eBay's Reputation Research Articles

Modeling Reputation as a Time‐Series: Evaluating the Risk of Purchase Decisions on eBay*

ABSTRACTReputation systems based on buyer feedback play an important role in today's online markets. In this article, we provide a rigorous methodology to establish a relationship between a seller's feedback history and risk of default. We validate this method against eBay's reputation system, using a dataset of terminated users (Not‐A‐Registered‐User or NARU) and the feedback left for them by buyers. By treating feedback rating data as a function of time, we characterize the tendency of change in seller feedback ratings in order to predict the behavior of a seller. We find that NARU sellers have significantly more negative feedback in their final weeks. Applying functional principal component analysis and classification tree methods, we find that when projecting the feedback data to an appropriate space, NARU and non‐NARU sellers can be distinguished at better than 92% accuracy. We use this to provide a quantitative mechanism for evaluating the risk of trading with a seller who has less than perfect feedback, and offer advice on how much a buyer should offer to pay, given an asking price on a commodity item and a seller's feedback history.

Book Reviews

Ben Brooks of the Becker Friedman Institute reviews “Game-Changer: Game Theory and the Art of Transforming Strategic Situations”, by David McAdams. The Econlit abstract of this book begins: “Introduces the reader to a game-theory approach to life, including business, and presents six ways to change games that people are confronted with. Discusses commitment; inviting regulation; merging or ""colluding"; enabling retaliation; building trust; leveraging relationships; how to escape the prisoners' dilemma; price comparison sites; the Newfoundland cod collapse; the real estate agency; addicts in the emergency department; eBay reputation; and antibiotic resistance. McAdams is Professor in the Fuqua School of Business at Duke University.”

Trust and privacy on the internet: The solution is ahead

Who has stopped and thought about their data out there on the internet? Data about a person: Twitter followers, Facebook friends, eBay reputation. Or data about one’s company: purchase record, customer lists, customer credit card numbers. Data are spread around, most often only protected with passwords and too easily accessible. The Digital Asset Grid (DAG) is a research project run by Innotribe (the innovation arm of SWIFT) in 2012, which envisions a solution to this problem using the following components: a crowd-sourced, user-centric identity; privacy and trust on the internet, based on the Respect Trust Framework; an infrastructure for secure storage and exchange of digital assets. This paper explores the DAG and looks at the status of the project.

Providing a Trustworthy Trading Environment with a New Authentication Protocol in Online Auctions

This research takes a two-pronged approach, identifying and then proposing a solution for the basic problem keeping many potential users from engaging in online auction trading, and discouraging those who do use these sites from buying or selling high-value items. The problem? ---The absence of a reliable user registration process. To remedy this situation, this study proposes a new authentication protocol adding two levels of protection and providing users with a trustworthy trading environment through stronger identification and authentication functions. Keywords: Online Auction, E-commerce Trust, User Authentication, eBay Reputation

The Impact of a Seller's Ebay Reputation on Price

eBay provides a large amount of data – enough data to investigate how markets address the problem of asymmetric information. This study analyzes 183 auctions of Apple iPod Touches on eBay.com to find the impact of a seller's rating on price. I found that a seller's percent of negative rating significantly impacts the price they ultimately receive.

THE DYNAMICS OF SELLER REPUTATION: EVIDENCE FROM EBAY*

We construct a panel of eBay seller histories and examine the importance of eBay's reputation mechanism. We find that, when a seller first receives negative feedback, his weekly sales rate drops from a positive 5% to a negative 8%; subsequent negative feedback ratings arrive 25% more rapidly than the first one and don't have nearly as much impact as the first one. We also find that a seller is more likely to exit the lower his reputation is; and that, just before exiting, sellers receive more negative feedback than their lifetime average.

The relationship between website quality, trust and price premiums at online auctions

This study measures the value of website quality in terms of its impact on trust, intention to transact and price premiums. Prior research on online auctions has focused on the use of reputation systems for building trust in online auction vendors and subsequently to generate price premiums. This study examines the extent to which trust can be induced by improving the quality of online auction listings. A survey of 701 eBay users is conducted which compares the price premiums of two nearly identical online auction businesses, one that has online auction listings with a perceived high quality and the other that has substantially lower perceived quality. Results of this study indicate that website quality can explain 49% of the variation in the trust for eBay sellers. In fact, it shows that sellers with good website quality are all perceived to be equally trustworthy regardless of their eBay reputation; whereas sellers with poor website quality are not perceived to be trustworthy even if they have a high eBay reputation score. The results also show that the trust resulting from increased website quality increases intention to transact and results in price premiums of 12% (on average) for sellers with higher quality listings. Theories from marketing, economics, and social psychology are used to explain why website quality induces trust in unknown vendors without providing any concrete evidence regarding the vendor's past history.

National and Cultural Differences in the C2C Electronic Marketplace: An Investigation into Transactional Behaviors of Chinese, American, and French Consumers on eBay

With the Internet connecting every part of the world, electronic marketplaces provide a world-wide transaction platform for global businesses. However, consumers from different countries demonstrate distinct purchasing behaviors online. This study uses eBay as the consumer-to-consumer electronic marketplace to investigate the eBay mechanisms and purchasing habits of consumers from the United States, France, and China to evaluate the effects of cultural and national differences on purchasing habits. This study uses data on national transactions and reputation profiles of sellers gathered from eBay.com, eBay.fr, and eBay.com.cn, including both cross-sectional data and panel data to provide a comprehensive comparison of cultural behavior in the trading process as well as the influence of eBay's reputation management system on online pricing in these three countries. Significant differences confirm that purchasing behavior differs in various national consumer-to-consumer electronic marketplaces and the importance of cultural behavioral patterns.

Buying reputation on eBay: Do recent changes help?

In this paper we use simple repeated field experiments to show how eBay's reputation system can be manipulated by purchasing cheap positive feedbacks. Our experiments confirm that 'feedback purchase' is still possible even under the new rules that eBay recently adopted to eliminate 'padding'. In discussing these rules we conclude that changes have strengthened the reliability of the feedback mechanism with respect to 'retaliatory' behaviours, but have weakened it with respect to 'feedback purchase'. Finally, we discuss the possible impact of feedback manipulation on eBay and analogous e-commerce platforms, and propose a simple alternative fee policy likely to improve the reliability of the mechanism.

Fuzzy computational models for trust and reputation systems

In the recent past, a considerable research has been devoted to trust and reputation mechanisms to simplify complex transactions for open environments in social networking, e-commerce, and recommender systems (RS). In real life, we come to know about others through our social circle according to their reputation which is a public view. However, it is not always adequate to depend solely on the public view and therefore a trust measure is required to give a personalized view of the future encounters with a specific partner. In this paper, we propose fuzzy computational models for both trust and reputation concepts. Reciprocity and experience are used for trust modeling while the proposed reputation model is a fuzzy extension of beta reputation model. A two-level filtering methodology is proposed to benefit to a large extent from both the concepts separately. In order to justify the proposed models, we compared them with the existing reputation models for movie RS. The experimental results show that the incorporation of trust and reputation concepts into RS indeed improves the recommendation accuracy and establish that our models are better than beta and the popular eBay reputation models.

What is Google's Reputation Score? A Method for Modified Self-Regulation of Search

Search engines have evolved along with the internet, and now control massive amounts of data that they often manipulate. In addition to these deceptive search practices, search engines are currently facing a slew of trademark lawsuits from owners of protected marks as well as criticism from privacy scholars. As gateways to the modern formulation of the internet, search engines resemble a bottleneck for internet traffic ideally suited for a streamlined regulatory approach to cure the current woes of search.This paper proposes a method of modified self regulation where search engines are required to create a feedback system for their search results not unlike eBay's reputation scores or Slashdot's comment and moderation system, effectively giving the community of users the ability to rate the relevance of search results, rather than a black-box algorithm. This, of course, would not be done without legislation. The simplest way to accomplish this feedback regime would be to require search engines to implement it or else lose their immunities provided under §230 of the Communications Decency Act and §512 of the Digital Millennium Copyright Act, which currently immunize search engines for torts that are committed by sites to which they link.Once a feedback regime is implemented, search engines would no longer be able to misrank search results, return results for counterfeit goods, or link to a false story damaging a person's reputation without being noticed, talked about, and rated by the watchful eye of the online community.

Online Art Auctions à la Française and à l'Américaine

This study extends previous research into online auctions by introducing a comparative cross-cultural dimension and focusing on heterogeneous goods. Samples of artwork auctions taken from two independent eBay sites—eBay USA and eBay France—are analyzed to detect culturally specific patterns of seller and bidder behavior. The analysis of these two samples reveals systematic differences in the art auctioned, winning bids, starting prices, and price dynamics, as well as bidders' “flocking” behavior and sellers' reliance on reserve auctions and other risk-management strategies. American sellers tend to offer more valuable items and to garner much better closing prices than French sellers. French auctions, by contrast, are dominated by domestic sellers; further, the number of bidders is more highly correlated with the final sale price in French than in American auctions. Finally, the data regarding art auctions disclose contrasting orientations to eBay's reputation system in that French bidders tend to be more attentive to negative feedback than Americans.