1097-0266(199902)20:2<195::AID-SMJ13>3.0.CO;2-7]
  • Iivari, J., & Erkki, K., (1987), “The PIOCO model for information systems design”, MIS Quarterly, p401-419. [https://doi.org/10.2307/248688]
  • Li, S., Kannan, S., & Baohong, S., (2009), “Internet auction features as quality signals”, Journal of Marketing, 73(1), p75-92. [https://doi.org/10.1509/jmkg.73.1.75]
  • Keeney, R. L., (1999), “The value of Internet commerce to the customer”, Management science, 45(4), p533-542. [https://doi.org/10.1287/mnsc.45.4.533]
  • Klobas, J. E., (1995), “Beyond information quality: Fitness for purpose and electronic information resource use”, Journal of Information Science, 21(2), p95-114. [https://doi.org/10.1177/016555159502100204]
  • Liu, C., et al. , (1997), “Web sites of the Fortune 500 companies: facing customers through home pages”, Information & Management, 31(6), p335-345. [https://doi.org/10.1016/S0378-7206(97)00001-3]
  • Matsumoto, Y., Satoru, F., (2002), “An Auction for Bidding on Combinations of Items”, '01 Proceedings of the fifth international conference on Autonomous agents, p552-559.
  • Nielsen, J., (1999), Designing web usability: The practice of simplicity, New Riders Publishing.
  • Nelson, P., (1970), “Information and consumer behavior”, The Journal of Political Economy, p311-329. [https://doi.org/10.1086/259630]
  • Nelson, P., (1974), “Advertising as information”, The Journal of Political Economy, p729-754. [https://doi.org/10.1086/260231]
  • O'Brien, J. A., (1998), Management information systems: Managing information technology in the networked enterprise, McGraw-Hill Professional.
  • Palmer, J. W., (2002), “Web site usability, design, and performance metrics”, Information Systems Research, 13(2), p151-167. [https://doi.org/10.1287/isre.13.2.151.88]
  • Park, Y. H., & Eric, T. B., (2005), “An integrated model for bidding behavior in Internet auctions: Whether, who, when, and how much”, Journal of Marketing Research, 42(4), p470-482. [https://doi.org/10.1509/jmkr.2005.42.4.470]
  • Paulo, F. B., Paulo, C. M., & Maria, C. F., (1999), “Hypercharts: Extended statecharts to support hypermedia specification”, Software Engineering, IEEE Transactions, 25(1), p33-49.
  • Peterson, R. A., Sridhar, B., & Bart, J. B., (1997), “Exploring the implications of the Internet for consumer marketing”, Journal of the Academy of Marketing science, 25(4), p329-346. [https://doi.org/10.1177/0092070397254005]
  • Petty, R. E., & John, T. C., (1986), The elaboration likelihood model of persuasion, Springer New York.
  • Pinker, E. J., Abraham, S., & Yaniv, V., (2003), “Managing online auctions: Current business and research issues”, Management Science, 49(11), p1457-1484. [https://doi.org/10.1287/mnsc.49.11.1457.20584]
  • Rafaeli, S., & Avi, N., (2002), “Online auctions, messaging, communication and social facilitation: a simulation and experimental evidence”, European Journal of Information Systems, 11(3), p196-207. [https://doi.org/10.1057/palgrave.ejis.3000434]
  • Srinivasan, K., & Xin, W., (2010), “Commentary-Bidders' Experience and Learning in Online Auctions: Issues and Implications”, Marketing Science, 29(6), p988-993. [https://doi.org/10.1287/mksc.1100.0581]
  • Stern, B. B., et al , (2008), “Consumer acceptance of online auctions: An extension and revision of the TAM”, Psychology & Marketing, 25(7), p619-636. [https://doi.org/10.1002/mar.20228]
  • Subramaniam, C., Michael, J. S., & David, M. G., (2000), “Product marketing and channel management in electronic commerce”, Information Systems Frontiers, 11(4), p363-378. [https://doi.org/10.1023/A:1010061924822]
  • Torkzadeh, G., & Gurpreet, D., (2002), “Measuring factors that influence the success of Internet commerce”, Information Systems Research, 13(2), p187-204. [https://doi.org/10.1287/isre.13.2.187.87]
  • Yen, C. H., & Hsi-Peng, L., (2008), “Factors influencing online auction repurchase intention”, Internet Research, 18(1), p7-25. [https://doi.org/10.1108/10662240810849568]
  • Zeithammer, R., (2007), “Research Note-Optimal Selling in Dynamic Auctions: Adaptation Versus Commitment”, Marketing Science, 26(6), p859-867. [https://doi.org/10.1287/mksc.1070.0269]

[Figure 1]

[Figure 1]
Persuasion Process of ELM (Petty et al. 1986)

[Figure 2]

[Figure 2]
Auction Data Collection Engine

Contents

ABSTRACT
Ⅰ. Introduction
Ⅱ. Literature review
Ⅲ. Theoretical Background and Research Hypotheses
Ⅳ. Research Methodology
Ⅴ. Results and Analysis
VI. Conclusion and Discussion
References
국문초록

<Table 1>

Review of Prior Auction Success Studies

Method Empirical Study Conceptual Study
Approach
Price or Cost Optimization Approach
(Market Auction Price, Transaction
Cost, Economics, etc.)
Gilkeson et al.(2003) Matsumoto et al.(2002);
Pinker et al.(2003)
Peripheral Cues Approach
(Bidder’s & Seller’s Experience,
Reputation, Satisfaction, Strategy, etc.)
Akula et al.(2004);
Bapna et al.(2009);
Bockstedt et al.(2011);
Goes et al.(2012);
Haruvy et al.(2013);
Matsumoto et al.(2002);
Srinivasan et al.(2010)
Chiou et al.(2009);
Halstead et al.(2009);
Li et al.(1997);
Matsumoto et al.(2002);
Srinivasan et al.(2010);
Zeithammer et al.(2007)
Central and Peripheral Cues Approach
(Persuasion Process)
My Study

<Table 2>

Operationalization of Research Variables

Variable Definition & Description
Auction
outcome
Auction Success Was the auction successful? Success (1) or Failure (0)
Winning Bid Price Final(high) bid, zero if no bids
Central
Cues
Number of Images The number of images for the auction item
Number of Bidders The number of bidder for auction item
Shipping Fee Cost of shipping, handling, and insurance
Delivery Days The days to deliver to buyer
Auction Duration The number of days from start to end
Starting Price The first price acceptable to opening bid
Number of Repeated Upload The number of repeated uploads
Peripheral
Cues
Seller Rating The feedback left for a seller by distinct users
Seller Positive Feedback Ratio The feedback ratio for a seller by distinct users
Seller Number of Other Items The number of other items sold by seller on eBay
Bidder Rating The number of feedback left for a bidder by distinct users

<Table 3>

Sample Characteristics and descriptive statistics

Item Name Frequency (n=3,697) Percent (%)
Golf Driver 3,200 86.6
Books 225 6.1
iPod 272 7.4

<Table 4>

Number of Auctions

Item Name Success (%) Failure (%) Total
iPod 176(64.7) 96(35.3) 272
Books 89(39.6) 136(60.4) 225
Golf Driver 2,840(88.8) 360(11.2) 3,200
SUM 3,105(84) 592(16) 3,697

<Table 5>

Summary of Data Collected by item (all and successful auctions)

Variable iPod Books Golf Driver All Types
Auction Scope
(ALL/SUCC)
ALL SUCC ALL SUCC ALL SUCC ALL SUCC
Number of auctions 272 176 225 89 3,200 2,840 3,697 3,105
Avg. Number of Images 3.51 3.06 1.51 1.46 3.69 3.52 3.55 3.44
Range Number of Images 0~12 0~12 0~12 0~12 0~12 0~12 0~12 0~12
Avg. Number of Bidders 11.01 10.56 1.35 2.72 8.26 9.04 8.04 8.94
Range Number of Bidder 0~74 0~73 0~30 1~21 0~85 1~85 0~85 0~85
Avg. Shipping Fee($) 8.36 6.95 11.78 6.97 30.62 30.47 27.84 28.46
Range Shipping Fee($) 0~69.59 0~69.59 0~73.25 0~73.25 0~297.25 0~297.25 0~297.25 0~297.25
Avg. Delivery Days 8.33 8.47 8.21 8.73 8.58 8.72 8.54 8.70
Range Delivery Days 4~12 4~12 2~12 2~12 1~12 1~12 1~12 1~12
Avg. Auction Duration 5.48 5.56 6.59 6.54 4.94 4.87 5.09 4.95
Range Auction Duration 1~10 1~10 1~10 1~10 1~10 1~10 1~10 1~10
Avg. Starting Price($) 47.45 31.21 28.77 28.12 48.71 39.47 47.40 38.68
Range Starting Price($) 0~250 0~250 0~1,250 0~600 0~1,270 0~700 0~1,270 0~700
Avg. Number of Repeated Uploads 6.07 6.07 2.80 3.06 2.76 1.38 1.33 1.70
Range Number of Repeated Uploads 1~33 1~33 1~7 2~6 1~49 1~13 1~49 1~33
Avg. Seller Rating 1,785 2,527 1.790 1,856 153,121 169,630 132,776 155,349
Range Seller Rating 0~249,209 0~249,209 0~52,764 0~24,551 0~487,447 0~487,365 0~487,447 0~487,365
Avg. Seller Positive Feedback Ratio 0.91 0.90 0.99 0.99 0.99 0.99 0.98 0.99
Range Seller Positive Feedback Ratio 0~1 0~1 0~1 0~1 0~1 0~1 0~1 0~1
Avg. Seller Number of Other Items 14.70 14.03 70.93 33.35 32.39 23.74 33.43 23.46
Range Seller Number of Other Items 0~872 0~872 0~893 0~473 0~932 0~932 0~932 0~932
Avg. Bidder Rating 399.22 616.98 708.64 1,791.52 464.50 523.38 474.55 565
Range Bidder Rating 0~26,845 0~26,845 0~73,468 0~73,468 0~30,626 0~30,626 0~73,468 0~73,468
Avg. Winning Price($) 53.80 43.44 33.33 39.64 83.73 78.93 78.46 75.49
Range Winning Price($) 0~299 0~204 0~1,250 0~600 0~1,270 0~700 0~1,270 0~700

<Table 6>

Logistic Regressions of Auction Success on Auction Features(All Auction Cases)

Variable iPod Books Golf Driver All Types
NOTE : Bold font indicates statistical significance at the 5% level.
Number of auctions 272 225 3,200 3,697
Number of Images -0.225
(0.273)
-2.822
(0.479)
0.078
(0.299)
0.021
(0.705)
Number of Bidders -0.046
(0.319)
0.208
(0.399)
0.068
(0.000)
0.046
(0.001)
Shipping Fee($) -0.303
(0.030)
-0.328
(0.381)
-0.020
(0.042)
-0.027
(0.005)
Delivery Days -0.043
(0.071)
-1.305
(0.435)
-0.013
(0.904)
-0.084
(0.338)
Auction Duration -0.742
(0.014)
2.026
(0.519)
-0.019
(0.849)
-0.075
(0.319)
Starting Price($) -0.080
(0.000)
0.025
(0.475)
-0.009
(0.002)
-0.013
(0.000)
Number of Repeated Uploads -0.142
(0.190)
1.595
(0.513)
-0.404
(0.031)
-0.266
(0.000)
Seller Rating -0.000
(0.976)
-0.001
(0.513)
0.000
(0.725)
0.000
(0.715)
Seller Positive Feedback Ratio -1.413
(0.400)
-23.255
(0.492)
3.371
(0.385)
0.509
(0.640)
Seller Number of Other Items 0.018
(0.182)
-0.003
(0.778)
-0.004
(0.041)
-0.003
(0.043)
Bidder Rating 22.116
(0.000)
10.671
(0.336)
8.382
(0.012)
10.338
(0.001)
Percent Concordant (%) 96.7 99.1 98.8 98.6
Likelihood Ratio Test 319.179
(0.000)
291.176
(0.000)
2060.977
(0.000)
2957.000
(0.000)
Hosmer-Lemeshow Test
( > 0.05)
0.260
(0.998)
0.015
(1.000)
0.549
(0.760)
0.347
(0.841)

<Table 7>

Linear Regressions of Winning Price on Auction (Successful Auction Cases)

Variable iPod Books Golf Driver All Types
NOTE : Bold font indicates statistical significance at the 5% level.
Number of auctions 176 89 2840 3,105
Number of Images 0.087
(0.304)
0.003
(0.955)
0.347
(0.495)
0.036
(0.015)
Number of Bidders 0.129
(0.110)
0.337
(0.000)
-0.090
(0.000)
0.444
(0.000)
Shipping Fee($) 0.185
(0.071)
-0.020
(0.806)
-0.130
(0.324)
-0.008
(0.563)
Delivery Days 0.271
(0.009)
0.073
(0.381)
-0.18
(0.194)
-0.019
(0.171)
Auction Duration -0.032
(0.667)
-0.098
(0.054)
-0.018
(0.456)
-0.014
(0.308)
Starting Price($) 0.191
(0.013)
0.914
(0.000)
0.886
(0.000)
0.854
(0.000)
Number of Repeated Upload -0.032
(0.708)
0.008
(0.866)
0.006
(0.528)
-0.042
(0.000)
Seller Rating -0.120
(0.392)
0.023
(0.705)
0.091
(0.000)
0.125
(0.000)
Seller Positive Feedback Ratio 0.105
(0.155)
0.065
(0.329)
0.006
(0.528)
0.016
(0.120)
Seller Number of Other Items 0.148
(0.308)
-0.008
(0.895)
-0.012
(0.256)
-0.017
(0.109)
Bidder Rating 0.051
(0.494)
-0.055
(0.263)
-0.021
(0.037)
-0.016
(0.129)
Adjusted R Square 0.089
(8.9%)
0.815
(81.5%)
0.719
(71.9%)
0.684
(68.4%)

<Table 8>

Hypothesis Testing Results

Hypo. Impact on Auctions Success or Winning Bid Price Remarks
H1a
H1b
The number of images → Auction Success
The number of images → Winning Bid Price
Not Supported
Supported
H2a
H2b
The number of bidders → Auction Success
The number of bidders → Winning Bid Price
Supported
Supported
H3a
H3b
The shipping fee → Auction Success
The shipping fee → Winning Bid Price
Supported
Not Supported
H4a
H4b
The delivery days → Auction Success
The delivery days → Winning Bid Price
Not Supported
Not Supported
H5a
H5b
The auction duration → Auction Success
The auction duration → Winning Bid Price
Not Supported
Not Supported
H6a
H6b
The starting price → Auction Success
The starting price → Winning Bid Price
Supported
Supported
H7a
H7b
The number of repeated upload → Auction Success
The number of repeated upload → Winning Bid Price
Supported
Supported
H8a
H8b
The seller’s rating → Auction Success Not
The seller’s rating→ Winning Bid Price
Not Supporte
Supported
H9a
H9b
The seller’s positive feedback ratio → Auction Success
The seller’s positive feedback ratio→ Winning Bid Price
Not Supported
Not Supported
H10a
H10b
The number of seller’s other items on eBay → Auction Success
The number of seller’s other items on eBay → Winning Bid Price
Supported
Not Supported
H11a
H11b
The bidder’s rating → Auction Success
The bidder’s rating → Winning Bid Price
Supported
Not Supported