May 15, 2007

Lit Review: A familiar Face(book): Profile elements as signals in an online social network (Lampe, Ellison, & Steinfeld)

In the third of a continuing series of Facebook research projects, Lampe et al; drew data from over 30,000 Facebook profiles at Michigan State University in order to uncover the relationships between the amount and type of profile elements presented and number of friends.

Walther's Social Information Processing Theory
: Online, lack of traditional cues leads to the development of new social cues, such as spelling ability.

Signaling Theory: Profile elements are signals used by individuals to communicate personal qualities that are interpreted by others in order to make judgments.
-Donath differentiates between assessment signals (which are observable qualities) and conventionial signals (indicated through social conventions). Online signals are generally conventional.
-In the world of Facebook, relationships are generally formed first offline. Thus, the structure of Facebook encourages honesty in profiles. Dishonesty is typically playful or ironic in nature.
-The researchers propose that the number of legitimate conventional signals included in Facebook profiles is proportionate to the size of one's online social network, as well as the signaling value of less verifiable cues (such as interests).

Common Ground Theory: Profile creation is motivated by a desire "to establish common frames or reference that enhance mutual understanding."
-Community membership is integral to assessing the amount of shared understandings, working to establish common frames of reference.
-Information derived from Facebook profiles works much in the same way as face-to-face "interviewing", indicating shared common ground that may enhance understanding between individuals (such as shared location or academic major).

Transaction Cost Theory: In establishing these common frames of reference through profiles, costly negotations ensue that work to enhance communication between interactants.
-Facebook profiles reduce the cost of connections by creating an easy way for individuals to search for those who share their interests or other attributes. Thus, the more information that is provided by an individual, the more likely they are to be found by others, enhancing the number of connections displayed by that individual.
-From this lens, the researchers suggest that the more verifiable elements and contact information is exhibited in one's Facebook profile, the greater the effects will be on the number of friends that person has.

The study used automated scripts to gather profile information, which was then encoded into four variables:
1. Control Variables: Network characteristics. (Sex, Length of Membership, Institutional Status, Last Updated)
2. Referents Index: Common points of reference, possibly assessment signals. (Hometown, High School, Residence, Concentration)
3. Interests Index: Conventional signals of identity. (Favorite Movies/Music/Books/TV Shows/Quotes, Interests, Political Views, About Me)
4. Contact Index: Willingness to share off-site connections. (Relationship Status, Looking For, Website, Address, Birthday, AIM, Email)
5. Dependent Variables: Total number of friends (Same School, Other School)

-Users completed 59% of fields on average.
-Median number of preferences listed: 5 interests, 1 book, 5 movies, 3 music, 0 TV shows, 36 characters in "About Me" section.
-Median number of friends: 75 same school, 68 other school, 0.53 ratio.
-Number of friends is highly correlated with undergraduate status, as well as how long the account has been active.
-The act of providing information on one's profile is highly correlated with number of friends, most notably High School (92:35), AIM (100:50), Birthday (80:26), Favorite Music (83:37), and About Me (88:56). The first three aid in supporting pre-existing bonds, such as high school bonds, while the former provide information about one's identity to all users.
-There is a weak correlation between the AMOUNT of information in profiles and the number of friends. The researchers posit two possible explanations: either a user with many friends feels social pressure to include more information, or such a user includes more information while also seeking out more people to add as friends.

One of the main limitations described by the researchers is that their study focused on the behaviors of Facebook users, but not their attitudes toward or motivations behind these behaviors, and that they did not examine the content of profile fields, but rather the existence of them.

My examination of online social networking communities will be considerably less quantitative than many of the studies I have been reviewing. An emphasis on qualitative interviewing of SNC members (both face-to-face and online) will be a considerable benefit to the current research in this field.

The Age of Egocasting?

Rosen, a technology journalist, discusses emerging online community practices in terms of a modern-day process she has coined “egocasting”. She documents the recent history of communicative technologies, which allow individuals to control with increasing precision the information they consume. Popular contraptions such as TiVo and the iPod allow individuals the capacity to avoid the sounds, images, and ideas we don’t agree with. She warns of the potential of this power for crafting a culture that is profoundly impatient and critical of all that does not align with their ideologies of choice.

TiVos and iPods will never destroy us. But our romance with technologies of personalization has partially fulfilled Krutch’s prediction. We haven’t become more like machines. We’ve made the machines more like us. In the process we are encouraging the flourishing of some of our less attractive human tendencies: for passive spectacle; for constant, escapist fantasy; for excesses of consumption. These impulses are age-old, of course, but they are now fantastically easy to satisfy. Instead of attending a bear-baiting, we can TiVo the wrestling match. From the remote control to TiVo and iPod, we have crafted technologies that are superbly capable of giving us what we want. Our pleasure at exercising control over what we hear, what we see, and what we read is not intrinsically dangerous. But an unwillingness to recognize the potential excesses of this power—egocasting, fetishization, a vast cultural impatience, and the triumph of individual choice over all critical standards—is perilous indeed.

The parallels to online communities are cutting. Though our culture frequently heralds the globalizing force of technology, there are darker implications that this technology may allow use to blind ourselves entirely to ideas and information that contest our beliefs and challenge our comfortable notions of ourselves, others, and the world at large.