METHODOLOGY
Our dataset of 180; 000 users was obtained using myPersonality1,
a Facebook application deployed in 2007. The application
allows Facebook users to complete a standard Five Factor
Model questionnaire [13, 14] and to obtain feedback regarding
their personality based on their responses. After filling
the questionnaire, users can give their consent to record their
Facebook profile information and personality scores for research
purposes. The list of features used in this research
is listed in Table 1. Many Facebook users had incomplete
profile information or security settings preventing us from accessing
some parts of their profile, and due to time constraints
and bandwidth limitations some of the features were recorded
for a fraction of the users. Consequently, not all of features
were available for all of the users, but we had at least 15; 000
data points per feature and over 50; 000 data points for most
of the features.
To a large extent our sample was representative of the general
Facebook population, with an average age of 24.15
(SD=6.55) and an overrepresentation of females (58% of females)
which may be attributed to the fact that they spend
more time on Facebook and that they are more interested in
getting feedback on their personality2.