Facebook’s Bad For You But Good For Me

Research recently published in Cyberpsychology, Behavior and Social Networking1 reveals that on average, people perceive Facebook to negatively affect other people, but do not believe themselves to be affected in the same way.

To examine this, the researchers provided an anonymous survey to 357 undergraduates.  They asked questions about Facebook usage, perceived negative effects toward others and perceived negative effects toward the self.  Using paired-samples t-tests, they found that participants believed the privacy of others was reduced due to Facebook use, but did not perceive their own privacy to be affected as much.  They also perceived later job opportunities for other people to be decreased due to a Facebook use, but did not perceive a decrease in opportunities for themselves.  There was no difference between perceptions about negative effects on self and others for personal relationships.

This does not suffer from many of the usual limitations of survey research because we are explicitly interested in user perceptions.  It doesn’t really matter if this captures their actual Facebook usage or not. The study aims to capture people’s beliefs, and that’s exactly what it does.

The researcher use this to support the presence of a third-person effect (TPE) on Facebook.  The third-person effect theoretical framework argues that people tend to view themselves as being immune to the negative effects of media, which traditionally includes forces like television, radio, film, and video games, but still believe others to be negatively affected by these media.  This stems from feelings of “biased optimism” – that you have sufficient control over yourself to avoid negative effects, but others don’t.  It seems Facebook falls into this paradigm just as easily – or perhaps it’s the Lake Wobegon Effect!

1. Paradise, A., & Sullivan, M. (2012). (In)visible threats? The third-person effect in perceptions of the influence of Facebook Cyberpsychology, Behavior, and Social Networking, 15 (1), 55-60 DOI: 10.1089/cyber.2011.0054 [↩]