Facebook & Self-Censorship

Facebook & Self-Censorship

“Self-censorship is the act of preventing oneself from speaking.”

For 17 days, from July 6th to July 22nd 2012, Facebook collected data from 3.9 million English speaking US and UK based users about whether or not they “last-minute self censor” comments and posts on Facebook. The primary aim of the report was to investigate how users are actually using social media and how to improve the Facebook platform in a way that encourages the open sharing of content without users feeling the need to self censor.

Future Tense, a partnership of Slate, was the first to break the news of the report.

The second paragraph of the Future Tense article reads as follows:

“Unfortunately, the code in your browser that powers Facebook still knows what you typed—even if you decide not to publish it.* It turns out that the things you explicitly choose not to share aren’t entirely private.” Future Tense/Slate

The author then goes on to passively dissect the implications of the aims of the Facebook study and actively draw attention to the alarming potential privacy violations that arise from in-depth studies of Facebook users.

This is where things get interesting. The Future Tense article gets picked up by Wired, the NY Daily News, and a multitude of other publications, and the next thing you know, my Facebook Newsfeed is cluttered with posts from friends wailing to the high heavens about how Facebook is reading the stuff that you’re not even writing! The hemming and hawing rose to a fever pitch in the form of a Care2 petition entitled “Facebook Stop Stalking Our Unposted Thoughts!.” The petition currently has 28,517 digital signatures.

Here’s the thing: In the onrush of alarmism brought on by news of the NSA and LifeLock’s marketing team, we actually missed a really interesting study on human behavior and how it plays out on social media networks.

Here are the Top 10 Takeaways from Sauvik Das and Adam Kramer’s Self-Censorship on Facebook:

  1. Self-censorship in general:
    It is something inherent to human social interaction that has been magnified with the advent of social media networks. Digitally, humans suffer from a condition called “last minute self censorship”. This is when a thought makes its way all the way out of your brain, down through your fingertips, and onto a screen somewhere, only to be reviewed and retracted at the last minute. This is what Facebook’s recent study was trying to get a pulse on.

    Facebook’s research looked something like this:

    • User types 5 or more characters
    • 10 minutes later those 5+ characters weren’t posted
    • User has officially last minute self-censored

    The overall results of the 17 day study look something like this:

    • 71% of Facebook users studied self-censored
    • It is likely that the remaining 29% didn’t self-censor simply because they didn’t have the chance. In other words, they didn’t post or didn’t post frequently enough in the 17 day period for accurate data to be collected.

    The general conclusion that was drawn from Facebook’s research is that people, all the people, are last minute self-censoring and as a result, Facebook users are probably missing out on some valuable opportunities to share content with each other.

  2. Posts v. Comments:
    Generally speaking, Facebook users will censor themselves more frequently when crafting a post (read: status update) than when commenting on an existing thread.

    • Users that censored “posts” censored 4.52 posts on average.
    • 33% of all potential “posts” were censored.
    • Users that censored “comments” censored 3.20 comments on average
    • 13% of all potential “comments” were censored

    Realistically, data like this is to be expected. When crafting a status update, a Facebook user is getting ready to share a thought that is meant to spark a conversation. There is a lot of pressure in throwing an idea out to the masses and hoping to get a response. The probability of on-line social failure is a lot lower when crafting a rebuttal to someone else’s brave initial post.

  3. Men vs. Women:
    Historically, women are more comfortable with sharing about themselves than men. So it comes as little surprise that:

    • Men censor 26% more frequently than women.
    • Men censored even more (11%) frequently than women as the proportion of males in either genders’ friend group increased.

    Interestingly, gender had little to no effect on the frequency of self-censorship in comments, only in posts.

  4. Age:
    Despite the researcher’s initial hypothesis that older Facebook users would censor more, research shows:

    • When crafting posts, older users censored 15% less** than younger users
    • When crafting comments, younger users censored less frequently

    This is perhaps an interesting commentary on “older” users’ understanding of the digital landscape. When reading statistics like these it is hard not to think about the phenomenon of a cranky grandfather sitting on a porch shouting: “get off my lawn” to whoever passes by, but politely addressing his friends over a game of bridge.

  5. Audience Influence:
    This was the largest focus of the data collected during Facebook’s research. Thanks to the “About Me” section on our profiles, users are able to self identify as liberal, conservative, straight, gay, male, female, old, young, etc. Facebook took a peek at this information and then did an analysis of their studied users’ friend groups asking questions like: How many of their friends share mutual friends? How long have these users been using Facebook? How diverse is this user’s friend group politically? All of these questions ultimately drive data that addresses the influence of a user’s audience on their likelihood to self-censor.

    • Users with a higher average number of friends-of-friends censored more posts. (+12%)
    • Users with more friends that were friends with each other censored fewer posts (-3%)
    • Users with more diverse friends (politically) censored less.
    • Users that utilize Facebook’s privacy features to limit what groups can see their posts censor less.

    The overall conclusion about audience’s influence on self-censorship is that the more unique circles of friends a user has within his overall collection of friends, the more they self-censor. However, the more varied, in terms of age and political orientation, a person’s overall pool of friends is the less they self-censor.

  6. The crossover between human interaction and digital interaction:
    Our social media networks exist in order to give us an expansive platform to communicate with one another. It is no surprise that natural human behavior would influence our digital behavior as well.
    In real life, people self-censor. You don’t say things around your boss that you might around your best friends, this is nothing new. Women have always been more apt to self share than men. Many people take serious consideration of their audience before they start sharing their experiences with a group.
    One idea posed by this study is that people last minute self -censor less when their audience is more politically and age diverse because they are prone to make statements that will appeal to the lowest common denominator. This is a common† technique used in broadcasting and public speaking;  hinting at an interesting crossover between face-to-face self-censoring and the extra filter that social media provides.
  7. The huge chasm between in real life and digital:
    Facebook appears to have a pretty serious problem with people only putting their best foot forward on social media. Modern conventional wisdom has taught Millennials that what they put on the internet could prevent them from getting a job one day. Everyone’s news feeds have taught us all that our peers are leading superior lives full of vacations, promotions, engagements, and puppies and our Facebook Page should be able to compete with that.
    You might assume that younger Facebook users would be less conscious about the information that they share and the older users would be more guarded. This is just the nature of maturity. However, at some point in their cyber development, Facebook users raised by Facebook users seem to have learned a valuable lesson about keeping their over-sharing to a minimum.
    You might also assume that women would be more conscious of what they post if most of their friends are male. Years of cotillion and Estee Lauder ads should back that up. Online, however, this is not the case.Moral of the story: there’s a whole new realm of psychology that exists to explain the differences between how we are online and how we are offline. It will only be a matter of time before one of us is seeing a Cyberpsychologist.
  8. Facebook is using your data?
    Absolutely, but not in the ways you might think.

    “Facebook, like many companies, runs analytics on its site. Sure, Facebook has much more sophisticated and specialized analytics than your average business site, but…If you’re looking for new reasons to hate on Facebook, I’m sorry to say that I couldn’t find any trace of the self-censored status update being sent. Nothing but meta data here.” –Faison Zutavern of Orion Web

    At the end of the day, Facebook (at least for now) isn’t violating your rights anymore than the next guy. You’re the one who chose to share your gender, political beliefs, and age on a 3rd-party-platform; Facebook is simply capitalizing on your lack of self-censorship to understand self-censorship more fully.

  9. Why Facebook CAN use your data and should:
    Ultimately, Facebook’s mission is to “give people the power to share and make the world more open and connected.” You have chosen to join a social media network whose mission is to give people the POWER to share. If the current nature of the platform is encouraging self-censorship in a way that accentuates normal human behavior, then Facebook is not doing its job. I imagine that Zuckerberg dreams of a Facebook where we are all comfortable sharing whatever we would like as long as it is valuable content that makes the world more open and connected. It is a tough balance, honest opinions and valuable opinions, but if Facebook needs to study my metadata to create a network that champions that balance (possibly in the face of in-real-life human nature) than so be it.
  10. Maybe Google+ is just the answer:
    According to Facebook’s study, Facebook users that implemented audience controls self-censored less. Google+ has the savviest of audience control tools and thus provides the perfect space for delegating who of who we know gets to see what of what we know. In theory Google+ gives its users a safe haven for self-expression.
    Unfortunately, Google+ really represents the pinnacle platform for self-censorship. You are always digitally conscious of your audience, you are never sharing your whole self with any one group of people, and you are always subconsciously aware of the implications of what you post on what happens when you Google search your own name. So in the meantime, stick to self-censoring on Facebook and maybe, eventually, they’ll tweak their UX to create a safer community for sharing valuable content. Either way, now you know they’ll at least be paying attention.
*Future Tense ultimately corrected its error (kind of): “Update, Dec. 16, 2013: This article was updated to clarify that it is the browser code, not Facebook, that reads whatever you type.”
** per 1% increase in age from the mean
†John Fiske, Introduction to Communication Studies: Studies in Culture and Communication, (Routledge, 2013), 82.
photo credit: Truthout.org via photopin cc
Enjoying our insights? Enter your email below to subscribe to our monthly newsletter.