April 25, 2013
Can Facebook Likes Be An Indicator Obesity?
Michael Harper for redOrbit.com — Your Universe Online
We give a lot of information to sites like Facebook and Google, even when we don´t realize it. It´s only when we pull back for a moment and see the big picture that we realize just how much information we´ve shared and what can be done with it. For instance, each year Google releases a list of the most searched terms from the previous year, revealing how each search for words like “iPhone” or “Beyonce” can paint a larger picture about what´s going on in the world as a whole.
Now, researchers at a Boston hospital have used information gathered from Facebook to map which areas of the nation are most likely to be affected by obesity.
According to a study by researchers at Boston Children´s Hospital, “liking” television shows and other similar interests is a pretty good indicator of obesity. Alternatively, clicking the “like” button on interests related to a healthy lifestyle, such as cycling, sports or a meat-free diet is a likely sign of a low obesity rate.
The researchers compared geotagged Facebook data with health surveys from New York City and other parts of the nation. Their results have been published in the online journal PLOS ONE.
Lead researchers Rumi Chunara, PhD and John Brownstein, PhD of Boston Children's Hospital's Informatics Program (CHIP) say Facebook has made it easier to conduct studies that require such large amounts of data. As seen each year with Google´s top searches, users give Facebook hoards of information each day, much of it about the businesses and hobbies that they frequent and enjoy. Researchers are able to collect this data, put it on a map, and observe various regional trends. This is much easier — and cost efficient — than conducting a large, nationwide poll online or over the phone, says Brownstein.
"Online social networks like Facebook represent a new high-value, low-cost data stream for looking at health at a population level," Brownstein said in a statement.
"The tight correlation between Facebook users' interests and obesity data suggest that this kind of social network analysis could help generate real-time estimates of obesity levels in an area, help target public health campaigns that would promote healthy behavior change, and assess the success of those campaigns."
Conducting the research was easy enough. The Boston team aggregated publicly available data from Facebook users and then compared it with other studies. The Facebook data resulted in a map which closely resembled that drawn by data from other health surveys and data from the Centers for Disease Control and Prevention's Behavioral Risk Factor Surveillance System-Selected Metropolitan/Micropolitan Area Risk Trends (BRFSS-SMART). In each of these studies, obesity is defined by body mass index (BMI).
In the end, the Boston team found that those areas where users more often “liked” television-related interests were also more likely to have a high rate of obesity. Those areas where users liked physical activities and healthier lifestyles had a lower rate of obesity.
For instance, in Myrtle Beach, North Carolina, 76% of Facebook profiles featured television related “likes.” People in this area were 3.9 percent more likely to be obese than people in Eugene and Springfield, Oregon, the area which had the lowest percentage of television-related likes.
This isn´t the first time a study has been conducted showing how much we can tell about people based on their Facebook likes. Earlier this year, the University of Cambridge developed a computer model that is able to accurately determine the age, gender, race and sexuality of a person based solely on Facebook likes.