Reputation Systems: eBay
When eBay first began in 1995, people worried about being able to trust strangers online. Perhaps more influentially, eBay worried that its users would not trust other users enough to buy things from them. eBay thus spent much effort creating a reputation system that would allow users to feel comfortable buying online. According to Michigan School of Information professor Paul Resnik and his colleagues, “A reputation system collects, distributes, and aggregates feedback about participants' past behavior. Though few producers or consumers of the ratings know one another, these systems help people decide whom to trust, encourage trustworthy behavior, and deter participation by those who are unskilled or dishonest” (link here). For many years, the eBay feedback and ratings policy has been of great interest to many different people, including users, bloggers, and academics. On a product page, an information box gives relevant history about the user selling the product:
The above box about user lestatmj appears to indicate that he is trustworthy. There is a star next to his name and 100% positive feedback. If you click on the ME icon next to the star, you can see lestamj's profile. He writes: "I am a 35 year old male with a beautiful wife and three wonderful children. I am a registered nurse. I love most all sports, computers, music, photography, and all things nature. Not much else to say. Happy Bidding!" Below lestamj's description of himself are all the reviews he has received:
People seem content with the user’s behavior. Overall, the UI and reputation system seem to indicate that lestamj is a trustworthy seller. However, on closer examination it becomes apparent that the UI actually obscures much relevant information. eBay uses a system of colored stars next to a username to in some way indicate the trustworthiness of the user. Look next to lestatmj's name. Next to the star, the number 27 is in parentheses. This is the number of positive ratings that the user has. It corresponds to the star color according to this chart:
The is fairly unintuitive. What looks like a gold star is first, with the fewest required ratings. Next is blue, followed by turquoise. The colors seem to have no relationship to the ordering. At a certain point, the star patterns change into shooting stars. These have the same colors in the same order, except that blue is missing. Overall, the star decisions seem arbitrary. Stanford HCI Professor Scott Klemmer thinks that they deliberately obscure reputability information in order to make users more readily trust sellers. In this sense the reputation is supposed to encourage users to trust the site, not to make informed decisions.
Journalist and writing teacher Alice LaPlante reported in her article EBay Feedback: Fatally Flawed? that this is only one of many problems with the feedback system. Look back at that review count for Lestatmj. Most of the positive reviews he received were actually as a buyer. On his profile there are actually no reviews that explicitly indicate his reputation as a seller. He has "Fast Pay" and one user "Would be Happy to Serve Again." Additionally, the 100% positive reviews say "in the last year" in small text below. A member 2000, this user has received only one review in the last year. eBay is either intentionally misrepresenting its users or simply failing to figure out what information will be most user for determining if a transaction should be made. Ultimately, for commercial sites the reputation system is as much to help the website's popularity has to help its users distinguish between trustworthy and untrustworthy sellers. Users must still be wary, and sellers should try to follow open suggestions such as LaPlante's about how to make their sites more helpful to their users.
The reviews help out the site in another way too. According to Resnik, "the expectation of reciprocity or retaliation in future interactions creates an incentive for good behavior." Merely having a reputation system means that people will be more likely to behave well. They feel more accountable for their actions. This insight is the basis for the reputation systems on many online communities, discussed in the next section.