Cool, Revealing, or Annoying? An Experimental Investigation of Facebook
Applications and Interpersonal Perceptions
Julia Woolley, Anthony Limperos, Dan Tamul (PhD Students)
This paper is based on a project from the "Psychological Aspects of Communication Technology" graduate course.
Dr. S. Shyam Sundar
For a complete report of this research, see:
Woolley, J., Limperos, A., Tamul, D. & S. Shyam Sundar(2008, August).
Cool, Revealing, or Annoying?: An Experimental Investigation of Facebook
Applications and Interpersonal Perceptions. Paper presented to the
Communication Technology Division of the Association for Education in
Journalism and Mass Communication (AEJMC) at the 91st annual convention
in Chicago, IL.
This research experimentally explored the relationship between (1) application
level (low, medium, high), (2) gender of the Facebook profile owner (male,
female), and (3) gender pairing of profile owner (male with female, female
with male) on levels of interpersonal liking, homophily, and behavioral
intent. To guide the current inquiry and advance hypotheses about the
probable mechanisms which influence perceptions of Facebook profiles,
literature involving computer-mediated communication, impression formation,
and the MAIN Model of technology heuristics was reviewed. Results indicated
that application level that applications elicit both distraction and coolness
heuristics which then differently effect feelings of interpersonal liking.
Overall, there were a variety of results and effects involving application
level, heuristic cumulation, and gender on the dependant measures.
Hypotheses & Research Question
H1: Higher levels of applications will lead to higher levels of interpersonal
homophily, and behavioral intent.
H2: Higher levels of applications will lead to lower levels of interpersonal
liking, homophily, and behavioral intent.
H3: Individuals who perceive the model profile as having applications
similar to their own will have higher levels of interpersonal liking,
homophily, and behavioral intent.
RQ1: What is the relationship between application level, model gender,
participant gender, and model pairing and interpersonal liking, homophily,
and behavioral intent.
All participants (N = 227) were randomly assigned into one condition
of a 16 cell fully-crossed between-subjects 2 (Model Gender; male, female)
X 2 (Pairing; paired, alone) X 4 (application level; none, low, medium,
high) factorial design. Before viewing the model profile, participants
were told that they were participating in a study to aid in the development
of a Facebook application which would help connect students in large lecture
classes. After providing their Facebook email address and the course for
which they were receiving credit (information which was immediately discarded),
they were “matched” with the model profile. The model profile
randomly varied in terms of model gender and whether or not the profile
picture contained one or two individuals. The number of applications on
the page also randomly varied between none, low, medium, and high. The
primary dependent variables, interpersonal liking, attitude homophily,
and behavioral intent, as well as profile coolness, distraction, and expectations
of liking were then measured by an online questionnaire after participants
had viewed the model profile.
Application level had no main effect on interpersonal liking, but a main
effect for model gender was obtained with females being liked more than
males. A significant relationship was found between application homophily
and interpersonal liking mediated by coolness. A similar relationship
was found between expectations and interpersonal liking mediated by coolness.
There was a significant main effect of application level on distraction
with high and medium levels of applications leading to significantly higher
distraction than low levels of applications. The distraction heuristic
then has an indirect effect on liking, with application level leading
to greater suspicion which partially mediates the relationship between
application level and distraction, and then with distraction mediating
the relationship between suspicion and systematic processing and systematic
processing leading to greater liking. There was also a significant Application
Level X Gender X Gender Pairing interaction on distraction with higher
application levels leading to higher levels of distraction for doubled
models, particularly when evaluated by male participants (See Figure 4).
This interaction may be interpreted in light of the fact that in evaluating
relational openness, pairing approached significance in its effect on
openness, with single models being viewed as more open than paired models.
Distraction and coolness heuristics show a cue-cumulation effect in their
effect on interpersonal liking where distraction and coolness have an
additive or “cue-cumulation” effect on interpersonal liking
Based on the findings of our study, it appears that enriching a Facebook
profile with applications may actually trigger effects that are not always
beneficial for impression formation in CMC contexts, contrary to social
information professing theory. These results also suggest that the MAIN
model heuristics approach to studying communication technology does have
utility. Further investigation is needed to investigate the specific antecedents
to the cueing of particular heuristics. Our results do show, however,
that these heuristics do play a role in how people perceive others. Also,
this finding supports the basic assumptions of the social information
processing theory as users were still able to perceive similarity and
tended to like the target profile because of perceived similarity even
in absence of all the cues which would normally be available in FTF interaction.
While the number of applications did yield some differences on our dependent
measures, differences based on type of applications could not be discerned.
Different types of applications may serve different social and personal
functions, such as broadcasting, interactive or single-person gaming,
gift-giving, and identity expression, to name just a few. These differences
may play a role in interpersonal evaluations. Also, given the fact that
application level did not directly influence evaluations of profile “coolness”,
exploratory research which attempts to ascertain what aspects of Facebook
profiles (or individual differences in evaluations of profiles) might
lead to differing evaluations of coolness would be beneficial.