Search Here

When Facebook first introduced the "Like" button, it became another way for people to express their opinions without needing to use words. Once someone "liked" a page, a status or some external link, it indicated to others that the user approves the subject. Since "liking" something can mean a lot more than just approval, researchers set out to find how "liking" something can influence other users. The researchers found that people who "liked" things can persuade others to do the same. According to this study, the "like" button holds a lot of power in swaying people's opinions and can greatly increase the popularity of a particular topic or person.

"Hype can work," Sinan K. Aral, one of the researchers, explained. Aral is a professor of information technology and marketing at the Massachusetts Institute of Technology. "[Hype] can feed on itself as well."

For this study, the researchers were focused on how something becomes popular and whether or not it is popular because it is good or popular due to hype. In order to test this, the researchers worked with a website from an anonymous company that did not want its information disclosed. Together, they created a page where users can submit news articles via links. Other users or readers can then comment on the links. Each comment receives a rating. Negative comments are subtracted from positive ones to achieve an overall score.

The researchers had two features to this comment section. In one section, the comments could be voted up or down. The other section, which acted as the control group, did not receive any voting. The voting was created to resemble the use of "likes." The researchers found that initially, there was more positive feedback than negative, which was predicted. The votes were 4,049 to 1942.

After utilizing the "up" or "down" feature for one group, the researchers calculated that the first person that reads a comment with a positive score was 32 percent more likely to give the same positive vote. For negative votes, however, people were not as quick to follow. The researchers found that between the two groups, the group that had the voting system had scores that were 25 percent higher than the posts in the control group. This indicates that using "like" and "dislike" features could increase traffic and help promote a particular topic.

"That is a significant change," Aral said. "We saw how these very small signals of social influence snowballed into behaviors like herding."

The study was published in Science.