Because data on Facebook are naturally occurring, we
avoided interviewer effects, recall limitations, and other sources
of measurement error endemic to survey-based network research
(17). Further, in contrast to past research that has used interaction
“events” such as e-mail or instant messaging to infer an
underlying structure of relationships (7, 18), our data refer to
explicit and mutually confirmed “friendships” between students.
Given that a Facebook friendship can refer to a number of
possible relationships in real life, from close friends or family to
mere acquaintances, we conservatively interpret these data as
documenting the type of “weak tie” relations that have long been
of interest to social scientists