Computers in Human Behavior 44 (2015) 148–155

Contents lists available at ScienceDirect

Computers in Human Behavior

journal homepage: www.elsevier.com/locate/comphumbeh

Too much Facebook? An exploratory examination of social media

fatigue
Laura F. Bright a,⇑, Susan Bardi Kleiser b,1, Stacy Landreth Grau b,2
a
School of Strategic Communication, Bob Schieffer College of Communication, Texas Christian University, TCU Box 298065, Fort Worth, TX 76129, United States
b
Department of Marketing, Neeley School of Business, Texas Christian University, TCU Box 298530, Fort Worth, TX 76129, United States

a r t i c l e i n f o a b s t r a c t

Article history: Social media usage levels continue to climb generating copious amounts of content. As more people
crowd social media (e.g. Facebook), and create content, some research points to the existence of a concept
called social media fatigue. Social media fatigue is defined as a user’s tendency to back away from social
Keywords: media participation when s/he becomes overwhelmed with information. Lang’s (2000) limited capacity
Social media model is used to understand the role of information overload for social media fatigue. This research
Consumer psychology examines the concept of social media fatigue and its proposed antecedents: social media efficacy, help-
Media effects
fulness, confidence and privacy concerns. Using confirmatory regression, this research determined that
Social media fatigue
Human–computer interaction
privacy concerns and confidence have the greatest predictive value for social media fatigue. This paper
Survey of social media usage has theoretical implications for not only LCM but also other technology acceptance models such as
TAM and UTAUT and UTAUT2. It also has implications for those trying to engage with online audiences
and their subsequent reactions to that attempt at engagement. Several future research ideas are explored
as well.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction inevitably this creates a great deal of content online. This content
can be in the form of wall posts, messages, tweets about deals,
Social media has changed the nature of communication – and constant status updates all with varying degrees of quality.
whether peer-to-peer, business to consumer or business-to-busi- But while Facebook has grown in popularity over the years, it
ness. Social networking sites like Facebook and micro-blogging appears it may have reached its peak in terms of usage. There
sites like Twitter have become some of the primary sources of appears to be a usage-related phenomenon setting in among users
communication for people to find out about people in their net- of Facebook and other social media websites, which claims users
work and brands with which they identify. Facebook has become are visiting social media sites less. This has been coined ‘‘social
the world’s largest social networking site with more than 802 mil- media fatigue’’ (Goasduff & Pettey, 2011). One explanation for this
lion active daily users on average for March 2014, a 21% year-over- phenomenon is that users are experiencing information overload.
year increase (Facebook., 2014). Similarly, Twitter has 100 million With Facebook and Twitter boasting their highest numbers ever
active user accounts and 1 billion tweets sent every seven days (Facebook, 2014; Sullivan, 2012) more people are using social
(Sullivan, 2011). Overall media usage has become fractured with media than ever before. Moreover, according to ‘‘Zuckerberg’s
people spending 24% of their time online (second only to television Law’’, the volume of status updates, photos, and other online mate-
at 40%). Part of this is due to the rise of polychronic media con- rial each individual posts on Facebook doubles every year
sumption whereby people consume multiple media types simulta- (Bradshaw, 2011). The Limited Capacity Model (LCM) (Lang, 2000)
neously, i.e. watching television while looking up information or suggests that people have a limited amount of mental resources
engaging with online friends via a laptop or smart phone (Brasel to process information. Whether it is grappling with the latest Face-
& Gips, 2011). With so many people on social networking sites book changes or dealing with a backlog of tweets, social media
and more consumers choosing to adopt a social media presence, users are flooded with information. Thus, LCM can be used to help
explain social media fatigue. In addition, several models rooted in
⇑ Corresponding author. Tel.: +1 817 257 5251. technology acceptance will be used to help explain this phenome-
E-mail addresses: [email protected] (L.F. Bright), [email protected] (S.B. Kleiser),
non including the Technology Acceptance Model (TAM) (Davis,
[email protected] (S.L. Grau). 1989) and the Unified Theory of Acceptance and Use of Technology
1
Tel.: +1 817 257 5485. (UTAUT) (Venkatesh, Morris, Davis, & Davis, 2003).
2
Tel.: +1 817 257 6540.

http://dx.doi.org/10.1016/j.chb.2014.11.048
0747-5632/Ó 2014 Elsevier Ltd. All rights reserved.
 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155 149

1.1. The social media landscape & Gladwell, 2013). In addition, the social compensation hypothesis
has been asserted as an explanation for increased usage among
Social media has many definitions. Many of these definitions those with lower self-esteem (e.g., Lee, Moore, Park, & Park,
come from bloggers and pundits – self-described experts in the 2012, p. 1037). Furthermore, McCord, Rodebaugh, and Levinson
field. Managers have struggled with not only how to benefit from (2014) find people with high anxiety on Facebook as well as high
social media strategies but even how to classify social media that social anxiety engage in more frequent social Facebook usage than
is meaningful to their goals. Kaplan and Haenlein (2010) define it those with high Facebook anxiety but low social anxiety. Neverthe-
as ‘‘a group of Internet-based applications that build on the ideo- less, what are the ramifications of this increased usage? Stress in
logical and technological foundations of Web 2.0 and that allow the workplace has been one consequence explored (Bucher,
the creation and exchange of user-generated content’’ (p. 61). In Fieseler, & Suphan, 2013) and social media fatigue is the proposed
general, this includes many technologies – from blogs to Facebook outcome investigated in this research.
profile pages; from social media games to Wikipedia. In order to
classify social media, Kaplan and Haenlein (2010) relied on a set
1.2. Social media fatigue
of theories, including social presence, media richness, self-presen-
tation and self-disclosure to serve as a theoretical foundation for a
Social media fatigue is defined as ‘‘social media users’ tendency
typology. As a result, they were able to classify six types of technol-
to back away from social media usage when they become over-
ogy according to the level of social presence/media richness (low,
whelmed with too many sites, too many pieces of content, too
medium and high) and self-presentation/self-disclosure (low and
many friends and contacts and too much time spent keeping up
high). The six technologies include: blogs, social networking sites
with these connections’’ (Technopedia., 2011). Social media fatigue
(e.g. Facebook), virtual social worlds (e.g. Second Life), collabora-
can also be linked to concerns about privacy and boredom among
tive projects (e.g. Wikipedia), content communities (e.g. YouTube)
social media users. According to market researcher Gartner, the
and virtual game worlds (e.g. World of Warcraft). Social network-
social media market is beginning to show signs of maturity as
ing sites, which is what most people think of as social media, is
some users in certain segments show signs of social media fatigue.
considered high on self-presentation/self-disclosure, defined as
The company surveyed 6295 users between the ages 13 and 74 in
any ‘‘type of social interaction people have the desire to control
11 developed and developing markets from December 2010 and
the impressions other people form of them’’ (Kaplan & Haenlein,
January 2011 (Goasduff & Pettey, 2011). Respondents were asked
2010, p. 62). It is classified as medium on social presence/media
about their use of and opinions of social media. Of those surveyed,
richness, defined as ‘‘the acoustic, visual, and physical contact that
24% said they were using their favorite social media site less than
can be achieved – that allows to emerge between two communica-
when they first signed up. Those who admitted to using social
tion partners’’ (Kaplan & Haenlein, 2010, p. 61). Given the number
media less were found to be in segments with a more practical
of people using social networking sites and that most people asso-
view of technology (Goasduff & Pettey, 2011). ‘‘The trend shows
ciate social media primarily with social networking sites, we have
some social media fatigue among early adopters, and the fact that
chosen to concentrate on this technology type. Therefore, the focus
31 percent of Aspirers [younger, more mobile, brand-conscious
of our research is on users of social networking sites primarily
consumers] indicated that they were getting bored with their
Facebook and Twitter. Table 1 outlines three of the major platforms
social network is a situation that social media providers should
as of January 2012 – Facebook, Twitter and Google+.
monitor, as they will need to innovate and diversify to keep con-
Indeed, social media usage rates continue to rise. The amount of
sumer attention’’, said Brian Blau, researcher at Gartner
time spent on social media and the amount of content encountered
(Goasduff & Pettey, 2011, p. 1). When asked why, interestingly,
on social media seems to be triggers for social media fatigue. For
the responses were mixed with online privacy concerns ranking
example, social networking accounts for 22% of all of the time
highest. However, teenagers and those in their twenties reported
spent online in the U.S. and global consumers spend an average
that they were significantly more likely to say that they had
of more than 6 h per week on social networking sites (Nielsen,
increased their usage. In fact, 37% of those in younger age groups
2011). People spend more than 700 billion minutes monthly on
and more tech-savvy segments said they used their favorite social
Facebook alone sharing more than 30 billion pieces of content
media site more (Goasduff & Pettey, 2011).
monthly (www.jeffbullas.com). Twitter adds more than half a mil-
These results were supported by yet another study indicating
lion users daily and handles 1.6 billion queries daily. Other sites
social media fatigue. Research firm Trendstream reported that
like YouTube generate more than 92 billion page views monthly
active behavior on Facebook has declined since July 2009, particu-
(www.jeffbullas.com). Time spent on social networking sites
larly in established markets like the United States
accounts for one in every 6 min spent online (Lipsman, 2011).
(GlobalWebIndex., 2011). Trendstream defines ‘‘active behavior’’
Research has investigated reasons for this increased usage of
as posting status updates, sharing content, messaging friends,
social media. For example, the ‘‘fear of missing out’’ contributes
and installing applications. The study revealed this trend is more
to more Facebook engagement (Przybylski, Murayama, DeHann,

Table 1
Social media platforms: the big three (2012).

Technology Usage information Engagement Advertising Implications for SMF

Facebook 850 million users; half of which ‘‘Like’’ associates user with a brand or Facebook ads: New layout caused many people to become angry;
2004 visit daily. Three quarters of the comment; this endorsement is then sponsored ads on The Ticker is considered distracting because it is
users are outside of the US broadcast in the newsfeed the side panels scrolling in real time (Gross, 2011)
Twitter 100 million monthly active users Retweet Promoted tweets; Interface is simple; but the sheer number of people
2007 who visit daily; more than 300 exploring rich to follow can be overwhelming
tweets since daily media ad formats
Google+ 90 million users in six months (20 Circles may take too much work from the user
2011 million in two weeks)
150 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155

visible in U.S. college educated users in their twenties, Facebook’s past (Barger, 2011). He argues it is not due to any inherent weakness
original users (GlobalWebIndex, 2011). Indeed, technology leaders in many of these platforms, but rather the saturation point that
are beginning to recognize that users have reached a point where many people feel with social networking. He argues that people
there is simply no more time. ‘‘We are definitely at a point where are spending as much time online as they can and the pressure of
the supply of things to do online is at a dramatic overcapacity rel- maintaining a presence on yet another platform is too burdensome
ative to what we can actually do in our lives as human beings. (Barger, 2011). The few new platforms that have gained popularity –
When you have technology that has an infinite supply of possibility namely Snapchat – have been adopted primarily by younger audi-
and you have finite time, at a certain point, people shut down. They ences. Snapchat, launched in late 2011 has more than 100 million
simply cannot handle anything else’’, said Ning co-founder, Gina active monthly users, 70% of which are women. Moreover, more
Bianchini (Boskers, 2011). This trend is indeed supported by the than 70% of users are under 25 years old (with 32% between the ages
LCM, which dictates that information consumers must make com- of 13 and 17). The active usage is likely due to its simple platform.
promises when it comes to their attention given its limited capac-
ity to process information.
One stream of research is dealing with the ways that this new 1.3. A theoretical framework to examine social media fatigue
media environment may be compromising learning and attention.
Research shows individuals pay less attention to messages when When it comes to examining the concept of social media fati-
faced with distractions (Lang, 2000) and retain less information gue, several models provide appropriate constructs to better
overall. Other studies have shown that recall of information can understand the underlying components that could lead to this
be inhibited (Drolet & Luce, 2004; Nasco & Bruner, 2009). Biocca, state of mind. Among them, the most applicable are TAM (Davis,
Owen, Tang, and Bohil (2007) have found that overwhelming 1989), UTAUT (Venkatesh et al., 2003) and LCM (Lang, 2000). Each
amounts of information makes it harder to comprehend complex of these models provides a framework for examining social media
messages and Palfrey and Gasser (2008) found that information fatigue from a variety of angles including perceived usefulness,
overload can cause people to use sensory filters to cope with the perceived ease of use, self-efficacy, information processing, and
sheer amount of information, making it impossible to pay attention overall technology acceptance.
to most messages (Hill & Moran, 2011). Furthermore, psychologists In his research to better understand technology acceptance
have linked text messaging and instant messaging with self- through the lens of predicting use, Davis (1989) established two
reported symptoms of depression and social anxiety among college constructs relevant to social media fatigue – (1) perceived ease of
undergraduates (Becker, Alzahabi, & Hopwood, 2013). Yet, none of use (PEU), and (2) perceived usefulness (PU). PEU is conceptualized
this research has dealt specifically with social media as a context. as ‘‘the degree to which a person believes that using a particular
Social media fatigue can come from a variety of sources. It can system would be free of effort’’ (p. 320). In the case of social media
come when old friends find users on Facebook and begin sending fatigue, PEU could be high for a given social media website if a per-
friend requests, posts about social games (i.e. FarmVille, Mafia son’s friends and family are all on the website; however, PEU could
Wars, CityVille), what they ate for breakfast and photos from their decrease if the consumer feels overwhelmed by privacy settings,
iPhone. It can come with the way that social networks are pro- friend requests, etc. Related to Bandura’s (1982) work on self-effi-
grammed. For example, when Facebook made changes to its inter- cacy, PEU is derived from a person’s perceptions of how well they
face, many people found that the new features were overwhelming would be able to do a certain task or deal with a given situation – in
and adaptation was difficult. For example, instead of the old way of this case, operating an account on one or several social media web-
defaulting to the most recent story, the new interface featured sites. Davis (1989) further defines PU as ‘‘the degree to which a
‘‘Top Stories’’ which combines factors such as friends the user person believes that using a particular system would enhance his
interacts with the most and the stories with the most ‘‘likes’’ or her job performance’’ (p. 320). As it relates to social media fati-
(Gross, 2011). So if a user feels that the ‘‘Top Stories’’ are incorrect, gue, PU might be conceptualized in terms of a person maintaining
it would take some time to scroll through hundreds of posts to find his or her own appearance through social media as a ‘‘job’’ – i.e.
more relevant information. Additionally, the new ‘‘Ticker’’ is a real how useful is it for me to maintain social media accounts? PEU
time scrolling update of information (similar to Twitter) and critics and PU could both potentially impact the level of fatigue one
have found it distracting and overwhelming if something is missed may experience on social media websites and, as such, will be
(Gross, 2011). The new Google+ has been criticized (and praised) components of the proposed model.
for the Circles feature where users may classify their friends and The Unified Theory of Acceptance and Use of Technology model
acquaintances and thereby controlling what they see. While some (UTAUT) (Venkatesh et al., 2003) is another foundational theory for
think this is efficient, others find it hard to classify friends and give social media fatigue. The UTAUT has four key constructs – perfor-
up (Boulton, 2011). ‘‘I’ve started to feel that Google Plus is asking mance expectancy, effort expectancy, social influence and facilitat-
too much from me. Rather than classify my contacts as I might sub- ing conditions that have been shown to influence technology use,
consciously in real life – as family, friends, or coworkers – I’ve been particularly in organizational settings. There have been several
forced to consciously determine my relationship with these people extensions and replications of this theory to address new contexts,
online . . . it is overwhelming’’, said critic Clint Boultan, from user populations and constructs (Venkatesh, Thong, & Xu, 2012).
eweek.com, a technology news website (Boulton, 2011, p. 2). Facilitating conditions is defined as ‘‘consumers’ perceptions of
Social media fatigue can come from not only interpersonal inter- the resources and support available to perform a behavior’’, and
action but also interaction with companies and brands. Studies show effort expectancy defined as ‘‘the degree of ease associated with
that consumers want things like discounts and reviews (with ‘‘being consumers’ use of technology’’, (Venkatesh et al., 2012, 159) are
a part of a community’’ ranked last) while brands want to use social both related to social media fatigue’s social media self efficacy.
media to provide information about new products and get opinions The UTAUT2 also extends this theory deeper into the consumer
about products and services (with ‘‘discounts’’ ranked last). There perspective while adding three new constructs: hedonic motiva-
seems to be a disconnection between what brands are delivering tion, price value, and habit. The concept of hedonic consumption
to consumers and what consumers want from brands (Shashank, (Holbrook & Hirschman, 1982) does play a role in social media fati-
2011). And despite the launch of many new social networking plat- gue since many people engage in social media for pleasure and
forms (e.g. Quora, Empire Avenue, Chime.in.) in 2011, many people other experiences. As such, this extension theory also plays a key
did not rush toward the ‘‘shiny new objects’’ like they have in the role in explaining social media fatigue.
 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155 151

However, we feel that one theory seems to explain the psycho- ance consumers have for dealing with content on social media
logical trigger of information overload. The Limited Capacity Model websites. Based on a recent study conducted by Pew Internet and
(LCM) assumes that individuals have a limited capacity to process American Life (2010), American consumers are becoming more
information. Primarily, LCM was developed to examine how people confident with social media websites as evidenced by an increase
process television messages (Lang, 1992, 1995; Lang & Basil, 1998). in the total number of sites to which consumers belong. In addi-
However, it has since been applied to studies on online forms of tion, the rate at which adult consumers share content online has
advertising (Lee & Faber, 2007; Macias, 2003). The model makes increased in recent years, reaching roughly 30% by September
two major assumptions: (1) people are information processors 2009 (Pew, 2010). These findings indicate that as consumers have
and (2) a person’s ability to process information is limited (Lang, an increased perception of their abilities to use social media web-
2000). Based on these assumptions, Lang (2000) proposes that sites they expand their networks and share more content – both of
there are three steps in information processing: encoding, storage, which are perceived as enjoyable experiences for social media
and selection. By following these steps, people are able to receive a website users. Thus,
stimulus like an online advertisement, analyze it, and then store it
for retrieval at a later date. Many things affect how well a message H1. Respondents with greater social media self-confidence will be
is processed and then ultimately retrievable. One primary factor is less likely to experience social media fatigue.
whether the recipient has sufficient processing resources to pro-
cess the message. This lack of processing power can be because
the recipient chooses not to allocate enough resources to process 2.2. Social media self-efficacy
the task or that the message may require more resources than is
available to allocate to the task. In either case, fewer resources According to Bandura (1986), self-efficacy is belief in one’s abil-
are allocated than are required to adequately process a message. ity to organize and execute a particular course of action – in this
In a social media context, the recipient (or user) may feel over- case, engagement and participation in social media. Self-efficacy
whelmed by the sheer number of messages and therefore not allo- is particularly relevant for novice users who have not yet mastered
cate enough cognitive resources to process the message. Or the skills needed to utilize social media sites like Facebook. Self-
depending on the type of message, it may just require too many efficacy suggests that as social media users become more self-effi-
resources from the user to adequately process the message. cacious, their expectations of obtaining specific outcomes will also
increase. As a result, their experience will encourage further usage.
Eastin and LaRose (2000) believe this is a result of enactive mastery
2. A conceptual model for exploring social media fatigue
(Bandura, 1986), which implies Internet users will gradually mas-
ter complex tasks. Users with low self-efficacy are less likely to
Social media fatigue is rooted in the idea that too much informa-
perform related behaviors in the future (Bandura, 1982). This
tion stemming from social media can lead to feelings of being over-
means they would be less likely to utilize and engage in social
whelmed. As such, LCM is being utilized in this research to
media related behaviors and activities than those with higher
determine whether information overload plays a conceptual role
degrees of self-efficacy. Eastin and LaRose (2000) clarify that self-
in social media fatigue. Research in psychology and organizational
efficacy is not a measure of skill, but instead a reflection of what
science have shown that humans have limited information process-
users believe they can do with the skills they possess. Related to
ing capacity and that information that exceeds this capacity will
this is the TAM construct PEU – based on a person’s perception
result in reduced performance (Hunter, 2004). Evidence shows that
of the ease of use or operation of the social media website, he or
large amounts of information can lead to information overload.
she will adjust their intentions to adopt the technology (Davis,
Information overload is defined as a state induced by a level of infor-
1989). In addition, the self-efficacy construct in this study relates
mation exceeding the ability of an individual to assimilate or process
to two constructs in the aforementioned UTAUT model – effort
a given unit of time (Jacoby, Speller, & Kohn, 1974). In the past, it has
expectancy and facilitating conditions (Venkatesh et al., 2003). Pre-
been reflected in two dimensions: increasing errors and negative
vious experience precedes self-efficacy, which means the more
affect (e.g. confusion or frustration). This research aims to examine
exposure to Facebook and other social media websites, the more
the latter – the negative affect associated with social media fatigue.
self-efficacious users will become. And, in turn, the more self-effi-
For example, changes to Facebook’s layout have moved users
cacious, the more likely the user will continue engaging and partic-
into a new era of sharing. While the layout provides more opportu-
ipating in social media. Therefore,
nity to know the interests and activities of one’s social network, it
may provide an overwhelming amount of information. Addition-
H2. Respondents with greater social media self efficacy will be less
ally, more activity on Facebook is transitioning from intentional
likely to experience social media fatigue
to automatic. For example, instead of users choosing to share infor-
mation, applications like Spotify, a digital music service, automat-
ically post information into the news feed for the user. If a user is
2.3. Concerns with privacy
listening to a great song, he no longer has to choose if or when he
wants to share that information with his social network. The appli-
As social media grows, the issue of privacy becomes ever more
cation will automatically post it for him. This transition from inten-
important. Facebook continues to have a liberal viewpoint on pri-
tional to automatic may increase the likelihood of feelings of social
vacy and Google recently announced major changes to privacy for
media fatigue with the addition of The Ticker, which provides real
all of its properties. In general, the transparent interaction between
time interactions. There are several antecedents that we want to
user and site raises concerns about privacy online (Karahasanovic
examine for social media fatigue.
et al., 2009). Indeed, research from Gartner (2001) shows that pri-
vacy is a major concern with users. We would expect those with
2.1. Social media confidence higher privacy concerns to also have social media fatigue due to
the ‘‘threshold beyond which social contact becomes irritating
Social media confidence describes a person’s perception of their for all parties’’ (Schwartz, 1968, p. 741) meaning that people might
ability to use social media in an effective manner. For the purposes experience being too accessible or receiving too much information
of this research, confidence refers to the level of certitude or assur- from too many people. Additionally, there are continued concerns
152 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155

about what social media sites do with the information collected. income information. In terms of education level, 26.1% (N = 195)
Therefore, of the sample had undertaken some college, 23.7% (N = 177) were
high school graduates, 19.7% (N = 147) received a bachelor’s
H3. Respondents with greater levels of privacy concern will be degree, 11.6% (N = 87) have a graduate degree, 10% (N = 75) com-
more likely to experience social media fatigue. pleted an associate’s degree, 5.1% (N = 38) have completed some
high school, and 3.7% (N = 28) have completed some graduate
school. Respondents belonged to a variety of social media websites
2.4. Social media helpfulness including Facebook, Twitter, Google+, Pinterest, and LinkedIn. On
average, respondents have active accounts on between one and
Social media helpfulness refers to the extent to which users two social media websites (M = 1.5, SD = 1.48).
gain resources and helpful information from their exploration of
social media sites. Research has explored why people participate
3.3. Measures
in social media and social networks (Foster, Francescucci, &
West, 2010). Much of the research centers on various theoretical
The survey instrument included measures related to social
platforms such as social capital (Ellison, Steinfield, & Lampe,
media fatigue, self-efficacy related to social media use, social
2007; Wasko & Faraj, 2005) or uses and gratifications theory
media helpfulness, self-confidence related to social media, and pri-
(Papacharissi & Rubin, 2000; Raacke & Bonds-Raacke, 2008).
vacy concerns experienced when using social media. Sub-items
Foster et al. (2010) found that people used social networks for its
within each scale were averaged resulting in composite scales.
information value (among others). Therefore we explore whether
Items are provided in Appendix A.
people find information on social media helpful in their daily life.
Therefore,
3.3.1. Independent variables
H4. Respondents with greater perceptions of social media help- Respondents’ levels of social media confidence, helpfulness and
fulness are less likely to experience social media fatigue. self-efficacy were measured by adapting existing scales to fit
within the social media context (Bearden, Netemeyer, & Haws,
2011). Participants were asked to respond to a variety of seven-
3. Methodology point, Likert type scales (1 = Strongly Disagree, 7 = Strongly Agree)
to assess their social media aptitude as it relates to confidence,
3.1. Design self-efficacy and helpfulness. Social media confidence was mea-
sured with a three-item social media confidence scale (a = 0.95,
A 210-item questionnaire was developed and pre-tested on a M = 5.60, SD = 1.31). Social media helpfulness was measured with
small group of academic professionals to ensure clarity. Once a four-item scale (a = 0.84, M = 4.99, SD = 1.31). An eight-item scale
tested, the online survey was created and administered using Qual- was used to assess each participant’s level of social media self-effi-
trics, a web-based survey management system. The survey was cacy (a = 0.92, M = 5.06, SD = 1.15). Social media privacy concern
administered to an opt-in subject pool recruited for web-based was measured on a three-item scale (a = 0.86, M = 4.99, SD = 1.36).
research (i.e. online panel). A representative sample of American
social media users between the ages of 18–49 was recruited for
3.3.2. Dependent variable
the survey. Data was gathered from a total of 750 participants over
Respondents’ level of social media fatigue was measured using
a seven-day period to insure an even distribution of respondents
a five-item scale adapted from a Gartner 2010 study (Technopedia,
on each day of the week (i.e., weekdays and weekend days).3
2011). Participants were asked to respond to a series of Likert-type
statements on a seven-point scale (1 = Strongly Disagree,
3.2. Sample
7 = Strongly Agree). A few example statements are as follows: ‘‘I
am frequently overwhelmed by the amount of information avail-
The final sample consisted of 747 current social media users
able on social media sites’’, ‘‘The amount of information available
with a Facebook account who currently reside in the United States.
on social media sites makes me feel tense and overwhelmed’’,
Among this sample, 47.5% (N = 355) were male and 52.5% (N = 392)
and ‘‘When searching for information on social media websites I
were female. Respondents ranged in age from 18 to 49 (M = 32.52,
frequently just give up because there is too much information to
SD = 9.1). With regard to ethnicity, 69.5% (N = 519) were Caucasian,
deal with’’. The social media fatigue scale achieved good reliability
8% (N = 60) were African-American, 7.6% (N = 57) were Asian
(a = 0.91, M = 4.14, SD = 1.39).
American, 6.2% (N = 46) were Hispanic American, 2.4% (N = 18)
were American Indian, 2.4% (N = 18) were Multiracial, 0.5%
(N = 4) were International, and 3.3% (N = 25) preferred not to indi- 4. Results
cate their ethnicity. The majority of the sample had an annual
household income level of $50,000 (USD) or above (38.4%, To test the hypothesized conceptual framework for social media
N = 287) with the remaining sample being evenly distributed fatigue, a confirmatory regression was undertaken using IBM SPSS
among an annual household income of $40,000–$49,000 (11.5%, Statistics 22 software. The overall model was significant (F(4,
N = 86), $30,000–$39,000 (14.6%, N = 109), $20,000–$29,000 742) = 72.93, R2 = 0.282, adjusted R2 = 0.278). Specifically, the
(12.3%, N = 92), and below $20,000 (14.7%, N = 110). A total of 63 results of this investigation indicate that all four antecedents,
respondents (8.4%) opted not to provide their annual household social media privacy, self-efficacy, confidence and helpfulness, all
had significant relationships with social media fatigue (see Table 3).
3
Roughly 107 respondents were collected each day from January 5th to 12th, 2012. However, two antecedents, self-efficacy and helpfulness, had posi-
Respondents were incentivized to complete the survey using a virtual credits system tive and significant predictive effects for social media fatigue
developed by ResearchNow, the panel administrator. ResearchNow enables research- which was the inverse of the hypothesized model. Privacy con-
ers to gather representative data by embedding surveys into social networking
websites and online communities. Respondents completed a profile questionnaire
cerns and social media confidence were significant in their ability
and were screened via a digital fingerprinting device prior to participation in the to predict social media fatigue in line with the hypothesized model
survey. (see Fig. 1).
 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155 153

Table 2
Descriptive statistics for model constructs.

Mean SD Cronbach’s alpha

Independent variables
Social media confidence 5.6 1.31 0.95
Social media helpfulness 4.99 1.31 0.84
Social media self-efficacy 5.06 1.15 0.92
Social media privacy 4.99 1.36 0.86
Dependent variable
Social media fatigue 4.14 1.39 0.91

Note: All scales were measured using seven-point Likert type scales. Composite
measures were computed as an average of the multi-item scales.

4.1. Hypothesis testing Fig. 1. A conceptual model of social media fatigue.

Hypothesis 1 posits that respondents with a greater degree of

social media self-confidence will be less likely to experience social
media fatigue. Using confirmatory regression to test this relation- have alluded to it, there has been no academic research that
ship within the realm of Facebook usage, a significant relationship explicitly examines social media fatigue and its antecedents.
was revealed (b = 0.35, p < .00). Thus, Hypothesis 1 is supported – While exploratory in nature, this research showed that there does
when respondents have a greater degree of social media confi- seem to be a relationship among several established constructs –
dence they are less likely to experience social media fatigue. Alter- social media confidence, social media self-efficacy, privacy con-
natively, those respondents with a lesser degree of social media cerns and social media helpfulness – and social media fatigue.
confidence were more likely to experience social media fatigue. Results showed that respondents with greater levels of confi-
Hypothesis 2 suggests that respondents with a greater social dence about their social media usage were less likely to experi-
media self-efficacy will be less likely to experience social media ence social media fatigue. However, we found a positive
fatigue. The data indicate a positive and significant relationship relationship between social media self-efficacy and social media
between social media self-efficacy and social media fatigue fatigue that was not the hypothesized relationship. This relation-
(b = 0.27, p < .00). Thus, Hypothesis 2 was not supported. ship may be the result of burnout experienced when using social
Hypothesis 3 posits that respondents with a greater privacy con- media. For example, respondents with a high degree of social
cerns will be more likely to experience social media fatigue. As media self-efficacy may use social media to a higher degree
shown in Table 2, social media privacy concerns have a positive rela- because they feel competent using this form of media. As such,
tionship with social media fatigue. Hypothesis 3 is supported – they may experience high levels of social media fatigue due to
when respondents have greater privacy concerns they are more increased usage. We found that there was a relationship between
likely to experience social media fatigue (b = 0.46, p < .00). This find- privacy concerns and social media fatigue whereas those respon-
ing suggests that privacy concerns may produce anxiety when visit- dents with higher levels of privacy concerns experienced social
ing social media websites thus increasing the likelihood of fatigue. media fatigue. Last, we found a positive relationship of social
Finally, Hypothesis 4 suggests that respondents with a greater media helpfulness and greater levels of social media fatigue.
perception of social media helpfulness are less likely to experience However, contrary to our hypothesis, when respondents experi-
social media fatigue. The positive and significant relationship ence high levels of social media helpfulness they also experience
shown in Table 2 shows that the inverse is actually the case. Thus, high levels of social media fatigue. Similar to social media self-
Hypothesis 4 is not confirmed (b = 0.21, p < .00). efficacy, it may be the case that respondents who find social
media to be helpful use it to a greater degree and thus experience
greater levels of social media fatigue. Other research has found
5. Discussion and conclusions mixed results regarding helpfulness and usefulness of social
media (Curran & Lennon, 2011). This finding could be further
The primary contribution of this research is to examine the explored using TAM or UTAUT to identify how the factors that
construct of social media fatigue. While bloggers and pundits

Table 3
Regression analysis.

Unstandardized Standard Standardized t p-level Collinearity

estimates error estimates statistics
B SE b Tol VIF
Intercept 1.36 0.23 5.97 0.00
Social media privacy 0.46** 0.03 0.45 13.40 0.00 0.85 1.17
Social media helpfulness 0.21** 0.05 0.20 4.48 0.00 0.48 2.09
Social media confidence 0.35** 0.05 0.32 6.75 0.00 0.42 2.37
Social media self-efficacy 0.27** 0.06 0.22 4.26 0.00 0.36 2.78
F(4, 742) 72.93
R2 0.282
Adjusted R2 0.278

Notes: Dependent variable is social media fatigue; predictors are social media privacy, social media helpfulness, social media confidence, and
social media self-efficacy.
**
p < .00.
154 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155

influence technology adoption impact the usage of social media Appendix A. (continued)
websites (Davis, 1989; Venkatesh et al., 2003).
Even as exploratory research, this has theoretical and manage- Independent variables
rial implications. First, this research extends the limited capacity Social media self-efficacy
theory further into a digital context where there is much more My interaction with social media is clear and
activity and the likelihood of becoming overwhelmed by content understandable
is higher than in traditional media. We can start to understand Interacting with social media does not require a lot of
the nuances of social media and how its opt in nature differs from mental effort
the more passive consumption of traditional media such as televi- I find social media websites 6 to use
sion and magazines. It also ties in other technology acceptance the- I find it 6 to get social media websites to do what I want to
ories – namely the TAM and UTAUT as foundational theories. do
Second, managers (particularly those who advertise online and I enjoy the technical improvements that social media bring
through social media) will find these results compelling. Many me
companies still struggle with the move from a top down commu- Social media websites make life easier
nication structure of traditional advertising to a more bottom up I feel that social media websites are superior to older
grassroots foundation of social media whereby other users create technologies such as websites
conversations around brands and thus do much of the promotion. I think social media websites give me more control over my
Many companies are still approaching social media as a collection life
of tactics – developing a Facebook presence, doing a few promo-
Social media privacy
tions, tweeting a few links and posting videos to YouTube without
I am concerned about my privacy on social networks
considering the larger strategic imperative. As such brands are
I believe that my personal information can easily be used by
guilty of creating much of that useless content that clutters social
marketers on social networks
media. Understanding how social media users perceive and thus
I feel that I have to give to much information to social
react to this content is important. This research is exploratory in
networks
nature and as such there are several avenues for further develop-
ment. Future research should continue to develop the concept of
social media fatigue and most importantly determine its effects Dependent variable
on information consumers. There are several moderators that can Social media fatigue
be examined. The concept of habit as used in technology accep- I am likely to receive too much information when I am
tance literature should be explored. Further work could identify a searching for something on social media sites
‘‘tipping point’’ whereby people start to experience greater social I am frequently overwhelmed by the amount of information
media fatigue. Research can also determine if social media fatigue available on social media sites
is higher or lower for various generational cohorts (e.g. Gen X vs. I find that social media sites do not have enough detail to
Gen Y vs. Digital Natives who have grown up with social media quickly find the information I am looking for
and Internet connectivity and explore the implications for brands. The amount of information available on social media sites
It would be interesting to see how far people take social media fati- makes me feel tense and overwhelmed
gue. As such it would be interesting to map the process from social When searching for information on social media sites, I
media adoption to social media fatigue to whatever ending behav- frequently just give up because there is too much to deal
ior happens – whether it is taking a social media break, or quitting with
altogether.

Acknowledgements
References
This research was supported by summer research grants pro-
Bandura, A. (1982). Self-efficacy mechanisms in human agency. American
vided by the College of Communication (first author) and Neeley
Psychologist, 37, 122–147.
School of Business (second and third author) at Texas Christian Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.
University. Englewood Cliffs, NJ: Prentice Hall.
Barger, C. (2011). The Top Ten Social Media Lessons of 2011: Part I <www.
forbes.com/sites/christopherbarger/2011/12/20/the-top-ten-social-media-lessons-
Appendix A. Scale items of-2011.htm>.
Bearden, W. O., Netemeyer, R. G., & Haws, K. L. (2011). Handbook of marketing scales:
Multi-item measures for marketing and consumer behavior research (3rd ed.).
Thousand Oaks, CA: Sage.
Becker, M., Alzahabi, R., & Hopwood, C. (2013). Media multitasking is associated
Independent variables with symptoms of depression and social anxiety. Cyberpsychology, Behavior, and
Social Networking, 16(2), 132–135.
Social media confidence Biocca, F., Owen, C., Tang, A., & Bohil, C. (2007). Attention issues in spatial
information systems: Directing mobile users’ visual attention using augmented
I feel comfortable using social media
reality. Journal of Management Information Systems, 23(4), 163–184.
I understand how to use social media Boskers, B. (2011). Gina Bianchini, Co-Founder of Ning on what’s next for social media.
I am confident in my ability to use social media <http://www.huffingtonpost.com/2011/10/11/gina-bianchini-ning-social-
media_n_1005676.html>.
Social media helpfulness Boulton, C. (2011). Google+ arrives, after social media fatigue <http://www.
eweek.com/c/a/Search-Engines/Google-Arrives-After-Social-Media-Fatigue-
Helps me keep in touch with family and friends
585397/>.
Helps me learn new things Bradshaw, T. (2011). Share and share alike: Social media vie for attention <http://
Helps me do my job www.ft.com/intl/cms/s/2/1671609a-d9f3-11e0-b199-00144feabdc0.html#axz
Helps me share ideas and creations with others z1bdYNylEn>.
 L.F. Bright et al. / Computers in Human Behavior 44 (2015) 148–155 155

Brasel, S. A., & Gips, J. (2011). Media multitasking behavior: Concurrent Lee, M., & Faber, R. J. (2007). Effects of product placement in on-line games on brand
television and computer usage. Cyberpsychology, Behavior, and Social memory. Journal of Advertising, 36(4), 75–90.
Networking, 14(9), 527–534. Lee, J.-E., Moore, D. C., Park, E.-A., & Park, S. G. (2012). Who wants to be ‘‘friend-
Bucher, E., Fieseler, C., & Suphan, A. (2013). The stress potential of social media in rich’’? Social compensatory friending on Facebook and the moderating role of
the workplace. Information, Communication & Society, 31(10), 1639–1667. public self-consciousness. Computers in Human Behavior, 28, 1036–1043.
Curran, J. M., & Lennon, R. (2011). Participating in the conversation: Exploring usage Lipsman, A. (2011). The network effect: Facebook, Linkedin, Twitter & Tumblr reach
of social media networking sites. Academy of Marketing Studies Journal, 15(1), new heights in May. comScore <http://blog.comscore.com/2011/06/
21–38. facebook_linkedin_twitter_tumblr.html>.
Davis, F. (1989). Perceived usefulness, perceived ease of use, and user acceptance of Macias, W. (2003). A preliminary structural equation model of comprehension and
information technology. MIS Quarterly, 13, 319–340. persuasion of interactive advertising brand web sites. Journal of Interactive
Drolet, A., & Luce, M. F. (2004). The rationalizing effects of cognitive load on Advertising, 3(2), 36–48.
emotion-based trade-off avoidance. Journal of Consumer Research, 31(1), 63–77. McCord, B., Rodebaugh, T., & Levinson, C. (2014). Facebook: Social uses and anxiety.
Eastin, M. S., & LaRose, R. (2000). Internet self-efficacy and the psychology of the Computers in Human Behavior, 34, 23–27.
digital divide. Journal of Computer-Mediated Communication, 6(1). <http:// Nasco, S. A., & Bruner, G. C. (2009). Perceptions and recall of advertising content
jcmc.indiana.edu/vol6/issue1/eastin.html>. presented on mobile handheld devices. Journal of Interactive Advertising, 7(2),
Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook ‘friends’: 1–22.
Social capital and college students’ use of online social network sites. Journal of Nielsen. (2011). State of Social Media: Social Media Report Q3, (accessed September
Computer-Mediated Communication, 12, 241–263. 10, 2011). [Available at <http://www.nielsen.com/us/en/insights/reports/2011/
Facebook. (2014). Facebook Statistics <https://www.facebook.com/press/ social-media-report-q3.html>].
info.php?statistics>. Palfrey, J., & Gasser, U. (2008). Born digital: Understanding the first generation of
Foster, M. K., Francescucci, A., & West, B. (2011). Why users participate in online digital natives. New York, NY: Basic Books.
social networks. International Journal of e-Business Management, 4(1), 3–19. Papacharissi, Z., & Rubin, A. (2000). Predictors of internet use. Journal of Broadcasting
GlobalWebIndex. (2011). Wave 5 trends <http://www.slideshare.net/ and Electronic Media, 44, 175–196.
Tomtrendstream/wave-5-trends-master-august-2011-slideshare-version>. Pew Internet & American Life Project. (2010). Social media and young adults,
Goasduff, L., & Pettey, C. (2011). Gartner survey highlights consumer fatigue with Washington, D.C. <http://www.pewinternet.org/~/media//Files/Reports/2010/
social media <http://www.gartner.com/it/page.jsp?id=1766814>. PIP_Social_Media_and_Young_Adults_Report_Final_with_toplines.pdf>.
Gross, D. (2011). Users not happy with new Facebook changes <http:// Przybylski, A., Murayama, K., DeHann, C., & Gladwell, V. (2013). Motivational,
articles.cnn.com/2011-09-21/tech/tech_social-media_facebook-changes-react_ emotional, and behavioral correlates of fear of missing out. Computers in Human
1_facebook-top-stories-users?_s=PM:TECH>. Behavior, 29, 1841–1848.
Hill, R., & Moran, N. (2011). Social marketing meets interactive media: Lessons for Raacke, J., & Bonds-Raacke, J. (2008). MySpace and Facebook: Applying the uses and
the advertising community. International Journal of Advertising, 30(5), 815–838. gratifications theory to exploring friend-networking sites. CyberPsychology &
Holbrook, M. B., & Hirschman, E. C. (1982). The experiential aspects of consumption: Behavior, 11(2), 169–174.
Consumer fantasies, feelings and fun. Journal of Consumer Research, 9(2), Schwartz, B. (1968). The social psychology of privacy. American Journal of Sociology,
132–140. 73, 741–752.
Hunter, G. (2004). Information overload: Guidance for identifying when sales Shashank (2011). Brands start to show signs of social media fatigue <http://
information becomes detrimental to sales force performance. Journal of Personal fastgush.com/marketing/brands-start-to-show-signs-of-social-media-fatigue.
Selling & Sales Management, 24(2), 91–100. html>.
Jacoby, J., Speller, D. E., & Kohn, C. A. (1974). Brand choice behavior as a function of Sullivan, D. (2011). Twitter CEO Dick Costolo’s State of the Union Address <http://
information load. Journal of Marketing Research, 11(February), 63–69. searchengineland.com/live-blog-twitter-ceo-dick-costolos-informal-business-
Kaplan, A., & Haenlein, M. (2010). Users of the world, unite! The challenges and address.htm>.
opportunities of social media. Business Horizons, 53, 59–68. Sullivan, D. (2012). Twitter CEO Dick Costolo’s State of the Union Address, (accessed
Karahasanovic, A., Brandtzaeg, P. B., Vanattenhoven, J., Lievens, B., Nielsen, T. K., & September 8, 2011). [Available at <http://searchengineland.com/live-blog-
Pierson, J. (2009). Ensuring etiquette, trust, and privacy when developing Web twitter-ceo-dick-costolos-informal-business-address.htm>].
2.0 applications. IEEE Computer, 42(6), 42–49. Technopedia. (2011). Definition of social media fatigue <http://www.techopedia.
Lang, A. (1992). A limited capacity theory of television viewing. Paper presented to com/definition/27372/social-media-fatigue>.
the information systems division of international communication association, Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of
Miami, FL. (May). information technology: Toward a unified view. MIS Quarterly, 27, 425–478.
Lang, A. (1995). Defining audio/video redundancy from a limited capacity Venkatesh, V., Thong, J., & Xu, X. (2012). Consumer acceptance and use of
information processing perspective. Communication Research, 22, 86–115. information technology: Extending the unified theory of acceptance and use
Lang, A. (2000). The limited capacity model of mediated message processing. Journal of technology. MIS Quarterly, 36(1), 157–178.
of Communication (Winter), 46–70. Wasko, M. M., & Faraj, S. (2005). Why should I share? Examining the social capital
Lang, A., & Basil, M. D. (1998). Attention, resource allocation, and communication and knowledge contribution in electronic networks of practice. MIS Quarterly,
research: What do secondary task reaction times measure anyway? In M. Roloff 29(1), 35–57.
(Ed.), Communication Yearbook 21. Beverly Hills, CA: Sage.