Online reviewers follow the crowd… unless they listen to friends
Knight and Day. The A-Team. Eat Pray Love. The Last Airbender. Toy Story 3. The Sorcerer’s Apprentice. Inception. Twilight: Eclipse. Wall Street: Money Never Sleeps.
How will you rate the movies of 2010?
A new study from the University of Washington Foster School of Business and the Wharton School of Business suggests that, if you pay attention to online movie reviews, you are likely to follow the crowd. If, on the other hand, you consider the online reviews generated by friends, you’ll probably come to your own conclusion.
Friends vs. the crowd
The study examined the behavior of 20,000 citizen movie reviewers on the site Flixster.com, considering 60,000 reviews of 17 films released in 2007. Flixster’s partnership with social networking giant Facebook allowed the researchers to compare the degree to which an individual reviewer’s judgment was influenced by prior reviews posted by “friends” versus the “crowd”—the average rating of the Flixster masses.
“We examined how social influences affect critical thinking,” says co-author Young Jin Lee, a doctoral student at the Foster School.
After controlling for relevant factors, the authors find that prior exposure to a crowd rating increases the likelihood that a reviewer will follow suit. So, for instance, if an individual reviewer first observes a crowd rating of 4.3 out of 5 stars for a film, the reviewer is likely to award a more positive rating as well. In other words, herding occurs—the same phenomenon that corrals diners to a crowded restaurant, regardless of its quality.
When a movie reviewer first consulted the reviews of friends in his or her social network, however, the herding effect was significantly moderated.
“Herding is irrational behavior,” explains co-author Yong Tan, an associate professor of information systems at the Foster School. “Learning from people you know helps you return to rationality and reduce herding behavior.”
Tan adds that when people examine their friends’ reviews, they see more than a number. Social networking allows for chat, e-mail, message boards—a chance to explain one’s position on a movie. All of this online word of mouth impacts an individual’s rating, but far less dramatically than does the verdict of the crowd.
Manipulation a risky strategy
The study has broad implications for firms doing business in the interactive era of Web 2.0. Online product reviews play a critical role in the success or failure of any new product—whether a film, a smart phone or a dog food. It can be tempting to a firm to intervene into the review process.
“In the short term, manipulating user reviews at the beginning of a product release to induce herding behavior can be good for business,” Lee says. “But in the long term, the entire integrity of the product can be questioned, and trust diminished, which would be bad for business.
“In the end, when everything converges,” adds Tan, “it is quality that matters.”
“Do I Follow My Friends or the Crowds? Informational Cascades in Online Movie Rating” is the work of Young Jin Lee and Yong Tan of the UW Foster School of Business, and Kartik Hosanagar of The Wharton School of Business at the University of Pennsylvania.