Drawing With Data: How News Organizations Are Using Facebook’s Data to Tell Stories
In a world where hundreds of new technology startups are vying for the attention of the masses, the collection of user data has become a prized possession. It is no wonder then that Facebook—a company with a market cap now over $400 billion and a monthly user base of nearly 2 billion people—has a strong foothold on data collection. Facebook gathers information from its users through their profiles, likes, check-ins, mutual friends, etc. With this wealth of information, news organizations have looked to Facebook to manipulate their data in a way that can be visualized and presented in new and exciting ways. However, there are still questions about how representative this data truly is in relation to the entire population. Here are some examples of stories that have been written using data from Facebook to present information and draw conclusions.
The Facebook Primary, by Matthew Conlen and Reuben Fischer-Baum (Five-Thirty-Eight)
For this story, Five-Thirty-Eight obtained their data from Dustin Cable, a Data Scientist at Facebook. They counted “likes” from each candidate’s official Facebook page and graphed them based on the county or ZIP code of the user liking the page. By associating each “like” with a user’s location, Conlen and Fischer-Baum were able to map out candidate support nationwide. Of course, the data was not predictive, as according to Facebook’s model Bernie Sanders would have won the election had it been conducted when the data was gathered. This makes sense considering that according to the article, only 58 percent of adults actually use Facebook. While Facebook seems to have a stronghold on information, there are still gaps that leave out important information about the collective opinions of the country.
Up Close on Baseball’s Borders, by Tom Giratikanon, Josh Katz, David Leohardt and Kevin Quealy (The Upshot)
The Upshot also used data straight from Facebook profiles in order to analyze regional differences in MLB team preferences. Facebook as a company released an analysis that went as far as calculating a fan’s “devotion”—as in how many fans like only their favorite team. They also calculated the relative conservatism of each fan base by looking at what pages the fans of each team liked in addition to their own.
This presented surprising results in North Carolina, where there is no official MLB team. Whereas one would think that the state would be split by neighboring teams such as the Washington Nationals, Baltimore Orioles, and Atlanta Braves, the data told a different story. In fact, the majority of the state is occupied by fans of the New York Yankees. While this may seem surprising considering that the states are over 400 miles apart from each other, it makes sense. The Yankees are one of the most commercially successful sports organizations in the world, with over 8.5 million likes on Facebook. In addition, the recent influx in residents moving from northern states to North Carolina could help explain the trend. The Yankees are also surprisingly popular in states like Florida, Virginia, and Louisiana, so it would be interesting to see how these anomalies are created.
It is common knowledge that Facebook collects an immense amount of data about each user that creates an account on its site. By knowing what to look for, news organizations have been able to create impressive visualizations that tells us a lot about ourselves on a local, or even national scale. However, it is important to remember that not only is Facebook only a sample of the actual population, but also a simple “like” on a page is not always indicative of ones opinions or preferences.