Psychological Effects of Frequency and Clutter in Web advertising
Student Researcher
Sang Lee (PhD Student)
Faculty Supervisor
Dr. S. Shyam Sundar
For a complete report of this research, see:
Lee, S. Y. , & Sundar, S. S. (2002, July). Psychological effects
of frequency and clutter in Web advertising. Paper presented at the
52nd annual conference of the International Communication Association
(ICA), Seoul, Korea.
Introduction
Frequency of exposure to advertising is considered one of the fundamental
determinants of advertising effectiveness in traditional mass communication.
However, in the realm of Web advertising, frequency remains a relatively
unexplored variable, despite phenomenal growth in Internet advertising.
Understanding the function of frequency is important because it is critical
to deciding the effectiveness of advertising, regardless of whether it
is traditional media advertising or Web-based Internet advertising. Unlike
frequency, advertising clutter has generally been viewed as a factor inhibiting
(rather than enhancing) advertising effectiveness, in terms of both memory
and attitude toward ads. Given the nature of the Web that often displays
multiple banners with same sizes and shapes, ad clutter can hinder the
processing of information contained in a specific banner. The current
investigation is a modest first attempt at pinpointing the distinct as
well as combinatory effects of frequency and clutter in a typical Web
advertising environment.
Hypotheses and Research Questions
Mere exposure effects, interference effects, three-hit theory and two-factor
theory were used to generate hypotheses for this study.
H1a: Banner advertising repetition will have a positive impact on ad
recall.
H1b: Banner advertising repetition will have a positive impact on ad
recognition.
H1c: Banner advertising repetition will have a positive impact on attitude
toward ad.
H1d: Banner advertising repetition will have a positive impact on attitude
toward brand.
H1e: Banner advertising repetition will have a positive impact on trial
intention.
H2a: Ad clutter will have a negative impact on ad recall.
H2b: Ad clutter will have a negative impact on ad recognition.
H2c: Ad clutter will have a negative impact on attitude toward ad.
H2d: Ad clutter will have a negative impact on attitude toward brand.
H2e: Ad clutter will have a negative impact on trial intention.
H3a: Level of Internet experience will have a negative impact on ad recall.
H3b: Level of Internet experience will have a negative impact on ad recognition.
RQ1: For users of a news Website, controlling for time spent viewing
a Webpage, what is the relationship between level of Internet experience
and attitude toward ad?
RQ2: For users of a news Website, controlling for time spent viewing
a Webpage, what is the relationship between level of Internet experience
and attitude toward brand?
RQ3: For users of a news Website, controlling for time spent viewing
a Webpage, what is the relationship between level of Internet experience
and trial intention?
Method
A fully-crossed 7 x 2 between-participants design (7 levels of frequency
and 2 levels of clutter) with a control variable (time) was used in the
experiment. Two hundred and fifty undergraduate students (150 females
and 100 males) enrolled in communication classes participated in the experiment
for extra credit. Participants were randomly assigned to one of fourteen
conditions and were exposed to an experimental Website. At the time of
recruiting, participants were told that the study concerned students'
learning on the Internet. All participants signed an informed consent
form prior to their participation in the experiment.
Results
H1a: Supported. Advertising frequency had a significantly positive effect
on recall.
H1b: Not supported. The effect for frequency on recognition was statistically
not significant.
H1c: Not supported. The statistical analysis failed to support the hypothesis
that frequency would have positive effects on attitude toward ad.
H1d: Supported. However, the effect for frequency on attitude toward
brand was marginally significant.
H1e: Supported. The effect for frequency on trial intention was statistically
significant.
H2a: Not supported. The analysis failed to support the hypothesis that
clutter would have a negative effect on recall.
H2b: Supported. A significant main effect for clutter on ad recognition
was found.
H2c: Not supported. No statistically significant effect for clutter on
attitude toward ad was found.
H2d: Not supported. No statistically significant effect for clutter on
attitude toward brand was found.
H2e: Not supported. No statistically significant effect for clutter on
trial intention was found.
H3a: Supported. However, the effect for level of Internet experience
was marginally significant.
H3b: Not supported. The effect for level of Internet experience on ad
recognition was not significant.
RQ1, 2, and 3: No significant effect for Internet experience on attitude
toward ad, attitude toward brand, and trial intention was found.
A couple of important two-way interactions were observed. There was a
statistically significant interaction between clutter and frequency on
attitude toward ad. This interaction showed that, while frequency did
not affect users' attitude toward ad in a non-cluttered environment, frequency
of exposure had a positive impact on users' attitude toward the stimulus
ads in a cluttered environment. Another interaction, between frequency
and level of Internet experience on recognition, approached significance.
This interaction indicated that, when the frequency is low, level of Internet
experience did not matter. However, when the frequency is high, level
of Internet experience had a positive impact on users' recognition of
the stimulus ads.
Conclusion
Results from the analyses suggest that frequency can be a powerful determinant
of advertising effectiveness. Specifically, it is found that the frequency
effects were significant on ad recall, attitude toward brand, and trial
intention. Contrary to our expectations, however, the effects of banner
clutter were weak across dependent variables. The clutter effects were
significant only on ad recognition. Additionally, the significant effect
of Internet experience was observed only on ad recall.
|