Content uploaded by Zongyuan Wang
Author content
All content in this area was uploaded by Zongyuan Wang on Jan 08, 2016
Content may be subject to copyright.
This article was downloaded by: [University of Illinois at Urbana-Champaign]
On: 03 September 2014, At: 15:04
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of Interactive Advertising
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/ujia20
Doing It All: An Exploratory Study of Predictors of
Media Multitasking
Brittany R.-L. Duffa, Gunwoo Yoona, Zongyuan (Glenn) Wangb & George Anghelcevc
a University of Illinois, Urbana-Champaign, Urbana, , Illinois, USA
b University of Missouri, Columbia, Missouri, USA
c Pennsylvania State University, University Park, Pennsylvania, USA
Accepted author version posted online: 30 Jan 2014.Published online: 13 Mar 2014.
To cite this article: Brittany R.-L. Duff, Gunwoo Yoon, Zongyuan (Glenn) Wang & George Anghelcev (2014) Doing It
All: An Exploratory Study of Predictors of Media Multitasking, Journal of Interactive Advertising, 14:1, 11-23, DOI:
10.1080/15252019.2014.884480
To link to this article: http://dx.doi.org/10.1080/15252019.2014.884480
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
Journal of Interactive Advertising, 14(1), 11–23
Copyright C
2014, American Academy of Advertising
ISSN: 1525-2019 online
DOI: 10.1080/15252019.2014.884480
Doing It All: An Exploratory Study of Predictors
of Media Multitasking
Brittany R.-L. Duff and Gunwoo Yoon
University of Illinois, Urbana-Champaign, Urbana, Illinois, USA
Zongyuan (Glenn) Wang
University of Missouri, Columbia, Missouri, USA
George Anghelcev
Pennsylvania State University, University Park, Pennsylvania, USA
Multitasking with media is increasing. This shift in media con-
sumption presents challenges to advertising practitioners and re-
searchers because it may affect people’s attention, perception, and
memory for advertising contained in those media. However, while
audience multitasking behavior has recently received increased
attention, the individual predictors of media multitasking are un-
derexplored. To better understand the audience factors associated
with heavy media multitasking we conducted a survey with sam-
ples from a student population (N=308) and a national consumer
population (N=501). Age and gender were significant predic-
tors only in the national sample, while personal control and need
for simplicity were predictors only in the student sample. Results
also indicated that sensation seeking and creativity were signifi-
cant predictors of multitasking in both samples. Interestingly for
advertisers, increased perception of advertising utility was also a
predictor of multitasking in both the student and national sample.
Keywords media multitasking, individual differences, advertising
utility, creativity, sensation seeking
Today, more than ever, portable media technology and flexi-
ble media devices allow people to access content at more times
Address correspondence to Brittany R.-L. Duff, Charles H. Sandage
Department of Advertising, University of Illinois, Urbana-Champaign,
103 Gregory Hall, 810 South Wright Street, Urbana, IL 61801. E-mail:
bduff@illinois.edu
Brittany R.-L. Duff (PhD, University of Minnesota), assistant pro-
fessor of advertising, Charles H. Sandage Department of Advertising,
University of Illinois, Urbana-Champaign.
Gunwoo Yoon (MS, KAIST), doctoral student, Institute of Com-
munications Research, University of Illinois, Urbana-Champaign.
Zongyuan (Glenn) Wang (MS, University of Illinois), doctoral stu-
dent, School of Journalism, University of Missouri.
George Anghelcev (PhD, University of Minnesota), assistant pro-
fessor of advertising/public relations, College of Communications,
Pennsylvania State University.
and places, using it when and where they want. Along with this,
media multitasking is on the rise. Multitasking—engaging in
two or more tasks simultaneously—is not new, but media mul-
titasking is a more recent phenomenon. For advertisers, media
multitasking is important because the changing way that peo-
ple interact with media has implications for ads placed in those
media. Yet despite its potential to affect ad perception, we still
know little about who is multitasking with media and why they
might be doing so.
Media multitasking consists of consuming media content
while simultaneously engaging in other tasks. Some have de-
fined media multitasking as consuming a medium while en-
gaging in a nonmedia task (e.g., watching TV while studying;
Jeong and Fishbein 2007), and others have considered it con-
suming two or more media simultaneously (e.g., listening to the
radio while on the computer; Ophir, Nass, and Wagner 2009;
Roberts and Foehr 2008). Recent academic research and indus-
try data reveal that both types of media multitasking (media
multitasking with nonmedia tasks and consuming multiple me-
dia simultaneously) are pervasive phenomena, with important
implications for advertising. In 2008, 45% of Internet users re-
ported that they used at least one other medium simultaneously
while they were online, a number that increased to 54% by
2010 (Moses 2010). Nielsen’s Cross Platform Report (2012)
also shows that simultaneous media use has become common-
place, with more than 80% of Americans reporting that they use
tablets or smartphones while watching TV. Currently, 90% of
tablet owners multitask while using their tablet; though there
are some generational differences in the amount of multitask-
ing, with millennials multitasking 44% of the total time they
use their tablets and pre-boomers multitasking 37% of their
tablet time (Moses 2012). The consumption of multiple media
simultaneously now adds six hours to the average media day: if
the same multitasked activities were performed sequentially, a
11
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
12 B.R.-L. DUFF ET AL.
30-hour “extended” day would be needed to fit in all of the me-
dia being consumed (Ipsos MediaCT and Internet Advertising
Bureau [IAB] 2012).
Noting the growth of media multitasking, scholars (e.g.,
Roberts and Foehr 2008; Pilotta and Schultz 2005) and media
and advertising professionals (e.g., New Zealand Broadcasting
Standards Authority 2007; Ipsos MediaCT and IAB 2012) have
called for research addressing consumer experience with mul-
titasking. To date, most studies on how consumers experience
media content and ads have primarily looked at exposure to one
medium/message, in isolation or as part of a sequential media
experience. However, this may no longer be how many message
exposures are actually experienced.
To begin addressing the research gap, we conducted a survey
on propensity for media multitasking and on individual factors
that may affect (or be affected by) the way media content and ad-
vertisements are processed during multitasking. The limitations
of survey research based on self-reports notwithstanding, it is
crucial that we begin to understand the phenomenon of media
multitasking as it affects ads. To begin this, we need to under-
stand audience-level predictors of media multitasking. There-
fore, the goal of this study is to lay much needed groundwork
on individual differences as predictors and potential moderators
of multitasking, which we hope will be useful for understanding
the multitasking audience mind-set and motivations as well as
set the stage for future advertising research designed to investi-
gate causal relationships.
To begin understanding the media multitasking audience, we
conducted a survey with a sample of college students (N=308)
as well as a sample from the general population (N=501). We
were interested in multitasking predictors in college students
because, as others have pointed out, college students are heavy
media multitaskers and therefore it makes sense to sample from
this particular group. Student samples are also more homoge-
neous (Calder, Phillips, and Tybout 1981), and a student sample
allows us to connect our findings with the existing studies of me-
dia multitasking. We were also interested in how certain traits
and preferences may predict media multitasking both within
the heavily multitasking, more commonly studied population of
college students as well as how multitasking predictors work
in a less homogenous, broader national sample, similar to what
is used in most industry studies. Our efforts were motivated by
past findings on multitasking performance, which suggest pos-
sible differences between these two populations in terms of the
propensity and ability to multitask.
MULTITASKING BACKGROUND
Extant research on media multitasking has concentrated on
the impact of multitasking on performance, examining differ-
ences in consumers’ ability to multitask rather than differences
that might lead to their motivation or propensity for engaging in
multitasking. Divided attention has been the most widely used
construct to explain performance in simultaneous task research
(Craik et al. 1996; Konig, Buhner, and Murling 2005). It is
thought that mental resources are shared by tasks competing
for attention and action (Kahneman 1973), and thus multiple
tasks may interfere with one another (Monsell 2003; Pashler
1994). Studies that have looked specifically at media multitask-
ing have found that competition for processing resources may
lead to lowered encoding of media content (e.g., Zhang, Jeong,
and Fishbein 2010; Voorveld 2011) and decreased counterar-
guing (Jeong and Hwang 2012; Petty, Wells, and Brock 1976).
This literature on the performance aspects of multitasking has
provided valuable insights into potential consequences of mul-
titasking and its implications for processing advertisements.
In contrast with the performance literature, very few studies
have looked at the individual predictors of multitasking with
media. Because some consumers are increasingly choosing to
multitask while consuming media, understanding what factors
may be causing them to engage in simultaneous media use in
the first place is important. Such knowledge of individual differ-
ences could enable future work on how people might respond
to ad messages in the media given their specific personality or
processing styles; it could also provide a roadmap for designing
advertising and media in ways that can appeal to multitaskers.
In fact, Ipsos Media and the IAB (2012) noted the need to under-
stand the multitasking audience’s “mind-set and motivations” to
aid with ad creation and format; similar calls have been issued
by scholars (e.g., Roberts and Foehr 2008).
MEDIA MULTITASKING AND INDIVIDUAL
DIFFERENCES
A handful of studies have examined individual differences
that may affect multitasking propensity, but they have primarily
utilized student samples. Variables that have previously been
found to predict multitasking behavior in college students in-
clude technology innovativeness and use of social networking
sites (Zhong, Hardin, and Sun 2011), neuroticism (Poposki,
Oswald, and Chen 2009; Wang and Tchernev 2012), and at-
tentional impulsiveness and sensation seeking (Sanbonmatsu
et al. 2013). One study that had a nonstudent (14–16 years old)
sample also found sensation seeking was a predictor of media
multitasking, as was gender (Jeong and Fishbein 2007). While
these studies have looked at some predictors for specific pur-
poses, we believe that a broader look at media multitasking, as
well as more work with nonstudent samples, is needed. In this
way we can better understand media multitaskers as an audience
and have a starting point for work on how they might interact
with advertising.
The individual preference or proclivity to engage in media
multitasking behaviors has been tied behaviorally to poor at-
tentional task performance. Ophir, Nass, and Wagner (2009) re-
ported that heavy media multitaskers (HMMs), who self-report
a higher propensity to use multiple media simultaneously (and
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
DOING IT ALL 13
therefore they should be more well practiced), actually per-
form worse on focusing tasks and switching between tasks com-
pared to light media multitaskers (LMMs). Slower, less efficient
task switching may result from HMMs’ involuntary attentional
breadth and reduced ability to filter out task-irrelevant interfer-
ence (Ophir, Nass, and Wagner 2009; Lin 2009). Multitasking
propensity has also been shown to negatively correlate with
working memory task performance (Sanbonmatsu et al. 2013).
Heavier multitaskers may use breadth of information rather
than depth, perhaps viewing information more democratically,
sampling from multiple sources to get more information, rather
than selectively deciding which information may be best (Lin
2009; Cain and Mitroff 2011). Indeed, other studies showed that
if the “irrelevant” information from an attentional task later be-
comes useful or predictive, then heavy multitaskers are better
at utilizing it than are light media multitaskers (Lui and Wong
2012; Cain and Mitroff 2011). Therefore, a key assumption
in our study is that higher propensity to multitask is connected
with a diminished ability to ignore irrelevant or distracting infor-
mation. Yet the attentional breadth and the decreased ability of
HMMs may well be associated with other individual differences
that could be beneficial in the context of consumer interaction
with media. Specifically, we identify demographics (age) and
traits connected to cognitive abilities (cognitive failure, personal
control) that might be related to audience (in)ability to attend or
use goal-directed attention. In addition to the demographic and
cognitive traits thought to be connected to media multitasking,
HMMs’ attentional breadth may contribute to, or be caused by,
a need for stimulation and increased associations. In particular,
it seems possible that heavier multitaskers may tend to have
more of a creative mentality and prefer more complexity (less
simplicity).
Finally, we were interested in how multitaskers specifically
relate to advertising. As discussed, HMMs seem to be unable
(or unwilling) to inhibit goal-irrelevant distractors in task per-
formance, but they are also more able than LMMs to utilize
these distractors later (e.g., better multisensory integration by
HMMs; Lui and Wong 2012). Advertising is frequently irrele-
vant to the specific media content in which it appears, and ads
have been conceptualized in the advertising literature as “dis-
tractors” when they are irrelevant to the audience goal (e.g.,
Duff and Faber 2011). However, the breadth-biased style of at-
tention in HMMs might help them take in more information that
is deemed irrelevant by LMMs (e.g., advertising). Therefore, it
seems possible that heavier multitaskers would perceive ads to
generally be more useful and relevant. In the next sections, we
detail our rationale for additional predictors of media multitask-
ing. These variables include demographics (age and gender),
executive control predictors (cognitive failure, personal con-
trol), personality and preferences (sensation seeking, creativity,
imagination, need for simplicity), and advertising (ad utility).
Demographics: Age and Gender
In looking at predictors of media multitasking in terms of
advertising and media audiences, demographic variables such
as age and gender are a basic starting point. Age differences
in media multitasking behavior are expected due to both eco-
logical and cognitive factors. Younger generations may report
higher levels of multitasking behavior because the current media
environment encourages multitasking with media (Carrier et al.
2009). Today’s media consumption environment may contribute
to generational differences by affecting the resources available
to younger people (Carrier et al. 2009) or even by fundamen-
tally altering their ability to split attention (Yap and Lim 2013).
Beyond generational differences, there are also cognitive ele-
ments of aging that are expected to interact with multitasking.
For older adults, lowered flexibility in allocation of attentional
resources (Prakash et al. 2009) could make multitasking more
difficult. Older adults show increased working memory disrup-
tion from multitasking when compared to younger adults (Clapp
et al. 2011). Most studies looking at predictors of media multi-
tasking have used a student sample only and therefore would not
have the ability to detect differences in age. Because a national
sample allows the examination of age differences, the following
hypothesis will be tested:
H1: Age is negatively related to propensity for media multitasking.
Some past studies have found that there may be gender differ-
ences in self-reported propensity to multitask. In a sample of
14- to 16-year-olds, Jeong and Fishbein (2007) found that fe-
males were more likely than males to report multitasking. In
a small sample (N=14) fMRI study correlating brain density
with distractibility, Kanai and colleagues (2011) found that fe-
males were more distractible than males. However, many other
studies have either not found (Sanbonmatsu et al. 2013; Zhong,
Hardin, and Sun 2011; Ophir, Nass, and Wagner 2009) or not
reported gender effects of multitasking with media. While the
majority of studies reporting gender state that there are no differ-
ences, it is important to note that these studies were conducted
primarily with college students. Given the inconsistent findings
relating to gender and media multitasking and lack of research
in nonstudent populations, the following research question is
posed:
RQ1: Are there differences by gender in self-reported media multi-
tasking?
Cognitive Variables: Cognitive Failures
and Personal Control
Cognitive failures consist of everyday lapses in attention,
memory, and perception (Broadbent, Cooper, and Parkes 1982).
In terms of heavy media multitasking behaviors, one partic-
ular dimension of the cognitive failures questionnaire (CFQ),
“distractability,” seems particularly relevant. The distractability
dimension measures the ability to avoid distraction and consists
of items such as “Do you daydream when you ought to be lis-
tening to something?” and “Do you read something and find
you haven’t been thinking about it and must read it again?”
(Broadbent, Cooper, and Parkes 1982). Recently, the dis-
tractability subscale of the CFQ was shown to predict perfor-
mance on a behavioral measure of attentional capture (Kanai
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
14 B.R.-L. DUFF ET AL.
et al. 2011). Thus HMMs, who are more susceptible to interef-
erence from irrelevant information, might also score higher on
distractability items than LMMs.
In sum, although the proposition that those who are heavier
multiaskers are higher in self-reported cognitive failures has not
yet been tested empirically, there is indirect evidence in favor of
it. For example, HMMs have been shown to have lesser ability
to filter out distractors (e.g., Ophir, Nass, and Wagner 2009).
Consequently, we expect the following:
H2: Cognitive failure is positively related to media multitasking.
Multitasking with media could be seen as a way for consumers to
take control of their media environment. However, with a lesser
ability to filter out irrelevant information or items not directly
related to their immediate goal (e.g., Ophir, Nass, and Wagner
2009; Sanbonmatsu et al. 2013; Lin 2009), it may be that heavier
multitaskers actually feel less in control of their responses. The
extent to which individuals feel that they are in control when
confronted with a problem or decision is termed personal control
(Heppner and Petersen 1982). We propose heavy multitaskers
may feel less personal control because of their lesser ability to
control what they are able to ignore or filter out. More formally,
we propose:
H3: Personal control is negatively related to media multitasking.
Personality and Preferences: Sensation Seeking,
Creativity, Imagination, and Need for Simplicity
Sensation seeking is a personality trait with a biological ba-
sis and is defined as “the need for varied, novel, and complex
sensations and experiences and the willingness to take physical
and social risks for the sake of such experiences” (Zuckerman
1979, p. 10). It has been used in communication literature as
an individual difference that can help predict media use (Dono-
hew, Lorch, and Palmgreen 2006). It may be that HMMs choose
to multitask more because they want an increase in arousal or
stimulation, something which is driven by trait-level differences
in sensation seeking. Perhaps because of this intuitive connec-
tion, sensation seeking is one of the few personality predictors
that has been explored in studies of media multitasking. Re-
cent work has shown evidence of a link between multitasking
and sensation seeking in college student samples (Strayer and
Watson 2012; Sanbonmatsu et al. 2013), and it has been shown
to be positively related to media multitasking among 14- to 16-
year-olds (Jeong and Fishbein 2007). It therefore seems very
likely that sensation seeking is significantly related to media
multitasking; therefore, the following is proposed:
H4: Sensation seeking is positively related to media multitasking.
Decreased ability to filter distractors provides clear expecta-
tions for a relationship between media multitasking propen-
sity, age, and cognitive factors, such as cognitive failure and
personal control. However, there may be other connected vari-
ables that do not relate directly to performance. Distractability,
a wider associative horizon, and defocused attention have also
been connected to trait creativity (Eysenck 1995). Creativity as
a construct embodies originality and appropriateness (Amabile
1996), but creativity can also be seen as a proclivity toward
associations, or bringing previously unconnected ideas together
and revealing their relationships (e.g., Reid and Rotfeld 1976).
Carson, Peterson, and Higgins (2003) found that high creative
achievement was strongly correlated with reduced ability to
screen out irrelevant stimuli. Thus, heavy multitaskers engaging
in breadth-based processing might be more likely to integrate the
“irrelevant” distractors into the formation of new connections
among concepts or ideas.
A person’s creative mentality—as indicated by the big five
factors of personality—is indicated by both perceived creativity
and imagination (Johnson 1994). As a personality trait, imagina-
tion is about an individual’s propensity for fantasy, daydreams,
and creative products (e.g., art), whereas creativity tends to be
more cognitive in nature, dealing with propensity for formation
of novel connections and challenging the status quo. Funda-
mentally, the creative process is tied directly to the ability to
generate new or unconventional ideas, rather than relying on a
“default” of the previously experienced (e.g., Berns 2008). The
ability to generate things not previously seen or experienced is
the essence of imagination. Therefore, it is possible that imag-
ination may be linked to multitasking propensity in a manner
similar to creativity. To date, there has been little work looking at
media attention and imagination; however, it may be an impor-
tant part of an individual’s larger creative mentality. Therefore,
we propose:
H5: Perceived creativity is positively related to media multitasking.
H6: Imagination is positively related to media multitasking.
Need for simplicity describes variation in preferences for com-
plexity versus simplicity. This need may be more situationally
affected than our previous variables, as experiencing messy or
complicated things in one’s immediate environment can induce
a higher need for simplicity (Liu, Smeesters, and Trampe 2012).
However, simplicity preferences are also affected by more sta-
ble individual differences. For example, highly creative indi-
viduals have been shown to prefer complex stimuli over simple
stimuli (Eysenck 1995). Only people that found disorganization
aversive responded to messiness by more strongly endorsing
simplicity (Liu, Smeesters, and Trampe 2012). It is possible
that being exposed to more media, with less ability to filter
distracters, could lead heavier multitaskers to desire simplicity.
However, with a more breadth-biased information processing
style, heavier multitaskers might not find disorganization aver-
sive and would therefore not be affected by being exposed to
disorganized or messy content configurations. Therefore, we
hypothesize the following:
H7: Need for simplicity will be negatively related to media multi-
tasking.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
DOING IT ALL 15
Advertising: Ad Utility
Our previous variables have largely dealt with audience-level
individual differences that may be connected to levels of me-
dia multitasking. While there have been calls to understand the
mind-set and motivations of multitaskers as a media audience,
we also wanted to explore multitaskers’ relationship with adver-
tising. Assuming that people can use media, including advertis-
ing, to fulfill varied goals and needs, a general utility measure
may be the best to start with in this initial exploration. Past
research has found that perceptions of advertising utility are
related to self-reported exposure to the medium in which the
advertising is placed, as well as to self-reported attention to the
ads (O’Keefe, Nash, and Liu 1981).
An intriguing but previously unexplored possibility is that
heavy multitaskers’ seemingly lesser ability to filter out irrel-
evant information may increase their perceived utility of the
advertisements they encounter while multitasking with media.
Indeed, while the lack of filtering of irrelevant information could
be seen as negative (in that it may make certain goal-directed
tasks more difficult to complete), it could also serve to allow
the irrelevant information to form new associations or connect
with previously unconnected concepts. For example, it has been
shown that an irrelevant sound, which helped to predict the on-
set of a target, was actually utilized more effectively by HMMs.
While heavy multitaskers were worse than light multitaskers at
performing an attention task in absence of irrelevant sounds,
they actually improved at the task when the additional (irrel-
evant) sound was present, therefore the seemingly irrelevant
sound ultimately became useful for heavy multitaskers rather
than a liability (Lui and Wong 2012).
HMMs’ broader attentional filters and lesser ability to ig-
nore distractors may mean that they see more advertising even
when it is not relevant to their immediate goals. Hence, heavy
multitaskers’ inability to filter out irrelevant information could
be seen as a propensity to allow that any of the information
encountered could have utility, even if it is not relevant to the
current goal. This may lead to heavy multitaskers perceiving ads
as having the potential for utility even if the ads are not always
goal relevant:
H8: Perceived advertising utility will be positively related to media
multitasking.
METHOD
Considering the increasing prevalence of media multitasking
behavior, as well as its possible effects on individual media
experience, it is essential to understand predictors that increase
or decrease the likelihood that an individual engages in media
multitasking (Jeong and Fishbein 2007). We therefore attempted
to find potential antecedents of media multitasking, which may
play important roles in shaping multitasking behaviors.
Particularly, we examine previously validated predictors
(e.g., sensation seeking) as well as some novel predictors (ad
utility, simplicity, imagination, personal control) of media mul-
titasking. Because past research has largely used student sam-
ples, and students are from an age group that multitasks heavily
(Carrier et al. 2009), we wanted to first test our variables with
college students. We therefore examined our first regression
model using a student sample aged from 18 to 29. However,
we were also interested in seeing if the patterns that appear
for students were also found in a sample from a broader pop-
ulation. We therefore built the second model using a national
sample (ages 18 to 75) by conducting the same regression anal-
yses with a sample collected through Amazon’s Mechanical
Turk (MTurk). Simons and Chabris (2012) compared MTurk
and professional telephone polling to examine whether the gen-
eral public’s beliefs about memory deviated from those of the
memory experts. They found similar results between MTurk
and telephone polling data. Therefore, overall, the MTurk sur-
vey system is capable of producing patterns of results compara-
ble to those from a nationally representative telephone survey.
MTurk samples produce reliable results in line with standard
biases in decision making observed in other representative sam-
ples (Goodman, Cryder, and Cheema 2012). MTurk produces
diverse samples, and the data that are produced by MTurk sam-
ples have been found to be as reliable as data from traditional
polling methods (Buhrmester, Kwang, and Gosling 2011). In
sum, we explored possible predictors of media multitasking be-
havior by comparing results from a college student sample and
a nationwide sample.
Samples and Survey Procedure
Model 1: College student sample. Data for Model 1 were col-
lected from a survey carried out in an introductory media course
at a Midwestern university. This class is a large survey course
for nonmajors. The survey was programmed into Qualtrics on-
line survey software. The total number of observations for the
final regression analysis was N=308 (Ntotal =313, Nmissing =
5). Students received extra credit for participation.
Model 2: National sample. Raw data for Model 2 were col-
lected via MTurk. We posted a “hit” on MTurk named “Media
Multitasking Survey” which was described as “a short survey
about media multitasking.” The post contained a brief intro-
duction (‘Answer a survey on your personality traits and media
use”), instructions, required maximum time limit for comple-
tion (three hours), population qualification (United States only),
and remuneration ($0.75). If potential participants were inter-
ested, they clicked the survey link. Then they were forwarded
to the online survey generated through Qualtrics (Ntotal =513,
Nmissing =12).
Measurement
Demographic variables. Two demographic variables, age
and gender, were used in each model. Age was a contin-
uous variable; gender was coded as a dummy variable (fe-
male =0; male =1). See Table 1 for a summary of participant
demographics.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
16 B.R.-L. DUFF ET AL.
TABLE 1
Demographic Statistics for Model 1 (Student Sample) and
Model 2 (U.S. Population Sample)
Model 1 Model 2
Demographic (n=308) (n=501)
Age M=20.37 M=34.43,
SD =1.86 SD =12.58
Range =18 to 29 Range =18 to 75
Gender 129 males, 179
females
241 males, 256
females, 4
missing
Independent variables. The explanatory variables that we in-
cluded in our analysis were audience-level traits and preferences
that may be associated with multitasking propensity, including
executive control, creative mentality, complexity and ad utility.
The executive control variables were two variables that il-
lustrate individuals’ executive control: cognitive failures and
personal control. Perner and Lang (1999) defined executive
control as “processes responsible for higher-level action con-
trol that are necessary in particular for maintaining a mentally
specified goal and for bringing it to fruition against distracting
alternatives” (p. 337). Cognitive failures (Broadbent, Cooper,
and Parkes 1982; e.g., “get confused easily”) were measured
by 10 questionnaire items with a 5-point Likert scale (1 =Ve r y
inaccurate,5=Very accurate), and personal control (Heppner
and Petersen 1982; e.g., “I make snap judgments and later regret
them”) was assessed by five items with a five-point Likert scale
(1 =Very inaccurate to 5 =Very accurate). The CFQ compre-
hensively examines constructs of executive control proposed by
Pineda and colleagues (1998), including decision making (“dif-
ficulty in making up mind”; Broadbent, Cooper, and Parkes
1982, p. 8), control of attention (“lack of concentration,” p. 8),
and self-regulation (“disorganized,” p. 8). The personal control
questionnaire is a complement to the CFQ in that it examines
self-control of cognition and behavior necessary for success in
task performance.
For the noncognitive predictors, we looked at sensation seek-
ing, perceived creativity and imagination, and need for simplic-
ity. Jeong and Fishbein (2007) and Sanbonmatsu et al. (2013)
found a connection between sensation seeking and multitask-
ing using the Sensation Seeking Scale. We were interested in
validating this connection in both samples. Sensation seeking
was measured with the Brief Sensation Seeking Scale (BSSS;
Stephenson et al. 2003; e.g., “I like to do frightening things”),
utilizing a 5-point Likert scale (1 =Strongly disagree,5=
Strongly agree).
Creativity in individuals has been measured in many ways,
including a variety of scales for self-report. The big five person-
ality measures contain several facets that are relevant. Factor
V of the Abridged Big Five-Dimensional Circumplex Model
(AB5C; Hofstee, de Raad, and Goldberg 1992) is described as
a measure of “creative mentality” (Johnson 1994). This creative
mentality measure has been shown to be closely related to other
personality measures, such as openness to ideas and aesthetics
(Johnson 1994) and one’s ability to make novel connections and
question the status quo. Therefore, perceived creativity was as-
sessed by using the 10-item creativity facet (e.g., “Ask questions
that no one else does”) from the AB5C (Hofstee, de Raad, and
Goldberg 1992) with a 5-point Likert scale (1 =Very inaccurate,
5=Very accurate).
In the big five model, imagination is measured with its own
personality facet, but imagination together with creativity has
been found to factor into an overall creative mentality assess-
ment (Johnson 1994). Therefore, we assessed individual self-
reported imagination with 10 items from the AB5C (e.g., “Have
a vivid imagination”; Goldberg 1999) with a 5-point Likert scale
(1 =Very inaccurate,5=Very accurate).
The construct of need for simplicity (Liu, Smeesters, and
Trampe 2012, e.g., “It upsets me to go into a complicated situ-
ation”) was measured by four items with 9-point Likert scales
(1 =Strongly disagree,9=Strongly agree).
Last, for advertising utility, we used a single item to measure
perceived ad utility (“I feel that advertising is:”). This item is
similar to O’Keefe, Nash, and Liu (1981) about whether ad-
vertisements in a given medium were “very useful, somewhat
useful, or hardly useful at all.” For Model 1, ad utility was
assessed by a 7-point scale (1 =Never useful to 7 =Always
useful); for Model 2, it was assessed on a 5-point scale.
Dependent variable. The dependent variable in our study was
multitasking behavior. To collect our dependent responses, we
used a subjective measure of multitasking behavior, which was
assessed in Model 1 by asking our participants to report on a 7-
point Likert scale (1 =Never,7=Always) and a 5-point Likert
scale for Model 2. All variables were standardized as z-scores.
All of the participants were required to answer two questionnaire
items: “How often do you multitask in general? (e.g., talk to a
friend while watching TV)” and “How often do you use multiple
media at the same time? (e.g., use computer while watching
TV).” These items combined to form a multitasking assessment
measure (α>.75 in both models). See Table 2 for descriptive
statistics.
RESULTS
To verify the categorizations that we made in the literature
review and hypotheses, we conducted a principal component
analysis (PCA) to see if the variables could be categorized into
larger factors. The variables that we included in the PCA were
personal control, cognitive failure, imagination, creativity, need
for simplicity, sensation seeking, and ad utility. A varimax ro-
tation was applied to the PCA, along with Kaiser’s stopping
criterion (i.e., all factors with eigenvalues greater than 1 were
retained). See Table 3 for results of the PCA.
Three factors accounting for 70.32% of the variance were
extracted from the seven independent variables included.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
DOING IT ALL 17
TABLE 2
Principal Component Analysis: Rotated Component Matrix
Component
Variable Name 1 2 3
BSS .480 .599 −.137
COGFAIL .855 −.112 −.016
CREATIVITY −.357 .739 −.102
IMAGINE .058 .787 .194
NEEDSIMP .292 −.608 .274
PERSCTRL .873 −.097 .091
ADUTILITY .011 −.035 .956
Note. Extraction method: principal component analysis; rotation
method: varimax with Kaiser normalization. Rotation converged in
four iterations.
Personal control and cognitive failure loaded heavily onto the
first factor. Sensation seeking loaded more heavily onto a second
factor. Therefore, we named the first factor executive control.
Factor 2 included imagination, creativity, need for simplicity,
and sensation seeking. Those variables appear to reflect person-
ality/preferences. The third factor was ad utility, which loaded
heavily onto its own factor.
We predicted that the demographic, executive control,
personality/preferences, and advertising utility variables would
be associated with multitasking behavior. To test our research
questions and hypotheses, estimates derived from multiple re-
gression analysis were used. Table 4 presents the estimation
results derived from the different models for hypothesis testing.
Each column depicts the specific model in which demographics,
executive control, sensation seeking, creative mentality, need
for simplicity, and advertising utility are included as the sets
of regressors. To ensure that our models did not have multi-
collinearity problems, the variance inflation factor (VIF) was
also tested. The mean VIFs derived from Models 1 and 2 were
1.78 (Model 1; ranged from 1.03 to 3.57) and 1.45 (Model 2;
ranged from 1.07 to 1.92), respectively, which were lower than
the recommended level (VIF <10; see Dielman 2001). Thus,
we concluded that our models did not violate the stability of
the parameter estimates, suggesting that each model was a lin-
ear combination of independent measures. Our aim was to look
at patterns of results within each model as well as to see how
similar the results were between the models.
For Model 1 (students age 18 to 29, N=308), a regression
with the two demographic variables, age (hypothesis 1, β=
−.06, t=−1.06, p=n.s.) and gender (research question 1, β=
.03, t=.50, p=n.s.), revealed there were no significant effects
on media multitasking propensity. One cognitive factor, personal
control (hypothesis 3), was significantly and positively related to
multitasking (β=.17, t=2.48, p<.05). Because higher scores
indicate lower levels of control, this generally indicated that less
personal control contributes to increased media multitasking.
However, the other executive control measure, cognitive failures
(hypothesis 2), was not a significant predictor, β=−.10, t=
−1.45, p=n.s.
As hypothesized, sensation seeking (hypothesis 4; β=.16,
t=2.44, p<.05) was validated as a significant predictor of
multitasking in college students. Creativity provided marginal
evidence that it was also related to multitasking (hypothesis 5;
β=.11, t=1.68, p<.1). This indicated that higher sensation
seeking and greater creativity contributed to increased propen-
sity to engage in multitasking behavior. However, despite work
that shows imagination and creativity both contribute to overall
creative mentality, imagination was not significant (hypothesis
6; β=−.15, t=−1.47, p=n.s.).
We proposed a significant relationship between need for sim-
plicity and multitasking behavior; the result supported this as-
sertion, indicating the need for simplicity (hypothesis 7) was
significantly negatively related to multitasking behavior (β=
−.22, t=−2.19, p<.05), meaning that less desire for sim-
plicity (more need for complicated or complex) significantly
predicts multitasking propensity.
The perception of advertising utility was clearly associated
with multitasking behavior, indicating a positive relationship
TABLE 3
Descriptive Statistics Across Two Models
Model 1 Model 2
Variable Name M(SD) Cronbach’s alpha M(SD) Cronbach’s alpha
Cognitive failures 2.7(.62) .759 2.47 (.63) .814
Personal control 2.98 (.73) .701 2.7(.89) .845
Sensation seeking 3.4(.73) .713 2.67 (.97) .893
Creativity 3.5(.69) .789 3.55 (.64) .790
Imagination 3.32 (.64) .776 3.66 (.67) .776
Simplicity 2.63 (.86) .640 2.98 (.95) .811
Ad utility 5.56 (1.21) — 4.10 (1.33) —
MTask (DV) 5.15 (1.18) .776 5.06 (1.32) .835
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
18 B.R.-L. DUFF ET AL.
TABLE 4
Multiple Regression Models (Student Sample and U.S. National Sample)
Model 1: Student sample Model 2: National sample
Beta (SE) tBeta (SE) t
Age (H1) −.06 (.04) −1.06 −.25 (.003)∗∗∗ −5.34
Gender (RQ1) .03 (.14) .50 −.17 (.08)∗∗∗ −3.81
Cognitive failure (H2) −.10 (.13) −1.45 −.07 (.09) −1.18
Personal control (H3) .17 (.11)∗∗ 2.48 .06 (.06) 1.10
Sensation seeking (H4) .16 (.10)∗∗ 2.44 .16 (.05)∗∗∗ 3.11
Creativity (H5) .11 (12)∗1.68 .13 (.08)∗∗ 2.45
Imagination (H6) −.15 (.18) −1.47 −.04 (.07) −.80
Need for simplicity (H7) −.22 (.14)∗∗ −2.19 .03 (.03) .58
Perceived ad utility (H8) .14 (.05)∗∗ 2.50 .10 (.04)∗∗ 2.40
R2.09 .14
Note. All entries were standardized regression coefficients with standard errors in parentheses.
∗p<.10; ∗∗p<.05; ∗∗∗p<.01.
between perceived ad utility and multitasking behavior
(hypothesis 8; β=.14, t=2.50, p<.01).
As shown in Table 4, Model 2 (national sample, age 18 to 75,
N=501) results appear in in the second column. Unlike the stu-
dent sample, the demographic variables, age and gender, were
both significant determinants of multitasking behavior. Specif-
ically, age (hypothesis 1) played a critical role in determining
participants’ multitasking behaviors. The negative association
between age and multitasking behavior (β=−.25, t=−5.34,
p<.01) indicated that as age increases, the tendency to multi-
task drops. Multitasking behavior was also influenced by gender
(research question 1). Compared to males, females were more
prone to rate themselves as high media multitaskers (β=−.17,
t=−3.81, p<.01).
As in Model 1, sensation seeking was validated in Model 2 as
being significantly associated with multitasking behavior. The
greater the sensation-seeking tendencies (hypothesis 4; β=.16,
t=3.11, p<.01), the greater the likelihood of heavy multitask-
ing behavior. Also similar to Model 1, the greater the creativity,
the more likely one was to multitask (hypothesis 5; β=.13, t=
2.45, p<.05). In addition, a regression analysis once again con-
firmed that advertising utility was positively and significantly
related to multitasking propensity (β=.10, t=2.40, p<.01).
No other slopes representing the association between personal
traits and multitasking tendency were significant, such that cog-
nitive failures (hypothesis 2, β=−.07, t=−1.15, p=n.s.),
personal control (hypothesis 3, β=.06, t=1.10, p=n.s.),
imagination (hypothesis 6, β=−.04, t=−
.80, p=n.s.), and
need for simplicity (hypothesis 7, β=.03, t=.58, p=n.s.)
all failed to be predictors in Model 2.
In sum, the results across the two models are somewhat con-
sistent, with a few key differences. Specifically, sensation seek-
ing, creativity, and perception of advertising utility revealed
predictive effects on multitasking behavior across models. In
contrast, personal control and need for simplicity served as sig-
nificant predictors of multitasking behavior only for the student
sample. Moreover, the two demographic predictors of multi-
tasking, age and gender, were found to be significant only in the
national sample.
DISCUSSION
Changing media usage trends and preferences have led to
the need for advertising scholars and practitioners to understand
the who and why behind media multitasking. Understanding
what motivates adopters (audience-level preferences and traits)
can be key for helping to develop messages and tailor media to
them (e.g., Gangadharbatla 2008). In this study, we focused on
individual differences in preferences for multitasking with me-
dia, which has been a fast-growing behavior in the past several
years and is of increasing importance to advertisers. There are
already specific ad placement venues that would likely be con-
sumed more by heavy media multitaskers. For example, 90% of
tablet users do something else while using their tablets (Moses
2012). Certain TV content (e.g., reality shows, movies, awards
shows, sports) also attract more simultaneous media use (Ipsos
MediaCT and IAB 2012). In terms of media planning, Ad Age
(2011) reported that CBS created a six-part segmentation model
for audiences based on how they interact with media. Yet for all
of this, we know surprisingly little about who media multitasks
and why.
Heavy media multitaskers—those who self-reported higher
levels of media multitasking—have been found in past research
to perform more poorly than low media multitaskers on lab
tasks testing attention and focus (Ophir, Nass, and Wagner 2009;
Sanbonmatsu et al. 2013). However, it has been pointed out that
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
DOING IT ALL 19
poor performance on focusing tasks could indicate breadth of
processing and an ability to integrate seemingly irrelevant infor-
mation (e.g., Lui and Wong 2012). As noted by Cain and Mitroff
(2011), it is an open question whether performance differences
in heavier versus light media multitaskers are due to underlying
differences that cause people to seek out increased distraction
and stimulation or whether increased multitasking with media
causes the lesser ability to block out irrelevant information and
stimuli.
The data reported here are survey data and are correlational;
however, certain variables may point to underlying differences
(e.g., sensation seeking), and others may more likely be a by-
product or part of a feedback loop with media multitasking
(e.g., creativity and need for simplicity). Our results showed
that some variables were consistent predictors of media multi-
tasking in both the student sample and the national consumer
sample (sensation seeking, creativity, and advertising utility).
Some predictors were found only for the college student sample
(need for simplicity, personal control), and some were signifi-
cant only in the national sample (age, gender). We also found that
a few of the expected traits were unrelated to media multitasking
in either sample (cognitive failures, imagination). Understand-
ing why these variables may or may not predict multitasking
is important. Therefore, we will discuss each variable in more
detail.
Demographic Variables: Age and Gender
Not surprisingly, age had differing predictive power between
the samples. Age was significant only in the broad national
sample, probably because there was limited variability in age in
the student sample. Few scholarly media multitasking studies
have looked at nonstudent samples (e.g., Carrier et al. 2009;
Jeong and Fishbein 2007). Going forward, it will be important
to understand how cognitive differences in age, as well as gen-
erational differences in media availability, contribute to media
multitasking. It is also possible that the age effect on high and
low media multitaskers may not be stable; the differences may
simply be an artifact of a shift in generations that are becom-
ing more accustomed to splitting attention (Yap and Lim 2013).
Therefore, differences currently being observed between heavy
media multitaskers and light media multitaskers could poten-
tially be less evident in the coming years.
Gender was also a significant predictor only in the national
sample. This may reflect something that is generational (e.g.,
gender roles learned in different decades) or it may be a by-
product of lay theories and misestimations about how much one
multitasks. Some work has shown similar results, with adoles-
cent females reporting more media multitasking behavior than
males (Jeong and Fishbein 2007), but other multitasking stud-
ies have not found a gender difference (e.g., Ophir, Nass, and
Wagner 2009). Beyond self-report, future studies should look at
behavioral differences to see if the gender effect holds for more
objective measures of multitasking behavior or if it is simply a
difference in perceptions.
Executive Control Variables: Cognitive Failure and
Personal Control
One of the most surprising results was the lack of relationship
between cognitive failures and multitasking propensity in both
samples; personal control was a significant predictor only among
the college student sample. Because much of the research on me-
dia multitasking looks at college students, researchers should be
wary when using these variables if they wish to generalize be-
yond that population. In performance-based media multitasking
research, there are differences in basic cognitive abilities as
shown on attention task performance (Cain and Mitroff 2011;
Sanbonmatsu et al. 2013; Ophir, Nass, and Wagner 2009). How-
ever, while the distractibility subscale of CFQ correlates with
brain differences that cause distractibility in attentional tasks
(Kanai et al. 2011), it did not predict multitasking propensity.
To say whether the lack of findings on executive control
variables might be due to the heterogeneity from age range
in the national sample, we narrowed the national sample to
include only those in the same age range as the college students
(age 18 to 29; N=234) and ran the analysis again. Neither
personal control nor cognitive failure was significant in this
matched age group national sample. Therefore, predictors in
college students might not even be able to be projected to the
same age range in a non–college student population despite
their similar stages in cognitive development and generational
relationship with media.
The lack of predictive power for executive control should
raise concern, because the few studies that have been done to
link media multitasking and message outcomes have largely
taken a cognitive perspective, noting that competition for pro-
cessing resources may lead to lowered encoding of media con-
tent (Zhang, Jeong, and Fishbein 2010; Voorveld 2011). It may
be that we simply did not use the correct executive measures,
or it may also be that there are not as many cognitive trait dif-
ferences that contribute to propensity for media multitasking.
Based on these results, advertising researchers should consider
looking at media multitasking in terms of perceptual or sensory
attention and competition. For example, if someone is looking
at a social network feed on their phone while also watching a
sitcom on television, they may not be employing much cognitive
load. However, he or she must divide visual attention between
the screens, so perceptual differences may be more key than
executive control in performance in certain tasks.
Personality and Preferences Variables: Sensation Seeking,
Creativity and Imagination, and Need for Simplicity
The predictors that we found consistently across samples
were important in terms of implications for the future as well
as for providing confirmation and extension of past research.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
20 B.R.-L. DUFF ET AL.
Sensation seeking is one of the few individual predictors of
media multitasking that has been previously studied (Jeong and
Fishbein 2007; Sanbonmatsu et al. 2013). Our study provides
confirmation of the finding with college students and extends
the role of sensation seeking in multitasking with media to a
broader adult population. The finding that sensation seeking is
a significant positive predictor in the national sample (age 18
to 75) is especially important because sensation seeking tends
to generally decline with age (Roth, Hammelstein, and Brahler
2007).
Sensation seekers are important for message creators because
as an audience they are more likely attracted to arousing content
and experiences, and they have a higher likelihood of engaging
in risky behaviors (Zuckerman 1979). They would also likely
be interested in products such as motorcycles or experiential
products. Content creators should also note the need to match
the arousal level of message content to the needs of this high
sensation-seeking audience (Donohew, Lorch, and Palmgreen
2006). It remains to be seen, however, whether the added arousal
involved in simultaneous media consumption might mean that
content would not need to be more arousing because the media
consumption itself provides arousal.
Creativity was also found to be a significant predictor of mul-
titasking with media in both the college student and national
samples. One previous study did look at creativity (remote as-
sociates test) and multitasking performance but did not find an
association with multitasking performance (Morgan et al. 2011).
However, our measure of creativity was more akin to “creative
mentality” or personality than a test of state creativity. Our mea-
sure picks up on elements of the ability to connect novel ideas
and concepts. Because a primary role of advertising in build-
ing brand equity is to create brand associations (Krishnan and
Chakravarti 1993), future work may want to look at how heavy
multitaskers form associations between a brand and ad content.
Imagination was not significant in either sample. This finding
is important because it shows that the creativity scale was able
to predict multitasking on its own rather than combined with
imagination to make a broader creative mentality construct. It
seems that media multitaskers report that they are more likely
to have the ability to make connections between previously
unconnected existing concepts but are not any more likely to
report that they tend to think about entirely new concepts. This
may have important implications not only for brand associations
and ad memory but also for how media multitaskers relate to
content in ads (e.g., narratives, elaboration, metaphors) and how
they are affected by campaign ads in multiple media (e.g., Pilotta
and Schultz 2008).
While need for simplicity was significant only for college
students, this finding may be partially due to the measure used.
Our measure looked more at the preference for general complex-
ity and simplicity (e.g., “I am bothered by complicated things”);
however, it may be better for future work to look specifically
at preferences for perceptual complexity and simplicity, as per-
ceptual (versus conceptual) complexity may more directly relate
to media multitasking differences. In line with this, sensation
seeking was found to be a significant predictor in both samples;
higher sensation seeking has also been shown to be associated
with increased interest in visual complexity (Looft and Bara-
nowski 1971). An additional issue is that complicated and simple
may be relative terms to media multitaskers of different levels.
A more objective, behavioral measure may help shed light on
this variable’s relationship with multitasking in the future.
Advertising Utility
It has been shown that heavy media multitaskers are worse
at filtering irrelevant information. However, that broader atten-
tional filtering means they may also benefit on tasks where
seemingly irrelevant items end up becoming useful (Cain and
Mitroff 2011; Lui and Wong 2012). In line with this, adver-
tising utility was a consistent predictor of multitasking across
both samples. This has strong implications for advertisers. It has
been pointed out that the majority of consumers in the United
States do not find online advertising useful, and perceived ad
relevance has been used in the debate about behavioral tar-
geting of ads and consumer data collection and privacy issues
(Hoofnagle, Urban, and Li 2012). The finding that higher ten-
dency to multitask was associated with perceived advertising
utility can also be helpful for those concerned about what hap-
pens to advertising when it is exposed during multitasking and
therefore is competing even more for limited attention. It may
also speak to a recent industry observation that multitasking con-
sumers have better recall of a TV show’s advertising sponsors
than do people who were watching the show without secondary
media (Ipsos MediaCT and IAB 2012).
It is important to note that due to the nature of the data
reported here (survey data) we are not making claims about
causality. It is possible that increased perception of advertis-
ing utility is related to a trait-level propensity to broadly view
information and to consciously utilize more varied and goal-
irrelevant items—the same trait that would lead an individual
to be a heavier media multitasker. Conversely, the association
could stem from a post hoc rationalization, in that the lack of
ability to filter leads heavier multitaskers to let in more irrelevant
stimuli and they want to believe that they do so for a reason. Fu-
ture research should tease out the distinction between genuine
broad attentional interest in multiple media content and a lack
of ability to narrow attention.
Previous work has also found that increased exposure to
a medium increases perception of the utility of ads in that
medium (O’Keefe, Nash, and Liu 1981). While we looked at
self-reported multitasking, there is evidence that those who mul-
titask with media may also be consuming more media gen-
erally (Ipsos MediaCT and IAB 2012); although that same
line of reasoning would mean that over the decades, as me-
dia use/exposure has increased, advertising utility would have
also increased—something that seems to be opposite to trends in
public opinion. Nonetheless, it should be explored with further
research.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
DOING IT ALL 21
Advertising utility is a novel variable to connect with me-
dia multitasking, and one that could potentially be important
for advertisers to understand. We used a one-item measure in
our studies, similar to one used by O’Keefe, Nash, and Liu
(1981). While this single-item measure was our only measure
in our student sample, in the national sample we also included
an ad utility measure adopted from Schlosser, Shavitt, and
Kanfer (1999). Advertising utility was measured in their study
by asking respondents if advertising is informative, entertaining,
and useful for making purchase decisions. This ad utility mea-
sure explained 43% of the variance in attitudes toward adver-
tising in their sample (Schlosser et al. 1999). This four-item ad
utility measure was highly correlated with our single-item utility
measure (r(501) =.60; p<.001. In addition, the four-item ad
utility scale was positively correlated with age (r(501) =.11,
p<.05. Upon further examination, it appeared that age acted
as a suppressor for ad utility in the national sample. With age in
the model, ad utility is a significant predictor of multitasking be-
havior; without age, ad utility becomes insignificant. However,
when the sample is segregated by tertiary split (i.e., separated by
the ages at 33rd and 67th percentile), ad utility is a significant
predictor of multitasking behavior in the younger age group,
which resembles the result in the student sample. Those results
are most likely because the student sample was composed of a
relatively homogenous and younger age group. Therefore, future
work looking at perceptions of advertising utility should con-
sider whether variations in age matter for the construct at hand.
We acknowledge that self-report has major flaws. People
may not always consciously realize when they are multitasking
or assess whether they have the ability to multitask. Instead,
people often have lay theories about their own abilities and
attention, particularly when that attention is divided (Simons
and Chabris 2011). People also generally underestimate how
often they switch between media (Brasel and Gips 2011). We did
see variation in reported multiasking, but it may be that certain
people are more predisposed to underestimations than others.
Media multitasking predictors found in this study should be
further vetted with additional samples and in controlled studies.
A major limitation in our study is the measure of multi-
tasking. Unfortunately, there is no current standard on how to
assess levels of media multitasking. Some studies have asked
participants to subjectively estimate how much they multitask
or engage in specific multitasking behaviors (Zhong, Hardin,
and Sun 2011). Ophir, Nass, and Wagner (2009) used a mea-
sure of media multitasking (MMI) that asked students to report
how often (Most/Some/A little of the time/Never) they spent us-
ing every possible combination of 12 media; others have also
adopted this measure (Sanbonmatsu et al. 2013). However, if
heavy multitaskers are less able to focus on tasks, this measure
could be problematic, as the MMI takes 20 minutes to com-
plete on its own. Jeong and Fishbein (2007) used a measure
asking about “how often” respondents used 13 different spe-
cific combinations of media. However, that brings up the issue
of subjectivity in rating “how often” one multitasks with two
media. In our national sample we therefore adapted a measure
similar to a shortened MMI in which we asked participants to
rate for each medium the number of hours that they multitasked
with other media each week. This more “objective” measure
of media multitasking was highly correlated with our two-item
self-report measure of multitasking, r(501) =.54, p<.001.
Therefore, while our measure is not perfect, we do feel that it
captured some of the variance in media multitasking behavior.
Compared to sequential presentation, simultaneous presen-
tation leads to changes in commercial argument cue effects
(Chowdhury, Finn, and Olsen 2007). Comparison between sin-
gle task and multiple tasks has also been used to show that
intrusive product placements may not have typical negative ef-
fects (Yoon, Choi, and Song 2011). It would be interesting to
see if some of the individual predictors found in this study affect
outcomes such as those. Because there were differences in dis-
tractor filtering ability, sensation seeking, creativity, desire for
complexity, and ad utility, it would seem likely that there would
be additional differences found that may moderate the effects in
experimental studies.
The goal of this research was to begin a preliminary ex-
amination of individual differences in preferences and process-
ing style and propensity to multitask. Typically, media research
has focused on effects models built around the viewing of a
medium/message as an isolated object of direct attention, but
media practitioners and scholars are realizing that media con-
sumption and exposure is changing. The individual differences
that we found in this study provide a starting point to help
untangle advertising and media effects in a multimedia world.
REFERENCES
Ad Age (2011), “CBS: Viewers Age and Sex Shouldn’t Matter to Mar-
keters,” March 23, http://adage.com/article/mediaworks/cbs-viewers-age-
sex-matter-marketers/149534/.
Amabile, Teresa M. (2001), “Beyond Talent: John Irving and the Passionate
Craft of Creativity,” American Psychologist, 56 (4), 333–36.
Berns, Gregory (2008), “Neuroscience Sheds New Light on Creativ-
ity,” Fast Company, September, http://www.fastcompany.com/magazine/
129/rewiring-the-creative-mind.html.
Brasel, S. Adam, and James Gips (2011), “Media Multitasking Behavior: Con-
current Television and Computer Usage,” Cyberpsychology, Behavior, and
Social Networking, 14 (9), 527–34.
Broadbent, Donald E., P. Fitzgerald Cooper, and Katharine R. Parkes (1982),
“The Cognitive Failures Questionnaire (CFQ) and Its Correlates,” British
Journal of Clinical Psychology, 21 (1), 1–16.
Buhrmester, M., T. Kwang, and S.D. Gosling (2011), “Amazon’s Mechanical
Turk: A New Source of Inexpensive, Yet High-Quality, Data?” Perspectives
on Psychological Science, 6 (1), 3–5.
Cain, Matthew S., and Stephen R. Mitroff (2011), “Distractor Filtering in Media
Multitaskers,” Perception, 40, 1183–92.
Calder, Bobby J., Lynn W. Phillips, and Alice M. Tybout (1981), “Designing
Research for Application,” Journal of Consumer Research, 197–207.
Carrier, L. Mark, Nancy A. Cheever, Larry D. Rosen, Sandra Benitez, and
Jennifer Chang (2009), “Multitasking across Generations: Multitasking
Choices and Difficulty Ratings in Three Generations of Americans,” Com-
puters in Human Behavior, 25 (2), 483–89.
Carson, Shelley H., Jordan B. Peterson, and Daniel M. Higgins (2003),
“Decreased Latent Inhibition Is Associated with Increased Creative
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
22 B.R.-L. DUFF ET AL.
Achievement in High-Functioning Individuals,” Journal of Personality and
Social Psychology, 85 (3), 499–506.
Chowdhury, Rafi, Adam Finn, and G. Douglas Olsen (2007), “Investigating the
Simultaneous Presentation of Advertising and Television Programming,”
Journal of Advertising, 36 (3), 85–96.
Clapp, Wesley C., Michael T. Rubens, Jasdeep Sabharwal, and Adam Gazzaley
(2011), “Deficit in Switching between Functional Brain Networks Underlies
the Impact of Multitasking on Working Memory in Older Adults,” Proceed-
ings of the National Academy of Sciences, 108 (17), 7212–17.
Craik, Fergus I.M., Richard Govoni, Moshe Naveh-Benjamin, and Nicole D.
Anderson (1996), “The Effects of Divided Attention on Encoding and Re-
trieval Processes in Human Memory,” Journal of Experimental Psychology:
General, 125 (2), 159–80.
Dielman, T.E. (2001), Applied Regression Analysis for Business and Economics,
Pacific Grove, CA: Duxbury Thomson Learning.
Donohew, Lewis, Elizabeth Pugles Lorch, and Philip Palmgreen (2006), “Ap-
plications of a Theoretic Model of Information Exposure to Health Inter-
ventions,” Human Communication Research, 24 (3), 454–68.
Duff, Brittany R.L., and Ronald J. Faber (2011), “Missing the Mark: Advertis-
ing Avoidance and Distractor Devaluation,” Journal of Advertising,40(2),
51–62.
Eysenck, Hans J. (1995), Genius: The Natural History of Creativity,NewYork:
Cambridge University Press.
Gangadharbatla, Harsha (2008), “Facebook Me: Collective Self-Esteem, Need
to Belong, and Internet Self-Efficacy as Predictors of the iGeneration’s At-
titudes toward Social Networking Sites,” Journal of Interactive Advertising,
8 (2), http://jiad.org/article100.html.
Goldberg, L.R. (1999), “A Broad-Bandwidth, Public Domain, Personality In-
ventory Measuring the Lower-Level Facets of Several Five-Factor Mod-
els,” in Psychology in Europe, I. Mervielde, I. Deary, F. De Fruyt, and F.
Ostendorf, eds., vol. 7, Personality, Tilburg, the Netherlands: Tilburg Uni-
versity Press, 7–28.
Goodman, Joseph K., Cynthia E. Cryder, and Amar Cheema (2013), “Data
Collection in a Flat World: The Strengths and Weaknesses of Mechanical
Turk samples,” Journal of Behavioral Decision Making, 26 (3), 213–24.
Heppner, P. Paul, and Chris H. Petersen (1982), “The Development and Impli-
cations of a Personal Problem-Solving Inventory,” Journal of Counseling
Psychology, 29 (1), 66–75.
Hofstee, Willem K., Boele de Raad, and Lewis R. Goldberg (1992), “Integration
of the Big Five and Circumplex Approaches to Trait Structure,” Journal of
Personality and Social Psychology, 63 (1), 146–63.
Hoofnagle, Chris Jay, Jennifer M. Urban, and Su Li (2012), “Privacy and Mod-
ern Advertising: Most U.S. Internet Users Want ‘Do Not Track’ to Stop
Collection of Data about Their Online Activities,” presented at the Amster-
dam Privacy Conference, Amsterdam, the Netherlands, October.
Ipsos MediaCT and Internet Advertising Bureau (2012), “Screens(n) What Are
People Doing. .. And Why?” http://www.iab.net/media/file/Simultaneous-
screen-IAB-Innovation-Day-v19.pdf.
Jeong, Se-Hoon, and Martin Fishbein (2007), “Predictors of Multitasking with
Media: Media Factors and Audience Factors,” Media Psychology,10(3),
364–84.
———, and Yoori Hwang (2012), “Does Multitasking Increase or Decrease Per-
suasion? Effects of Multitasking on Comprehension and Counterarguing,”
Journal of Communication, 62, 571–87.
Johnson, J.A. (1994), “Clarification of Factor Five with the Help of the AB5C
Model,” European Journal of Personality, 8, 311–34.
Kahneman, Daniel (1973), Attention and Effort, Englewood Cliffs, NJ: Prentice-
Hall.
Kanai, Ryota, Mia Yuan Dong, Bahador Bahrami, and Geraint Rees (2011),
“Distractibility in Daily Life Is Reflected in the Structure and Function of
Human Parietal Cortex,” Journal of Neuroscience, 31 (18), 6620–26.
Konig, Cornelius J., Markus Buhner, and Gesine Murling (2005), “Working
Memory, Fluid Intelligence, and Attention Are Predictors of Multitasking
Performance, but Polychronicity and Extraversion Are Not,” Human Per-
formance, 18 (3), 243–66.
Krishnan, Shankar H., and Dipankar Chakravarti (1993), “Varieties of Brand
Memory Induced by Advertising: Determinants, Measures, and Relation-
ships,” In D. A. Aaker & A. L. Biel (Eds.) Brand equity and advertising:
Advertising’s role in building strong brands (Vol. 1), Psychology Press.
Lin, Lin (2009), “Breadth-Biased Versus Focused Cognitive Control in Media
Multitasking Behaviors,” Proceedings of the National Academy of Sciences,
106 (37), 15521–22.
Liu, Jia, Dirk Smeesters, and Debra Trampe (2012), “Effects of Messiness on
Preferences for Simplicity,” Journal of Consumer Research,39 (1), 199–214.
Lui, Kelvin F. H., and Alan C.-N. Wong (2012), “Does Media Multitasking Al-
ways Hurt? A Positive Correlation Between Multitasking and Multisensory
Integration,” Psychonomic Bulletin & Review, 19 (4), 647–53.
Looft, William R., and Marc D. Baranowski (1971), “An Analysis of Five
Measures of Sensation Seeking and Preference for Complexity,” Journal of
General Psychology, 85 (2), 307–13.
Monsell, S. (2003), “Task Switching,” Trends in Cognitive Sciences, 7 (3),
134–40.
Morgan, Brent, Sidney D’Mello, Karl Fike, Robert Abbott, Michael Haass,
Andrea Tamplin,Gabriel Radvansky, and Chris Forsythe (2011), “Individual
Differences in Multitasking Ability and Adaptability,” Proceedings of the
Human Factors and Ergonomics Society Annual Meeting, 55 (1), 919–
23.
Moses, Lucia (2010), “Media Multitaskers Abound,” Adweek, September
8, http://www.adweek.com/news/television/media-multitaskers-abound-
103228.
——— (2012), “Data Points: Two-Screen Viewing,” Adweek, Novem-
ber 7, http://www.adweek.com/news/technology/data-points-two-screen-
viewing-145014.
Nielsen Cross Platform Report (2012), “The Cross-Platform Commu-
nity: A New Connected Community,” November 13, Newswire,
http://www.nielsen.com/us/en/newswire/2012/the-cross-platform-report-a-
new-connected-community.html.
New Zealand Broadcasting Standards Authority (2007), Children’s Media Use
and Responses: A Review of the Literature, Wellington, New Zealand:
Broadcasting Standards Authority.
O’Keefe, Garrett J., Kathaleen Nass, and Jenny Liu (1981), “The Perceived
Utility of Advertising,” Journalism Quarterly, 58 (4), 535–42.
Ophir, Eyal, Clifford Nass, and Anthony D. Wagner (2009), “Cognitive Control
in Media Multitaskers,” Proceedings of the National Academy of Sciences,
106 (37), 15583–87.
Pashler, Harold (1994), “Dual-Task Interference in Simple Tasks: Data and
Theory,” Psychological Bulletin, 116 (2), 220–44.
Perner, Josef, and Birgit Lang (1999), “Development of Theory of Mind and
Executive Control,” Trends in Cognitive Sciences, 3 (9), 337–44.
Petty, Richard E., Gary L. Wells, and Timothy C. Brock (1976), “Distraction
can Enhance or Reduce Yielding to Propaganda: Thought Disruption Versus
Effort Justification,” Journal of Personality and Social Psychology, 34(5),
874.
Pilotta, J., and Schultz, D. (2005), “Simultaneous Media Experience and Synes-
thesia,” Journal of Advertising Research, 45 (1), 19–26.
Pineda, David, Alfredo Ardila, Mo’Nica Rosselli, Clemencia Cadavid, Silvia
Mancheno, and Silvia Mejia (1998), “Executive Dysfunctions in Children
with Attention Deficit Hyperactivity Disorder,” International Journal of
Neuroscience, 96 (3–4), 177–96.
Poposki, Elizabeth, Frederick Oswald, and Hubert Chen (2009), “Neuroticism
Negatively Affects Multitasking Performance through State Anxiety,” Navy
Personnel Research, Studies, and Technology, 9(3).
Prakash, Ruchika Shaurya, Kirk I. Erickson, Stanley J. Colcombe, Jennifer S.
Kim, Michelle W. Voss, and Arthur F. Kramer (2009), “Age-Related Differ-
ences in the Involvement of the Prefrontal Cortex in Attentional Control,”
Brain and Cognition, 71 (3), 328–35.
Reid, Len N., and Herbert J. Rotfeld (1976), “Toward an Associative Model of
Advertising Creativity,” Journal of Advertising, 5 (4), 24–29.
Roberts, Donald F., and Ulla G. Foehr (2008), “Trends in Media Use,” Future
of Children, 18 (1), 11–37.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
DOING IT ALL 23
Roth, Marcus, Phillip Hammelstein, and Elmar Brahler (2007), “Beyond a
Youthful Behavior Style: Age and Sex Differences in Sensation Seek-
ing Based on Need Theory,” Personality and Individual Differences, 43,
1839–50.
Sanbonmatsu, David M., David Strayer, Nathan Medeiros-Ward, and Jason
Watson (2013), “Who Multi-Tasks and Why? Multi-Tasking Ability, Per-
ceived Multi-Tasking Ability, Impulsivity, and Sensation Seeking,” PloS
ONE, 8 (1), e54402.
Schlosser, Ann, Sharon Shavitt and Alaina Kanfer (1999), “Survey of Inter-
net Users’ Attitudes Towards Internet Advertising,” Journal of Interactive
Marketing, 13 (3), 34–54.
Simons, Daniel J., and Christopher Chabris (2011), “What People Believe about
How Memory Works: A Representative Survey of the U.S. Population,”
PLoS ONE, 6, (8), e22757.
———, and ——— (2012), “Common (Mis) Beliefs about Memory: A Replica-
tion and Comparison of Telephone and Mechanical Turk Survey Methods,”
PLoS ONE, 7 (12), e51876.
Stephenson, M., Rick H. Hoyle, Philip Palmgreen, and Michael D. Slater (2003),
“Brief Measures of Sensation Seeking for Screening and Largescale Sur-
veys,” Drug and Alcohol Dependence, 72 (3), 279–86.
Strayer, David, and Jason Watson (2012), “Supertaskers and the Multitasking
Brain,” Scientific American Mind, 23 (1), 22–9.
Voorveld, Hilde (2011), “Media Multitasking and the Effectiveness of Combin-
ing Online and Radio Advertising,” Computers in Human Behavior,27(6),
2200–06.
Wang, Zheng, and Tchernev (2012), “The ‘myth’ of media multitasking: Recip-
rocal dynamics of media multitasking, personal needs, and gratifications,”
Journal of Communication, 62, 493–513.
Yap, Jit Yong, and Stephen Wee Hun Lim (2013), “Splitting Visual Focal At-
tention? It Probably Depends on Who You Are,” Proceedings of the 2nd
Annual International conference on Cognitive and Behavioral Psychology,
Singapore, February.
Yoon, Sukki, Yung Kyun Choi, and Sujin Song (2011), “When Intrusive Can
Be Likable,” Journal of Advertising, 40 (2), 63–76.
Zhang, Weiyu,SeeHoon Jeong, and Martin Fishbein (2010), “Situational Factors
Competing for Attention: The Interaction Effect between Multitasking and
Sexual Explicitness on TV Recognition,” Journal of Media Psychology,22
(1), 2–13.
Zhong, Bu, Marie Hardin, and Tao Sun (2011), “Less Effortful Thinking Leads
to More Social Networking? The Associations between the Use of Social
Network Sites and Personality Traits,” Computers in Human Interaction,
27, 1265–71.
Zuckerman, Marvin (1979), Sensation Seeking: Beyond the Optimal Level of
Arousal, Hillsdale, NJ: Erlbaum.
Downloaded by [University of Illinois at Urbana-Champaign] at 15:04 03 September 2014
