Predictive Prevention: Can social networks improve social science?

Author(s)
Published on
July 20, 2014

You look at Facebook and you see what your friend had for breakfast.

Psychologists, behavioral scientists and others in the social sciences look at Facebook and they see a wealth of information about how people think and act in the real world, information that has enormous potential to answer questions that have been difficult to answer because of the costs and other limitations of conducting studies with subjects enrolled one by one. Information that could inform questions about how to prevent bullying, how to combat discrimination, how to protect people from domestic violence.

Scientists outside of Facebook and within Facebook have been quietly working for years to allow researchers to turn the everyday “likes,” comments and other data from Facebook into a range of published studies. But a 2012 study about the spread of emotions on social media that was described recently by a former Facebook scientist in a piece by Reed Albergotti and Elizabeth Dwoskin in the Wall Street Journal has led to outrage in almost every quarter.

Jason Abbruzzese at Mashable summed up the sentiment in a piece titled “Will Facebook Backlash Kill Scientists' Data Dream?

The difference between Facebook users and lab rats? Lab rats can see who's conducting the experiment. Facebook's 2012 experiment on 689,003 unsuspecting users has stoked controversy for breaking ethical norms that are widely recognized by academics and researchers. Facebook has apologized, though this is far from the first study that the social network has performed on its users.

“Far from the first” is an understatement. A review by Robert Wilson at Washington University in St. Louis in Perspectives on Psychological Science in May found 412 articles that covered “descriptive analysis of users, motivations for using Facebook, identity presentation, the role of Facebook in social interactions, and privacy and information disclosure.” Wilson wrote that his analysis was prompted by the sheer volume of research making use of Facebook data. He wrote:

A rapidly growing body of research has accompanied the meteoric rise of Facebook as social scientists assess the impact of Facebook on social life. In addition, researchers have recognized the utility of Facebook as a novel tool to observe behavior in a naturalistic setting, test hypotheses, and recruit participants. However, research on Facebook emanates from a wide variety of disciplines, with results being published in a broad range of journals and conference proceedings, making it difficult to keep track of various findings.

I don’t want to be the only online commentator defending Facebook, but in this discussion, should we really be calling Facebook users “unsuspecting”? When Facebook is showing me advertisements for things I have bought on entirely different websites? When Facebook ads include recommendations from my friends? When I receive solicitations from 10 people I know on Facebook for the same thing on the same day? All of these things tell me that I have willingly jumped into a fish bowl. Anything I post is being monitored, and every connection I make is a new piece of data about me. I do it because Facebook is a great way to stay in touch with a broad group of people and to catch up in seconds on important events in their lives.

It’s a tradeoff. A little bit of my privacy is compromised for a lot of social engagement that would be impossible with just a phone and email.

Now, does this tradeoff mean that I should be part of a social experiment? The key is consent. And this goes back to Angry Birds. We consent to having our information gathered, packaged, and sold so that we can play a game that involves flinging birds at pigs. Most of us do this without giving any thought to how that information is later used. Similarly, we agree to all kinds of things in order to keep using Facebook.

Denise Chow of LiveScience, writing on the emotional contagion study, reported that “the researchers said Facebook's Data Use Policy constitutes informed consent for the study.”

Facebook users must agree to the company's Data Use Policy before they are able to set up an account on the social media site. But, critics say the policy should not be used as part of blanket principles of informed consent, and that the study researchers did not adequately allow for users to opt out of the social experiment.

What obligation does Facebook, or Angry Birds, or the big data companies that sell our credit card purchase histories and other information have to tell us when our data are being used for scientific research?

If regulators are going to step into this area, which undoubtedly will happen, they shouldn’t single out social science or predictive analytics in health care for undue restrictions. Why not simplify the system more generally?

Instead of privacy policies that are longer than the U.S. Constitution, companies should be required to break down in fewer words how the information they gather about you may be used. They could also explain some of the benefits of allowing your information to be used for large-scale human behavior research or to help your physician better understand your health status. My guess is that if the information is presented in the right way, it will seem less scary and people would opt in.

But I may be wrong. I’ll ask my network on Facebook and see what they say.

Cropped photo by Dimitris Kalogeropoylos via Flickr.