[Air-l] viewing American class divisions through Facebook and MySpace

[Zeynep Tufekci]

Hi everyone,

I just wanted to chime with some preliminary data.

I have survey data collected on two separate occasions (May of 06 and
December of 06.) These are indeed convenience samples, so they are not as
strong as Estzer's data. However, they were collected from multiple sections
an introductory social science course that is a popular choice for students
looking to fulfill required coursework. The sample characteristics are
representative of my university as a whole, which is a mid-sized public
university with a racially-diverse but academically high-achieving student
body.

The May 2006 sample had 304 and December 2006 sample had 241subjects.

I can report the following:
1- About 85 percent of the total sample uses social network sites in
general.
2- Facebook is the most popular site (about 90 percent of the SNS using
students use Facebook.)
3- About 55 percent of those who use SNS use Myspace.
4- As you can tell from the above, many students use both Facebook and
Myspace. For May 06, about 36 percent reported using Facebook and not
Myspace and 39 percent reported using both. Dec 06 numbers were similar.
(There is a very small group, less than five percent that uses Myspace and
not Facebook. )

So, I compared "highest degree completed by the father" response between
those that solely use Facebook and those that use both Facebook and Myspace.


1- Indeed, the father's educational levels are higher for those that only
use Facebook.
2- The differences are statistically significant, and moreover, the
divergence between the groups, and thus the level of statistical
significance, increased between the May 06 sample and the Dec 06 sample.
(From .01 level to .000 level). In other words, the kids whose parents had
higher levels of education were even more likely to be using solely Facebook
in the Dec 06 sample. This may be a sign of the shift danah was observing in
the field. (The distribution of the parental education levels of the two
samples are quite similar so I don't think this is due to a shift in the
sample characteristics between the two data collection points.)

While many caveats are required, most important is that the term "class" as
used in danah's blog essay and parental education level are loosely
connected. Also, this is just a snapshot of a rapidly changing terrain; I
suspect the opening of Facebook to more general audiences, as well as the
flood of new Facebook "apps" that are introducing Myspace-like aesthetics
and functionality to Facebook might change this dynamic (if they haven't
already). However, fairly significant differences are there in my data and
they support the analysis danah has from her fieldwork and Eszter's data
from her survey.

-z

Zeynep Tufekci, Ph.D.
Department of Sociology and Anthropology
University of Maryland, Baltimore County
1000 Hilltop Circle
Baltimore, MD 21250
Phone: (410) 455-8064