Humor Effectiveness in Social Video Engagement

Abstract

This study extends research on humor effectiveness to videos intended for social media engagement. A survey and content analysis of 2911 videos revealed an attitude favorability toward humor over serious entertainment especially when the message is intended for goods classified as low involvement and emotionally motivated. The study also suggests, however, that humor may dissuade social media participants from imparting comments. Although examination of structural characteristics suggests no significant differences in the performance of humor types suggested in Speck’s (1991) taxonomy, results showed that content is more favorably received when humor dominates than when humor is subordinate to message information.

Full text

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=mmtp20
Journal of Marketing Theory and Practice
ISSN: 1069-6679 (Print) 1944-7175 (Online) Journal homepage: http://www.tandfonline.com/loi/mmtp20
HUMOR EFFECTIVENESS IN SOCIAL VIDEO
ENGAGEMENT
James M. Barry & Sandra S. Graça
To cite this article: James M. Barry & Sandra S. Graça (2018) HUMOR EFFECTIVENESS IN
SOCIAL VIDEO ENGAGEMENT, Journal of Marketing Theory and Practice, 26:1-2, 158-180, DOI:
10.1080/10696679.2017.1389247
To link to this article: https://doi.org/10.1080/10696679.2017.1389247
Published online: 02 Mar 2018.
Submit your article to this journal
Article views: 2
View related articles
View Crossmark data

HUMOR EFFECTIVENESS IN SOCIAL VIDEO ENGAGEMENT
James M. Barry and Sandra S. Graça
This study extends research on humor effectiveness to videos intended for social media engagement.
A survey and content analysis of 2,911 videos revealed an attitude favorability toward humor over
serious entertainment, especially when the message is intended for goods classified as low involve-
ment and emotionally motivated. The study also suggests, however, that humor may dissuade social
media participants from imparting comments. Although examination of structural characteristics
suggests no significant differences in the performance of humor types suggested in Speck’s (1991)
taxonomy, results showed that content is more favorably received when humor dominates than
when humor is subordinate to message information.
Breaking through content clutter on social media
remains a challenge for marketers. In their attempts to
stand out, brands are embarking on social influence
strategies that show promise in garnering followers
that ultimately rewards the influencer with their
patronage. But the route to building this fan base
requires more than helpful tips or timely news.
Leading social media influencers are now discovering
the power of entertainment in boosting the enthu-
siasm of their followers (Barry 2015). A study of social
media influence archetypes (Barry and Gironda 2017a)
in fact found the use of humor to be a distinguishing
characteristic of the content shared by leading influen-
cers. Rather than using informative messaging strate-
gies to capture and hold their audiences’attention,
these leading influencers are banking on social enter-
tainment to build lasting emotional connections with
their followers.
Fueling this trend toward entertaining of emotion-
ally charged content are videos which are shared more
often than any other kind of social content (Frasco
2015). Now rapidly approaching three-quarters of all
internet traffic (Syndacast 2014), brands are recogniz-
ing high returns on their investments in social video
content. A study of over 2,000 professional marketers
and small and midsize business owners, for example,
found that 77 percent of those who use social video
marketing confirmed its direct impact on their busi-
nesses (O’Neill 2016). Moreover, digital metric evidence
suggests that audiences expect some form of amuse-
ment or enjoyment from their social community influ-
encers (Barry and Graça 2013; Barry and Hale 2014).
Yet despite this growing popularity of humorous
social videos, research in the field is limited in scope
and accuracy. The following study addresses these lit-
erature gaps along with this study’sowncontribution
to the conceptual understandings and empirical test-
ing of humor effectiveness in content marketing. In
particular, this study extends the research on humor
by examining how, and in what situations, humor
contributes to the performance of video content cast
in social media settings. The research explores the
connection between different types of humor and
their successful execution from the afterlife engage-
ment of commercials recasts hosted on YouTube. As
detailed further, this study venue is well suited to
addressing prior research limitations in sampling and
testing, digital marketing relevance and performance
measurement.
Beginning with sampling limitations, prior research
in humorous advertising lacks a sufficient amount of
data available to draw meaningful conclusions. Studies
conducted during pre-internet times, for example,
required researchers to record content from scheduled
programming or from available magazine publications.
The relatively small samples of captured content, how-
ever, rendered these earlier seminal studies inconclu-
sive when attempting to extend their findings across a
comprehensive humor taxonomy and situational vari-
ables (Speck 1991;Spotts et al. 1997; Weinberger et al.
1995). One variable in particular includes the role of
James M. Barry (D.B.A., Nova Southeastern University),
Associate Professor of Marketing, H. Wayne Huizenga College
of Business and Entrepreneurship, Nova Southeastern
University, Fort Lauderdale, FL, jmbarry@huizenga.nova.edu.
Sandra S. Graça (D.B.A., Nova Southeastern University),
Assistant Professor of International Business, The
Collegium of Comparative Cultures, Eckerd College, St.
Petersburg, FL, gracass@eckerd.edu.
Journal of Marketing Theory and Practice, vol. 26, nos. 1–2 (Winter–Spring 2018), pp. 158–180.
Copyright ÓTaylor & Francis Group, LLC
ISSN: 1069–6679 (print) / ISSN 1944–7175 (online)
DOI: https://doi.org/10.1080/10696679.2017.1389247

audience purchase involvement and motivation asso-
ciated with the advertised product. As described
further, this study examines a sufficient number of
humorous videos to draw meaningful conclusions on
humor message taxonomies in the context of this situa-
tion variable. Moreover, with the aid of digital metric
evidence, the research extends the scope of audience
reactions to millions of actual social video viewers who
engaged with the content.
A second limitation from using print and scheduled
television recordings in pre-internet studies relates to
questionable testing approaches. Examinations to date
have been restricted to measures of humorous advertis-
ing effectiveness in nondigital media outlets where
experimental study participants were exposed to fic-
tional advertisements that lacked any recorded perfor-
mance data (Cochran 2004; Flaherty et al. 2004; Zhang
and Zinghan 2006). Other studies examined the fre-
quency of humorous executions employed by adverti-
sers as a proxy for perceived humor (Hatzithomas et al.
2009; Speck 1991; Spotts et al. 1997; Toncar 2001;
Weinberger et al. 1995). One notable drawback of
these approaches, however, is the assumption that the
humor intended by advertisers mirrors the actual audi-
ence perceptions of humor effectiveness. This study
overcomes this limitation by examining actual and
unsolicited audience interest from social media engage-
ment metrics (e.g., comments posted after viewing a
YouTube video).
A third limitation relates to digital marketing rele-
vance. A shortcoming of prior humorous advertising
research includes its restriction to nondigital settings.
Consequently, brands lack information on the effec-
tiveness of humor as a mechanism to encourage con-
tent engagement while capturing the unsolicited
reactions of large audiences. In effect, the recasting of
commercials as social videos permits an afterlife
engagement experience that boosts the exposure,
brand experience, and advertising feedback desired by
brands. The sharing of videos across social media chan-
nels, along with the search engine impact from posted
comments and likes, creates a new opportunity for
brands to expand their social media influence (Barry
and Gironda 2017b). Moreover, brands can gain real-
time insights on the effectiveness and audiences inter-
ested in their commercial recasts. This research exam-
ines the effects of commercials recasts so as to enrich
the literature with insights on the role that humor
plays in boosting social engagement.
Finally, a fourth limitation of humorous advertising
research conducted to date relates to effective performance
measurement. Little, if any, research examines the effec-
tiveness of humor relative to alternative forms of enter-
tainment (e.g., inspiration, amazement, or creative fun).
Furthermore, by not having social media engagement
statistics, prior research lacks a mechanism for qualify-
ing high performance videos. The fact that most studies
found less than 30 percent of the captured advertise-
ments to be humorous suggests a sample skewed toward
low performing informational advertisements. This
study, on the other hand, examines only high perform-
ing videos for both humorous and non-humorous con-
tent. The combination of both forms of content in this
case represented over 95 percent of the content exam-
ined in this range of viewership.
To address these limitations in scope and study
approach, this research combines content analysis
and attitude surveys to answer the following four
research questions:
1. Does humor outperform other forms of entertain-
ing social content?
2. Are the advertising attitude results in studies of
humor message taxonomies extendable to social
engagement behaviors?
3. Does the humor intended by advertisers mirror
the actual audience perceptions of humor effec-
tiveness? That is, are advertisers spending their
money wisely?
4. To what degree is the effectiveness of humor in
social content dependent on the goods being
advertised? That is, does the level of humor inter-
est vary depending on the audience involvement
and motivation associated with the advertised
product?
To effectively answer these questions, this study first
examines the literature and a theoretical framework for
evaluating the effectiveness of humor in advertising
content. We then outline empirical testing used to
evaluate research hypotheses concerning the relative
performance of humor in the context of social media
engagement and advertising attitudes. The theoretical
framework is then examined across product classifica-
tions that reflect audience purchase involvement and
motivation in order to validate the context-specific
effects of humor proposed in the study. Implications
for theory and practice are then discussed.
Winter–Spring 2018 159

IMPORTANCE OF HUMOR
The importance of using humor in advertising has
been well researched over the past three decades. In
the case of advertising, the effectiveness of humor is
often attributed to its capacity to boost message
understanding and acceptance (Alden and Hoyer
1993). But when viewed in the context of social
media and content marketing, humor is also seen as
a welcoming diversion from the content clutter over-
whelming social media audiences. Barry and Gironda
(2017b), for example, demonstrated that humor
impacts social media influence indirectly through
inspiration. In effect, the levity taps into an emotional
sentiment that sets the stage for marketers to spark an
inspirational appeal for their brands in an era of info-
besity. And since social media has exponentially
increased the size of audiences now engaged in enter-
taining content, it behooves marketing scholars and
practitioners to examine formulas for what makes
advertising and digital content humorous.
Despite the estimated $50 billion spent worldwide
and annually just on commercials intended to be
humorous, the complex nature of humor still con-
founds comedians, playwrights, and marketers in
search of universal formulas for applying humor. Of
the attempts to classify humor for more concentrated
research, much of the research to date lacks a universal
theoretical framework in which to examine its effec-
tiveness. Attempts to categorize humor often result in
an unmanageable number of humor techniques sug-
gested for skit storyline manipulations and character
portrayals (Berger 1993; Buijzen and Valkenburg 2004).
To more fully appreciate the manner in which
humor stirs audience interest in an advertiser’s video
advertisement or other content, these next sections
discuss a more parsimonious approach to understand-
ing humor and its impact on advertising. In particular,
light is shed on the interplay of cognitive and affective
factors driving its effective usage. Using theories of
incongruity, superiority, and relief as a baseline for
rationalizing humor effectiveness, the ensuing section
examines a comprehensive humorous message taxon-
omy to determine how humor is best placed in the
video and related to the message. Consistent with
prior research, a content analysis and attitude survey
is then used to determine if humor effectiveness
depends on the audience motives and involvement
associated with the type of goods advertised.
Cognitive Factors Driving Humor Effectiveness
When viewed from a cognitive perspective, marketers
appreciate the role of humor in capturing audience
attention while helping them process information in
the intended marketing message. According to Eisend
(2011), “humorleadstheconsumertoelaboratemore
on the message, thereby enhancing cognitive
responses”(p. 116). By increasing advertising atten-
tion, humor can better ensure that the message will
be comprehended. A number of humor researchers
argue that the cognitive structure of humorous adver-
tisements revealed in its problem-solving processes
can also enhance audience recall (Alden and Hoyer
1993). Moreover, this problem-solving process can
“distract consumers from processing counterargu-
ments”(Eisend 2011, p. 116) which, in turn, can
increase the persuasion capacity of the advertising
message.
Substantial evidence supports this cognitive argu-
ment for why humor works. In their studies of
humorous promotions, for example, Scott et al.
(1990)foundthat“when well integrated with the
objective(s) and message of the ad”(p. 498), humor
can both capture audience attention and enhance
recall. Similarly, Weinberger and Spotts (1989)
confirmed in their study that humor increased audi-
ence attention and comprehension. Finally,
Weinberger and Gulas’(1992) review on the impact
of humor in advertising found that 94 percent of
advertising practitioners and 55 percent of advertis-
ing research executives consider humor an effective
and superior choice to increase audience’s attention.
They report that the impact of humor on attention
is mostly positive in both advertising and nonadver-
tising studies. And, despite the moderating influ-
ences of factors such as age, gender, education,
style of humor and relatedness to product, there is
some empirical support that humor impacts compre-
hension as well.
The impact that humor has on persuasion, how-
ever, is not so clear when viewed from a cognitive
perspective. In their study, Weinberger and Gulas
(1992)foundthat“humor does not appear to offer
an advantage or disadvantage over non-humor at
increasing persuasion”(p.56).Thissupports
Sternthal and Craig’s (1973) conclusion that “humor
enhances persuasion to the same extent as serious
appeals”(p. 15).
160 Journal of Marketing Theory and Practice

Affective Factors Driving Humor Effectiveness
Despite these mixed results from a cognitive perspec-
tive, insights from an affective perspective suggest that
humor enhances persuasion, advertising attitudes, and
engagement. When audiences engage with advertising
at an emotional level, humor serves as a mediating
variable that can influence the audience attitude
toward the message (Unger 1995). Eisend’s(2011)
meta-analysis revealed that “humor effects are primar-
ily based on affective processes”(p. 128) and that
humor increases advertising favorability which, in
turn, impacts attitudes toward the brand. In their
cross-cultural study, Unger (1995) found support for
this affect-based model and the “effectiveness of
humor in persuasion”(p. 68). Their study concluded
that funny advertisements increase advertising atti-
tudes which, in turn, increases product attitudes. In a
similar vein, Alden et al. (2000) concluded that effec-
tive humorous attempts will result in stronger liking of
the marketing communications using the humor.
Finally, in their study of perceived humor, Flaherty
et al. (2004) confirmed that advertisements perceived
as humorous outperformed advertisements not per-
ceived as humorous on both advertisement attitude
and brand attitude. Thus, we posit that:
H
1a
:Audience favorability toward social videos per-
ceived as humorous is greater than the interest of
serious entertainment advertisements when mea-
sured across advertising attitude.
H
1b
:Audience engagement in social videos per-
ceived as humorous is greater than the engage-
ment of serious entertainment advertisements
when measured across comments generated.
THEORETICAL FRAMEWORKS FOR
EXAMINING HUMOR EFFECTIVENESS
This interplay of cognitive and affective factors driving
humor effectiveness is justified by the theories pro-
posed as to why we laugh. Three in particular are
most often cited in the literature. Collectively, they
suggest that laughter comes from:
1. Seeing something out of sorts like irony, gross
exaggeration, or an unexpected surprise (incon-
gruity theory).
2. Enjoying others’inferiorities and misfortunes
(superiority theory).
3. Releasing ourselves from repressed desires or
other nervous energy (relief theory).
The incongruity theory suggests that humor is
derived from a cognitive process, while the superiority
and relief theories assume an affective process. In this
next section, we discuss the relevant theoretical frame-
works used to evaluate humor effectiveness so as to
better understand its structural characteristics. In so
doing, insights can be gained on how humor can be
used to stir audience interest and engagement.
Humor Theories and Mechanisms
The incongruity theory assumes that a cognitive stimu-
lus is required for mirth. This witty form of humor
results when our intellect first recognizes an incongru-
ity between what we expect and what actually hap-
pens. Among the anomalies that can trigger this
contradiction are logical discords, odd behaviors, and
visual aberrations. Once a resolution to the contradic-
tion is recognized, we often laugh in the form of an
“Ah-hah!”In effect, we are rewarding ourselves for
clarity and wit understanding. Since laughter requires
us to process the incongruity as well as resolve it, Speck
(1991) and others refer to this dual process as incongru-
ity-resolution.
A more disparaging or aggressive form of humor
stems from the theory of superiority. This theory
assumes that a socio-emotional stimulus is required to
make us laugh. In effect, the theory contends that
humor arises from the sudden glory we feel after wit-
nessing a well-deserved putdown, the enjoyment of
others’misfortune, or the awkwardness that relieves
us from experiencing the same discomfort. In this
case, laughter in the form of a “Hah-hah!”results
from a contrasting of ourselves with those of perceived
lower status. This disparaging humor starts with a mock-
ery mechanism like sarcasm, outwitting, parodies, or
paybacks. When seen as well-deserved or innocent,
the seemingly innocent interpersonal attack can then
result in sensations of sudden glory manifested in mal-
icious joy or schadenfreude.
Finally, the relief theory attributes humor to the
tension released in the form of aggressive liberation,
sexual allusion, or fear anxiety relief. The theory con-
tends that humor is used mainly to overcome
Winter–Spring 2018 161

sociocultural inhibitions or other suppressed desires.
This pent up nervous energy is released when we vicar-
iously imagine our own rebellion through fantasized
rule breaking, dream exploits, or unruly outbursts. But
unlike the cognitive and socio-emotional forms of
humor represented by the incongruity and superiority
theories, respectively, the relief theory assumes humor
is derived from a physiological state. It follows an arou-
sal-safety narrative, where laughter in the form of an
“Ahhh!”results from the resolution of a stressful
situation.
Linking Audience Involvement/Motivation to
Humor’s Structural Characteristics
Although each of these humor theories contributes to a
better understanding of how humor works, missing are
universal frameworks that capture the essence of each
theory. Universality is also challenged by the wide range
of processed cues stemming from humor. This has led
many researchers to examine humorous advertising fra-
meworks that shed light on varying degrees of audience
involvement and motivation. The Elaboration
Likelihood Model (ELM) (Petty and Cacioppo 1986),
for example, suggests that humor often serves as a per-
ipheral cue whose success largely depends on specific
processing invoked by a recipient of the message
(Zhang 1996). The greater the involvement in the cue
and motivation in argument processing, the greater the
opportunity for humor effectiveness in creating adver-
tising attention and persuasion.
Humor Message Taxonomies
Universality is also challenged by the myriad of dimen-
sions used to analyze humor effectiveness in the con-
text of its message relatedness. One promising
framework, however, is the humorous message taxon-
omy offered by Speck (1991). According to
Hatzithomas et al. (2009), “the various humor types
and processes described by Speck’s taxonomy are
linked to consumer involvement and motivation”(p.
43). Beginning with a framework for measuring incon-
gruity-resolution, disparagement and arousal-safety,
the taxonomy lends itself to an examination of
humor’s structural relationships to an intended mes-
sage. Speck (1991) and others (Hatzithomas et al.
2009; Spotts et al. 1997), for example, analyzed the
degree to which humor-dominated advertisements out-
performed message-dominated advertisements and
whether the intended humor is more effectively exe-
cuted at the beginning, end, or throughout an
advertisement.
Product Type Relatedness to Humor
Besides its relatedness to an advertising message, “a
number of situational factors may play a role in deter-
mining the effectiveness of humor”(Weinberger and
Gulas 1992, in Flaherty et al. 2004, p. 26). The affilia-
tion of humor to products, for example, has been
found to be a strong predictor of advertising success
(Cline and Kellaris 2007). This has led humor research-
ers to consider a number of product typologies that
reflect the ELM involvement/motivation dimensions
to consider when processing an advertising message
(Eisend 2009). In particular, the brand attitude grid
proposed by Rossiter and Percy (1997) classifies goods
into four quadrants: high involvement/emotionally
motivated (red goods), low involvement/emotionally
motived (yellow goods), high involvement/rationally
motivated (white goods), and low involvement/ration-
ally motivated (blue goods).
According to Rossiter and Percy (1997), advertisers
should employ strategies that transform an audience’s
mood when motives are positive or emotionally
impacted. This would be the case for purchasing alco-
hol or jewelry. But when motives are rationally-driven
(negative)—as in the case of buying office supplies—
advertisers should employ informational strategies that
help a target audience address a perceived problem
(Hatzithomas et al. 2009). The dimension characterized
by high and low involvement suggests that audiences
process messages differently depending on the per-
ceived financial risks at stake. Buying laundry detergent
(blue goods) for example would be considered low
involvement; whereas life insurance (white goods)
would be considered a high involvement purchase.
Use of Rossiter and Percy’s brand attitude grid has
spurred a number of humorous advertising studies to
consider how the nature of goods advertised may
impact the way humor is received (Cochran 2004;
Djambaska et al. 2016; Flaherty et al. 2004;
Hatzithomas et al. 2009; Spotts et al. 1997; Toncar
2001; Weinberger et al. 1995; Zhang and Zinkhan
2006). This combination of Speck’s(1991) taxonomy
with Rossiter and Percy’s(1997) brand attitude grid
162 Journal of Marketing Theory and Practice

permits examination of how humor type and message
relatedness impact audience interest across the span of
audience involvement and motivation associated with
the advertised product.
Examination of Speck’s(1991) Taxonomy
This next section describes the manner in which
Speck’s(1991) Humorous Message Taxonomy can be
used as a framework for examining humor’s commu-
nication effects. The taxonomy accounts for differences
in humor type, message type, and humor-message
interaction.
Humor Performance across Humor Types
The examination of humor effectiveness starts with the
incongruity-resolution, arousal-safety, and humorous
disparagement methods used to generate humor.
Speck (1991) identified five distinct types of humor
derived from various combinations of these three
humor methods: comic wit, sentimental humor, satire,
sentimental comedy, and full comedy. Comic wit and
sentimental humor are solely based on one humor
process. Comic wit, for example, is derived from the
incongruity-resolution often exemplified in unexpected
twists, odd behaviors, unusual settings, or unrealistic
exaggerations. Sentimental humor is derived from arou-
sal-safety often exemplified through childlike fantasies,
naughtiness, or inner secrets. As shown in Table 1, the
remaining humor types involve combinations of
humor methods. Satire, for example, combines ele-
ments of incongruity-resolution and disparagement. A
sudden glory may be realized in this case after witnes-
sing a celebrity featured in an unusual setting (incon-
gruity-resolution) through a stereotyping parody that
humbles the celebrity (disparagement). A combination
of arousal-safety and incongruity-resolution leads to
what Speck (1991) calls sentimental comedies. This
type of humor is often exemplified through hysteria,
belligerence, and other forms of unleashed repression.
For example, the exercising of improprieties and anxi-
ety build-up from witnessing someone’s outbursts
(arousal-safety) is juxtaposed on odd behaviors (incon-
gruity-resolution). A combination of this sentimental
comedy with satire then leads to what Speck (1991)
calls full comedy. This rich form of humor is often
exemplified through social order deviancy that com-
bines taboos (arousal-safety), disparagement (society
satire), and unexpected behaviors (incongruity-resolu-
tion). Another form of full comedy is realized in scenes
of awkwardness where the displacement of embarrass-
ment onto others (arousal-safety) is combined with an
unlucky happenstance (disparagement) and an unusual
predicament (incongruity-resolution).
As referenced in the limitations of prior research, the
effectiveness of humor types has traditionally been
measured from the frequency of humor executions
used by advertisers (Hatzithomas et al. 2009; Speck
1991; Spotts et al. 1997). Although this proxy for per-
ceived humor reflects what advertisers believe to be
effective, this study will examine the degree to which
these earlier research results can be duplicated when
measured from an audience’s perspective. Starting
with an examination of humor types, prior research
found comic wit to be the most popular type of
humor (31.2 percent of advertising placements in
Speck’s study) and (33.1 percent of placements in
Hatzithomas et al.’s study). The remaining humor
types were found to be fairly evenly distributed across
both studies. Assuming that advertiser intentions
reflect true audience perceptions, we therefore posit
the following:
H
2a-b
:Audience favorability toward and engage-
ment in social videos perceived as comic wit will
outperform all other forms of humor when mea-
sured across a) advertising attitude and b) com-
ments generated.
Table 1
Types of Humor (Speck 1991)
Incongruity-Resolution Humorous Disparagement Arousal Safety
Comic Wit (HMT1) ✓
Sentimental Humor (HMT2) ✓
Satire (HMT3) ✓✓
Sentimental Comedy (HMT4) ✓✓
Full Comedy (HMT5) ✓✓✓
Winter–Spring 2018 163

Humor Performance across Structural Characteristics
Speck’s(1991) taxonomy further serves as a foundation
in which to examine the structural relationship of humor
to product/service-related messages. This and other
research (Hatzithomas et al. 2009;Speck1991;Spotts
et al. 1997) examines the degree to which humor dom-
inates or is subordinate to non-humorous elements in an
advertised message. Humor dominance implies, for
example, that the absence of humor in the content
would render the message meaningless. The alternative
of message dominance implies that the humor is subor-
dinate to the overall message. In this case, the taxonomy
further analyzes whether the message dominant adver-
tisements are image intensive or information intensive.
Image intensity suggests that the message is primarily
intended to shore up an emotional sentiment to the
brand; whereas information dominance implies a focus
on product or service knowledge such as discount claims
or new feature announcements. Finally, this study and
other research (Hatzithomas et al. 2009;Speck1991;
Spotts et al. 1997) further characterizes message domi-
nance based on the positioning of humor in the adver-
tisement or social content (e.g., initial, embedded, or
closing humor).
Studies of humor’s structural relatedness to advertis-
ing messages show mixed results. Hatzithomas et al.
(2009), Speck (1991) and Spotts et al. (1997)allshowed
a higher frequency of humor dominance compared to
both image dominant and information dominant adver-
tisements. Assuming that advertiser intentions reflect
actual audience perceptions of humor across its struc-
tural characteristics, we therefore posit the following:
H
3a-b
:Audience favorability toward and engage-
ment in social videos perceived as humor domi-
nant will outperform humor perceived as message
dominant when measured across a) advertising
attitudes and b) comments generated.
An examination of humor placement and its impact
on advertising effectiveness, however, suggests that
further research is required. Results of humor place-
ment tabulations in the Speck (1991) and
Hatzithomas et al. (2009) studies, for example, revealed
the following:
1. Initial humor represented 13.6 percent of adver-
tisements in Speck’s study and 36.7 percent of
advertisements in Hatzithomas et al.’s study.
2. Embedded humor represented 21.6 percent of
advertisements in Speck’s study and 11.3 percent
of advertisements in Hatzithomas et al.’s study.
3. Closing humor represented 64 percent of adver-
tisements in Speck’s study and 52.1 percent of
advertisements in Hatzithomas et al.’s study.
But consideration of some social media trends may
suggest that embedded humor works best in light of a
growing attention deficit among social media enthu-
siasts (Ruedlinger 2012). Accustomed to seeing fifteen-
second video spurts on Instagram and Snapchat, for
example, millennials in particular may not tolerate a
postponed injection of humor. At the other extreme,
early placement of humor may lead to quick abandon-
ment of the social video should the remaining content
become more message focused. This leads us to the
following:
H
4a-b
:Audience favorability toward and engage-
ment in social videos where humor is embedded
will outperform initial and closing humor when
measured across a) advertising attitudes and b)
comments generated.
Humor Performance across Product Types and Speck’s
Taxonomy
Considerable research has been devoted to the exam-
ination of humor across product types. Humor in
high involvement products, for example, could nega-
tively impact the brand sentiment sought by a red or
white goods supplier. Given the higher risks asso-
ciated with the purchase of luxury goods (red) or
expensive appliances (white), audiences will likely
prefer messaging that reinforces their more complex
decision-making process as evidenced by the higher
proportion of humorous ads found among viewers of
yellow compared to red products (Toncar 2001).
Audiences of yellow goods (e.g., beer and snacks)
messaging, on the other hand, will likely embrace
the humor as a welcoming distraction from an other-
wise insignificant message about a routine product
purchase. Decision making in these cases relies more
heavily on peripheral cues such as music, humor,
celebrities, and color (Flaherty et al. 2004; Petty
et al. 1983). Moreover, their emotional sentiments
to the brand imply that these audiences expect to
be entertained throughout the overall message. This
may not be the case for goods involving rational
164 Journal of Marketing Theory and Practice

motives, however, where humor may overtake and
even disrupt the processing of product benefits or
other informative claims. In their study, for example,
Hatzithomas et al. (2009) found that the lower the
involvement and the more emotional the motive, the
more embedded humor was used in messaging.
Consistent with these findings, we therefore posit
the following:
H
5a-b
:Audience favorability toward and engage-
ment in social videos for low involvement goods
(yellow and blue) will outperform humor shown
for high involvement goods (red and white) when
measured across a) advertising attitudes and b)
comments generated.
H
6a-b
:Audience favorability toward and engage-
ment in social videos for emotionally motivated
goods (yellow and red) will outperform humor
shown for rationally motivated goods (blue and
white) when measured across a) advertising atti-
tudes and b) comments generated.
H
7a-b
:The use of humor will be a) proportionally
higher for yellow goods and b) proportionally
lower for red goods.
METHODOLOGY
Empirical testing for this study combines a content ana-
lysis for measuring engagement with a self-administered
survey for measuring advertising attitudes. Beginning
with the content analysis of 1,777 videos qualified as
humorous and 1,134 videos qualified as non-humorous,
an examination is made of the effect that humor has on
engagement when measured by comments posted by
social video followers. These videos are restricted to the
recast of brand sponsored commercials whose television
broadcasts have expired. In effect, the now social videos
permit brands to examine the afterlife effects of their
commercial content. Only videos garnering over 50,000
views and posted over the past decade were selected for
analysis. This criteria ensures that the examined humor-
ous and non-humorous attributes have been vetted for
high performance and temporal proximity.
From the search of television commercial recasts
hosted on YouTube, a total of 2,762 videos were origin-
ally identified by the reviewers (i.e., two authors) as
having content intended to be humorous. Each of the
brands sponsoring the video were then categorized into
four product categories labeled as blue, white, yellow,
and red. Shown in Table 2 is an example of product
Table 2
Product Color Matrix with Sample Videos Used in Humor Evaluation
Negative Motivation (Rational) Positive Motivation (Emotional)
White Goods (Big Tools) Red Goods (Big Toys)
High Involvement Category Example (URL) Year Category Example (URL) Year
Banking, Credit Cards & Insurance bit.ly/2qnZxMm 2016 Club Memberships bit.ly/1f3SR6U 2011
Cable, Phone & Utility Services bit.ly/1oEqhVg 2011 Fashion Apparel bit.ly/2oHxEeK 2010
Computer & Software Products bit.ly/2knsTG4 2016 Cosmetics bit.ly/oifGl5 2011
Home Improvement Products bit.ly/1Qz84yT 2013 Entertainment Systems bit.ly/39uWl 2009
Furniture & Appliances bit.ly/2rE6ESv 2011 Sports & Luxury Vehicles bit.ly/cS47Rd 2010
Vehicles (Non-Sports/Luxury) bit.ly/2kYSqES 2017 Travel & Vacationing Services bit.ly/1R55ovA 2014
Trading & Consulting Services bit.ly/2rRqGJy 2010 Resorts & Themed Attractions bit.ly/kZdqWo 2011
Blue Goods (Little Tools) Yellow Goods (Little Toys & Treats)
Low Involvement Category Example (URL) Year Category Example (URL) Year
Essential Diet Items bit.ly/19esaPH 2013 Alcoholic & Soft Beverages bit.ly/vMVOtY 2009
Essential Clothing bit.ly/1WXJ7Ef 2015 Games, Toys & Electronic Arts bit.ly/1emh7nA 2013
Health, Beauty & Hygiene Products bit.ly/9tbsds 2010 Restaurants bit.ly/PioWhb 2008
Household Goods bit.ly/2kxD5ii 2017 Snacks & Desserts bit.ly/A8f6Ow 2012
Office Supplies bit.ly/1mCCb 2009 Sporting Goods bit.ly/om2tE 2008
Routine Maintenance & Repair Services bit.ly/1ihDsDv 2011 Sports Venues bit.ly/21r26pp 2008
Winter–Spring 2018 165

categories and sample URL links used in assigning each
video to its proper grid quadrant.
The commercials were then defined and sorted into
categories of comic devices that dominated the com-
mercial. Up to three comic devices were recorded by
each of the two authors for each video. These devices
(e.g., exaggerated stories, miscast personality,
exercising humility, etc.) were then consolidated in
accordance with the three theories of advertising
humor (incongruity, superiority, and relief) and their
contribution to Speck’s(1991)five humor types.
Shown in Figure 1, for example, is the approach used
to sort high order comic devices into humor type cate-
gories. To ensure consistency in humor type
Figure 1
Conceptual Model of Humor Taxonomy
Humor Types (Speck, 1991)
HMT1: Comic Wit (Incongruenc e)
HMT2: Sentimental Humor (Arousal-Safety)
HMT3: Satire (Inc ongruence + Disparagement )
HMT4: Sentimental Comedy (Incongruence + Arousal-Safety)
HMT5: Full Comedy (Incongruence, Disparagement + Arousal-Safety)
Response to Cognitive Shift
Incongruity Mechanisms
•PD: Per ceptu al Disp lacement
•IJ: Iro nic Ju xtap osit ion
•H: Hy pe rbole
•S: Surpr ise
Odd Behaviors
•PD: Un usual Pers onification (A, C)
•IJ: Iro nic Temp erament (A, C) *,#
•IJ: Ironic P ersona (A, C) *,#
•H: Ov er-reactive Behaviors (A, C , D)
Vis ual Aberrations
•PD: Bizarre Substitutions (A, C)
•IJ: Visual Irony (A)*,#
•H: Exaggerated Qualit ies (A)* ,#
•S: Transformations (A)*,#
•S: Visual Surprise (A)*,#
Logi cal Di s cor d
•PD: No nsens e (A)*,#
•PD: M isrep resent ed Context (A)* ,#
•IJ: Situ atio nal Iron y (A, C)*,#
•H: Exaggerated Outcomes (A)*,#
•H: Understatements (A)
•S: Concep tu al Surp rises (A, C, D)*
•S: Plot Trick ery (A, C, D )
•Puns (A)*,#
Mocke ry Mechanisms
•P: Putdowns*,#
•MJ: M alicious Joy
APerceptual Displacement
•P: Stereot y p ing (C)*,#
•MJ: Bungling Behaviors (C)
Ironi c J uxtapos iti on
•P: Lofty Conquest (C, E)
•MJ: Unlu cky H app ens tance (C , E)
•MJ: Deserved Repercussions (C)
Hype r bol e
•P: Mo cked Peculiarities (C)* ,#
•P: Society Satire (C, E)
•MJ: Cret ins (C)
Surprise
•MJ: Unan ticip at ed Spo iler (C, E)*,#
Feelings of Sudden Glory
Arousal
Me ch a n is m s
•A: Anxiety
•T: Taboos
•I: Infant ilism*, #
C
Unr u l i n es s
•A: H ys teria ( D, E)
•A: Impulsive Outbursts (D, E)
•A: Dis placed Irr itat ion (D)
•T: Exercising Imp ropr ieties (D , E)
Drama
•A: Fear & Anxiety Relief ( D)
•A: Melod rama (D)
Tran s ce nden ce
•T: Inner Secrets (D, E)
•I: Child Innocence (D )
Re pre ss ed Ene rgy Re lea se
D
B
Social Order Deviancy
•Socie ty Irr everenc e
•Forbidden Behaviors,
•Offensive Conduct
•Unleashed Mania
Awkwardnes s *,#
•Remors eful Regret s
•Uncomf ort able Sett ings
•Exercising Humilit y
•Revealed Secret s
Arousal & Mocke ry
E
* Humor technique adopted from
Buijzen and Valkenburg (2004)
# Humor technique adopted from
Berger (1993)
A
B
C
D
E
166 Journal of Marketing Theory and Practice

Table 3
Comparative Favorability and Engagement Results across Product Grid
Example Video Reviewer #1’s Top Comic Device Selections Reviewer #2’s Top Comic Device Selections
URL Year
High Order
Device #1
1
(Low Order Device)
2
High Order
Device #2
1
(Low Order Device)
2
High Order
Device #3
1
(Low Order Device)
2
High Order
Device #1
1
(Low Order Device)
2
High Order
Device #2
1
(Low Order Device)
2
High Order
Device #3
1
(Low Order Device)
2
Red bit.ly/ckFWvK 2010 Exaggerated Outcome
(Exaggerated Results)
Exaggerated Outcome
(Exaggerated Results)
Overreaction (Taking
Extreme Measures)
Yellow bit.ly/fcvsXr 2011 Bizarre Substitution
(Anthro-pomorphism)
Situational Irony
(Unexpected
Outcome)
Misrepresented
Context
(Humanized
Depiction)
Unusual Personification
(Unlikely Animal
Behavior)
Misrepresented Context
(Humanized
Depiction)
Bizarre Substitution
(Animal
Substitution)
Blue bit.ly/39oIy6 2009 Situational Irony
(Misplaced Routine)
Exaggerated Outcome
(Exaggerated
Results)
Exaggerated Outcomes
(Exaggerated
Response)
Exaggerated Qualities
(Supernatural
Performance)
Overreaction (Over
Intense)
White bit.ly/1ho4kAR 2013 Exaggerated Qualities
(Incredible Allure)
Exaggerated Qualities
(Incredible Allure)
Exaggerated Qualities
(Supernatural
Performance)
Exaggerated Outcomes
(Exaggerated
Stories)
Red bit.ly/PbjQ6X 2010 Inner Secrets (Dream
Exploits)
Inner Secrets (Dream
Exploits)
Yellow bit.ly/1QgKXyM 2014 Fear & Anxiety Relief
(Trepidation)
Inner Secrets (Dream
Exploits)
Fear & Anxiety Relief
(Trepidation)
Blue bit.ly/e7PmDd 2011 Child Innocence
(Youthful Discoveries)
Child Innocence (Child
Mimicry of
Adulthood)
Child Innocence
(Youthful Discoveries)
White bit.ly/ntc8Cp 2011 Child Innocence
(Youthful Discoveries)
Fear & Anxiety Relief
(Barely Escaped
Detection)
Melodrama (Fervent
Behaviors)
Fear & Anxiety Relief
(Barely Escaped
Detection)
Red Disparagement bit.ly/ip1J3N 2010 Mocked Peculiarities
(Quirkiness)
Society Satire (Cultural
Nuance)
+ Incongruity Visual Irony (Unusual
Pairing)
Visual Irony (Unusual
Pairing)
Exaggerated Qualities
(Speed & Scale
Distortion)
Yellow Disparagement bit.ly/AicWpQ 2012 Ironic Temperament
(Soft Tough Guy)
Society Satire (Cultural
Nuance)
Lofty Conquest (Macho
Gone Sour)
+ Incongruity Exaggerated Qualities
(Speed & Scale
Distortion)
Understatement
(Unrattled)
Visual Irony (Unusual
Pairing)
Blue Disparagement bit.ly/mm3jdU 2011 Society Satire
(Parodies)
Society Satire (Parodies)
+ Incongruity Exaggerated Outcomes
(Exaggerated Stories)
Nonsense (Irrelevance) Situational Irony
(Anachronisms)
Misrepresented
Context (Unusual
Setting)
White Disparagement bit.ly/5OzFmX 2009 Society Satire (Cultural
Nuance)
Society Satire (Cultural
Nuance)
+ Incongruity Nonsense (Baffling
Dialog)
Situational Irony (Mis-
communications)
Situational Irony (Mis-
communications)
Plot Trickery (Twist of
Fate)
(Continued)
Winter–Spring 2018 167

Table 3
(Continued)
Example Video Reviewer #1’s Top Comic Device Selections Reviewer #2’s Top Comic Device Selections
URL Year
High Order
Device #1
1
(Low Order Device)
2
High Order
Device #2
1
(Low Order Device)
2
High Order
Device #3
1
(Low Order Device)
2
High Order
Device #1
1
(Low Order Device)
2
High Order
Device #2
1
(Low Order Device)
2
High Order
Device #3
1
(Low Order Device)
2
Red Arousal Safety bit.ly/48F9A5 2008 Child Innocence
(Youthful
Discoveries)
Child Innocence
(Youthful Discoveries)
+ Incongruity Ironic Persona (Adult
Acting Child)
Ironic Temperament
(Miscast
Temperament)
Ironic Persona (Adult
Acting Child)
Yellow Arousal Safety bit.ly/553ql8 2009 Fear & Anxiety Relief
(Narrow Escape)
Fear & Anxiety Relief
(Narrow Escape)
+ Incongruity Exaggerated Outcomes
(Exaggerated Stories)
Understatement
(Unrattled)
Blue Arousal Safety bit.ly/NyvftI 2011 Fear & Anxiety Relief
(Narrow Escape)
Fear & Anxiety Relief
(Narrow Escape)
+ Incongruity Bizarre Substitution
(Anthro-pomorphism)
Conceptual Surprises
(Absurd Chain
Reaction)
Unusual Personification
(Unlikely Animal
Behavior)
Exaggerated Outcomes
(Exaggerated
Stories)
White Arousal Safety bit.ly/2rqIR7r 2011 Hysteria (Extreme
Screaming)
Hysteria (Extreme
Screaming)
+ Incongruity Overreaction (Over
Intense)
Visual Surprise (Creature
Appearance)
Nonsense (Ignorance)
Red Arousal Safety bit.ly/14IKbn 2009 Exercising Improprieties
(Recalcitrance)
Society Irreverence (Rule
Breaking)
+Disparagement Overreaction (Over
Intense)
Mocked Peculiarities
(Background
Mockery)
Yellow Arousal Safety bit.ly/N1km2O 2008 Offensive Behaviors
(Bleeped Language)
Offensive Behaviors
(Repulsive Behaviors)
+Disparagement Mocked Peculiarities
(Illusory Superiority)
Society Satire
(Parodies)
Stereotyping
(Stereotyped
Professions)
Blue Arousal Safety bit.ly/GIGew2 2012 Fear & Anxiety Relief
(Narrow Escape)
Hysteria (Extreme
Screaming)
+Disparagement Unlucky Happenstance
(Unforeseen
Consequences)
Unanticipated Spoiler
(Unexpected
Danger)
Society Satire (Cultural
Nuance)
Unanticipated Spoiler
(Unexpected
Danger)
White Arousal Safety bit.ly/2rqVTlv 2009 Displaced Irritation
(Ending the
Annoyance)
Displaced Irritation
(Ending the
Annoyance)
+Disparagement Lofty Conquest
(Arrogant
Knockdowns)
Deserved Repercussions
(Paybacks)
1) One of 41 high order classifications derived in Figure 1 for humor type sorting.
2) Lower level classifications derived from comic devices referenced in Berger (1993) Buijzen and Valkenburg (2004).
168 Journal of Marketing Theory and Practice

classification, while permitting the reviewers a comic
device description readily evident in the skit, each
reviewer qualified the videos with a high order and
low order descriptor as displayed in Table 3.
Where identical or closely related selections were
made by both reviewers at the high order level, the
video was coded for second stage qualification. To
maintain inter-reliability across reviewers, only
matched evaluations were included in this set. This
eliminated 316 videos from the original 2,762. On the
second stage, reviewers then examined the remaining
2,446 for humor dominance and placement.
Specifically, each reviewer examined if the advertise-
ment was humor, image, or information dominant.
Mismatches in this case reduced the set of qualified
videos to 2,024. Finally, for image and information
dominant videos, reviewers examined the placement
of humor as initial, embedded, or closing. A further
reduction of 247 videos from mismatched evaluations
resulted in a final set of 1,777 humorous videos selected
for testing. Shown in Table 4 are sample video links
used to categorize each social video along dimensions
of dominance and placement.
In qualifying non-humorous videos, reviewers fol-
lowed a similar approach of selecting low order descrip-
tors subsequently categorized under higher order
classifications for matched assessments. Shown in
Table 5 is an example of the coding process used
along with sample video links across each non-humor
type (creative fun, inspiration, and amazement) and
product type. From the originally identified list of
1,286 candidate videos, an elimination of 152 videos
from mismatched reviewer evaluations resulted in a
final set of 1,134 non-humorous videos.
From this set of humorous and non-humorous
videos, engagement statistics were then recorded
using the number of comments relative to the number
of views. This ratio of comments to views allowed for
an improved measure of engagement as very high
views could otherwise skew the average number of
total comments.
Measurement of advertising attitudes was based on
surveys hosted on SurveyGizmo and taken by 584
MBA and undergraduate students from two Florida-
based universities. Each respondent was asked to evalu-
ate from four to eight different videos in an attempt to
measure all qualified videos. A distribution of age and
gender is shown in Table 6. Although the distribution is
skewed more heavily toward a millennial age group in
comparison to overall YouTube viewers (Blattberg 2015),
this sample is highly representative of age groups most
likely to view entertaining content (Fitzgerald 2015).
The choice of advertising attitudes in the surveys is
consistent with prior research in advertising humor
Table 4
Sample Videos for Humor Message Taxonomy
Example Video Example Video
URL Year URL Year
Intention Relatedness Humor Position
Humor Dominant Initial
Red bit.ly/te2YH 2008 Red bit.ly/dmnqPx 2010
Yellow bit.ly/vMVOtY 2009 Yellow bit.ly/jwGti 2008
Blue bit.ly/9MOEO 2008 Blue bit.ly/1Ma17c3 2007
White bit.ly/1MLb0IY 2015 White bit.ly/uTl3D 2008
Image Dominant Embedded
Red bit.ly/d1OvDI 2010 Red bit.ly/1k5BSLb 2012
Yellow bit.ly/KIzEwP 2012 Yellow bit.ly/JE2zl 2008
Blue bit.ly/7gXo54 2009 Blue bit.ly/1cH80zg 2013
White bit.ly/1isij6n 2012 White bit.ly/dzsblD 2010
Information Dominant Closing
Red bit.ly/1pF0eMn 2011 Red bit.ly/8YTLVS 2010
Yellow bit.ly/cZfXae 2010 Yellow bit.ly/g2KogO 2011
Blue bit.ly/PnHRVQ 2008 Blue bit.ly/1PtkTxL 2015
White bit.ly/95czGn 2010 White bit.ly/1QpQ61R 2015
Winter–Spring 2018 169

Table 5
Example Coding Process and Videos for Non-Humor over Product Grid
Example Video Reviewer #1 Selections Reviewer #2 Selections
URL Year
High Order Concept #1
1
(Low Order Concept)
2
High Order Concept #2
1
(Low Order Concept)
2
High Order Concept #1
1
(Low Order Concept)
2
High Order Concept #2
1
(Low Order Concept)
2
Non-Humor Types
Inspirational
Red bit.ly/2r2gYz7 2012 Inspirational (Pushing Limits) Inspirational (Overcoming
Obstacles)
Inspirational (Pushing Limits)
Yellow bit.ly/2pSFwu8 2017 Inspirational (Overcoming
Obstacles)
Inspirational (Overcoming
Obstacles)
Blue bit.ly/2iglsTy 2012 Inspirational (Pushing Limits) Inspirational (Overcoming
Obstacles)
Inspirational (Pushing Limits) Inspirational (Overcoming
Obstacles)
White bit.ly/2oH16VS 2017 Inspirational (Reassurance) Inspirational (Reassurance) Inspirational (Overcoming
Obstacles)
Amazement
Red bit.ly/2qYc3UI 2009 Amazement (Natural Beauty) Amazement (Natural Beauty) Amazement (Human Potential)
Yellow bit.ly/1du5SXA 2010 Amazement (Masterful
Craftsmanship)
Amazement (Human Potential) Amazement (Masterful
Craftsmanship)
Blue bit.ly/2aRM9XX 2012 Amazement (Human Potential) Amazement (Human Potential) Inspirational (Pushing Limits)
White bit.ly/1qUgVkY 2009 Amazement (Masterful
Craftsmanship)
Amazement (Masterful
Craftsmanship)
Creative Fun
Red bit.ly/2aS2fAN 2010 Creative Fun (Performance) Creative Fun (Concept Imagery) Creative Fun (Concept Imagery) Creative Fun (Performance)
Yellow bit.ly/2aNnMJs 2009 Creative Fun (Participative Skits) Creative Fun (Performance) Creative Fun (Performance)
Blue bit.ly/2aP8enZ 2008 Creative Fun (Concept Imagery) Creative Fun (Participative Skits) Creative Fun (Concept Imagery) Creative Fun (Performance)
White bit.ly/1WXKNMi 2015 Creative Fun (Participative Skits) Creative Fun (Participative Skits) Creative Fun (Performance)
1) Higher Order Concepts 2) Lower Order Concepts
Inspirational Reassurance Overcoming Obstacles Pushing Limits
Amazement Natural Beauty Masterful Craftsmanship Human Potential
Creative Fun Performance Concept Imagery Participative Skits
170 Journal of Marketing Theory and Practice

(Cochran 2004; Flaherty et al. 2004; Zhang and
Zinghan 2006) along with these and other studies
(Unger 1995) suggesting a high correlation between
advertising attitudes and brand attitudes. Using the
same four-item, seven interval semantic differential
scale adopted by these studies, respondents were
asked to describe if their reaction to the video was 1)
pleasant/unpleasant, 2) likeably/unlikeable, 3) inter-
ested/bored, and 4) good/bad. The reliability of the
instrument was validated with a Cronbach Alpha of
0.93.
Finally, to measure the comparisons implied in
Hypotheses 1–6, independent-samples t-tests in SPSS
were conducted to determine if the associated popula-
tion means for advertising attitudes and posted
comments are significantly different. Shown in
Table 7 and 8are the comparative engagement and
attitude results found across Speck’s(1991) taxonomy
and the product grid, respectively.
RESULTS
A summary of all hypothesized results is shown in
Table 9. Beginning with H
1
, humor was found to out-
perform non-humor in advertising attitude, thereby
supporting H
1a
. But non-humor significantly outper-
forms humor in generated comments, thereby rejecting
H
1b
.
As displayed in Figure 2, humor is more favorably
received (i.e., scores higher in advertising attitude) than
the combination of all forms of non-humorous
entertainment; but non-humorous social videos have
more comments per viewed episode. In-depth exami-
nation of the different types of non-humorous enter-
tainment shows that videos seen as astonishing (e.g.,
through amazing views of natural beauty, masterful
craftsmanship, or extraordinary human potential)
underperformed humor in advertising attitudes.
Videos perceived as inspirational (e.g., heartfelt stories
of overcoming obstacles), on the other hand, signifi-
cantly outperformed the average of humorous videos
on engagement. Surprisingly, the number of videos
classified as inspirational were very low. The bulk of
non-humorous entertainment videos involved crea-
tive fun as displayed in concept imagery, perfor-
mances, or audience participative skits. Humor
significantly outperformed this category in advertising
attitudes but not in the percent of comments gener-
ated by audience viewers. This may suggest that
humor is not as inviting to ongoing dialogs as is
non-humorous entertainment. When audiences wit-
ness heartfelt moments, daring adventures or playful
exercises, for example, they may be more willing to
share their personal stories.
Despite the strong global literature support for the
dominance of incongruity-resolution, Figure 3 suggests
that the weak separation of comic wit (based solely on
incongruity-resolution) from the other types of humor
does not support this conclusion. In fact, no significant
differences were found in the advertising attitudes or
generated comments of comic wit when compared to
all other types of humor. Therefore H
2a
and H
2b
are not
supported. This would suggest that all three humor
theories (incongruity, superiority, and relief) have
merit. Moreover, the fact that incongruity-resolution
is associated with almost all types of humor suggests
that its cognitive style of amusement is matched or
overwhelmed by arousal-safety and disparagement
effects.
As displayed in Figure 4, humor dominant videos
significantly outperformed image and information
dominant videos in engagement but not in advertising
attitudes. Therefore, H
3b
is supported, but H
3a
is
rejected. With regard to humor placement, Figure 5
reveals no significant differences in attitudes or engage-
ment when comparing humor placed at the beginning
of, throughout, or at the end of videos. Therefore,
neither H
4a
nor H
4b
are supported.
Table 6
Age and Gender Distribution of Surveyed Respondents
0%
10%
20%
30%
40%
50%
60%
70%
< 20 20-29 30-39 40-49 >49
Percent of Age Distrib ution
Male Female
Winter–Spring 2018 171

Table 7
Comparative Engagement and Attitude Results across Speck’s(1991) Taxonomy
n = Mean
1
n = Mean
1
n = Mean
1
n = Mean
1
n = Mean
1
Entertaining Ads Humor Non-Humor Amazement Inspiration Creative Fun
Ad Attitude 1177 4.74 485 4.56 119 4.15 187 4.91 179 4.46
Comments as % Views 1777 0.078% 1134 0.103% 148 0.088% 252 0.098% 734 0.108%
Significant Differences (ns - not significant)
Humor vs. Non-Humor Humor vs. Amazement Humor vs. Inspiration Humor vs. Creative Fun
Ad Attitude t = 2.0, p < 0.05 t = 3.4, p < 0.01 ns t = 2.0, p < 0.05
Comments as % Views t = –5.8, p < 0.001 ns t = –2.5, p < 0.05 t = –5.6, p < 0.001
n = Mean
1
n = Mean
1
n = Mean
1
n = Mean
1
n = Mean
1
Humor Types Comic Wit (HMT1) Sentimental Humor (HMT2) Satire (HMT3) Sentimental Comedy (HMT4) Full Comedy (HMT5)
Ad Attitude 484 474 38 4.53 676 4.66 186 4.85 393 4.83
Comments as % Views 484 0.080% 38 0.082% 676 0.075% 186 0.089% 393 0.075%
Significant Differences (ns - not significant)
HMT1 vs. Other Humor HMT2 vs. Other Humor HMT3 vs. Other Humor HMT4 vs. Other Humor HMT5 vs. Other Humor
Ad Attitude ns ns ns ns ns
Comments as % Views ns ns ns ns ns
HMT1 vs. Non- Humor HMT2 vs. Non-Humor HMT3 vs. Non-Humor HMT4 vs. Non-Humor HMT5 vs. Non-Humor
Ad Attitude ns ns ns t = 2.0, p < 0.05 t = 2.4, p < 0.05
Comments as % Views t = –4.3, p < 0.001 ns t = –4.3, p < 0.001 ns t = –4.6, p < 0.001
n = Mean
1
n = Mean
1
n = Mean
1
Intention Relatedness Humor Dominant Image Dominant Information Dominant
Ad Attitude 1006 4.77 346 4.67 413 4.68
Comments as % Views 1006 0.069% 346 0.084% 413 0.095%
Significant Differences (ns - not significant)
Humor vs. Image Dominant Humor vs. Information Dominant Image vs. Information Dominant
172 Journal of Marketing Theory and Practice

Ad Attitude ns ns ns
Comments as % Views t = 2.3, p < 0.05 t = 4.5, p < 0.001 ns
n = Mean
1
n = Mean
1
n = Mean
1
Humor Position (for message dominance only) Initial Embedded Closing
Ad Attitude 36 4.48 623 4.68 100 4.70
Comments as % Views 36 0.086% 623 0.093% 100 0.076%
Significant Differences (ns - not significant)
Initial vs. Embedded Initial vs. Closing Embedded vs. Closing
Ad Attitude ns ns ns
Comments as % Views ns ns ns
1) Based on Likert score (1-7, 7= strongly agree)
Winter–Spring 2018 173

Table 8
Comparative Engagement and Attitude Results across Product Grid
Emotionally Motivated Rationally Motivated
Red Yellow White Blue Total
n = Mean
1
n = Mean
1
Significant Difference n = Mean
1
n = Mean
1
Significant Difference n = Mean
1
Humor
Ad Attitude 180 4.82 573 4.62 ns 596 4.85 428 4.70 ns 1777 4.74
Comments as % Views 180 0.061% 573 0.081% t=3.0; p<0.01 596 0.070% 428 0.091% t=3.3; p<0.01 1777 0.078%
Non-Humor
Ad Attitude 130 4.57 83 4.40 ns 149 4.69 123 4.48 ns 485 4.56
Comments as % Views 310 0.084% 237 0.112% t=2.8; p<0.01 386 0.115% 201 0.098% ns 1134 0.103%
Total
Ad Attitude 310 4.71 656 4.59 ns 745 4.82 551 4.65 ns 2262 4.70
Comments as % Views 490 0.076% 810 0.090% ns 982 0.088% 629 0.093% ns 2911 0.088%
High Involvement Low Involvement Total
Red White Yellow Blue
n = Mean
1
n = Mean
1
Significant Difference n = Mean
1
n = Mean
1
Significant Difference n = Mean
1
Humor
Ad Attitude 180 4.82 596 4.85 ns 573 4.62 428 4.70 ns 1777 4.74
Comments as % Views 180 0.061% 596 0.070% ns 573 0.081% 428 0.091% ns 1777 0.078%
Non-Humor
Ad Attitude 130 4.57 149 4.69 ns 83 4.40 123 4.48 ns 485 4.56
Comments as % Views 310 0.084% 386 0.115% t=4.3, p<0.001 237 0.112% 201 0.098% ns 1134 0.103%
Total
Ad Attitude 310 4.71 745 4.82 ns 656 4.59 551 4.65 ns 2262 4.70
Comments as % Views 490 0.076% 982 0.088% t=2.7, p<0.01 810 0.090% 629 0.093% ns 2911 0.088%
1) Mean value of Likert scale (1-7; 7 = strongly agree)
174 Journal of Marketing Theory and Practice

Table 9
Summary of Hypotheses Testing
Comparative Means from Independent-Samples T Test
Test Hypotheses Comparison Result
H
1a
Supported H
1a
: Audience favorability toward social videos perceived
as humorous is greater than the interest of serious
entertainment advertisements when measured across
advertising attitudes.
Humor vs. Non-Humor 2.0*
H
1b
Unsupported H
1b
: Audience engagement in social videos perceived as
humorous is greater than the engagement of serious
entertainment advertisements when measured across
comments generated.
Humor vs. Non-Humor -5.8***
H
2a
Unsupported H
2a
: Audience favorability toward social videos perceived
as comic wit will outperform all other forms of humor
when measured across advertising attitudes.
Comic Wit vs. All Other Humor ns
H
2b
Unsupported H
2b
: Audience engagement of social videos perceived as
comic wit will outperform all other forms of humor
when measured across comments generated.
Comic Wit vs. All Other Humor ns
H
3a
Unsupported H
3a
: Audience favorability toward social videos perceived
as humor dominant will outperform humor perceived as
message dominant when measured across advertising
attitudes.
Humor vs. Image Dominant ns
Humor vs. Information Dominant ns
H
3b
Supported H
3b
: Audience engagement of social videos perceived as
humor dominant will outperform humor perceived as
message dominant when measured across comments
generated.
Humor vs. Image Dominant 2.3*
Humor vs. Information Dominant 4.5***
H
4a
Unsupported H
4a
: Audience favorability toward social videos where
humor is embedded will outperform initial and closing
humor when measured across advertising attitudes.
Embedded vs. Initial ns
Embedded vs. Closing ns
H
4b
Unsupported H
4b
: Audience engagement in social videos where humor
is embedded will outperform initial and closing humor
when measured across comments generated.
Embedded vs. Initial ns
Embedded vs. Closing ns
H
5a
Unsupported H
5a
: Audience favorability toward social videos for low
involvement goods (yellow and blue) will outperform
humor shown for high involvement goods (red and
white) when measured across advertising attitudes.
Yellow (Low Involvement) vs. Red (High
Involvement)
ns
Blue (Low Involvement) vs. White (High
Involvement)
ns
H
5b
Supported H
5b
: Audience engagement in social videos for low
involvement goods (yellow and blue) will outperform
humor shown for high involvement goods (red and
white) when measured across comments generated.
Yellow (Low Involvement) vs. Red (High
Involvement)
3.3**
Blue (Low Involvement) vs. White (High
Involvement)
3.0**
H
6a
Unsupported H
6a
: Audience favorability toward social videos for
emotionally motivated goods (yellow and red) will
outperform humor shown for rationally motivated
goods (blue and white) when measured across when
measured across advertising attitudes.
Yellow (Emotional) vs. Blue (Rational) ns
Red (Emotional) vs. White (Rational) ns
H
6b
Unsupported H
6b
: Audience engagement in social videos for
emotionally motivated goods (yellow and red) will
outperform humor shown for rationally motivated
goods (blue and white) when measured across when
measured across comments generated.
Yellow (Emotional) vs. Blue (Rational) ns
Red (Emotional) vs. White (Rational) ns
H
7a
Supported H
7a
: The use of humor compared to non-humor will be
proportionally higher for yellow goods.
% Humor for Yellow Goods 71%
% Non-Humor for Yellow Goods 29%
H
7b
Supported H
7b
: The use of humor compared to non-humor will be
proportionally lower for red goods.
% Humor for Red Goods 37%
% Non-Humor for Red Goods 63%
* p< .05, ** p < .01; *** p< .001; ns - not significant
Winter–Spring 2018 175

To evaluate the impact that humor has on differ-
ent product types, Table 8 is arranged to permit pro-
duct type performance comparisons across the
motivation and involvement dimensions of Rossiter
and Percy’s(1997) brand attitude grid. As demon-
strated in Figure 6, the predicted outperformance of
humor aimed at low involvement (yellow and blue)
goods over humor aimed at high involvement (red
and white) goods is partially supported. For example,
humor aimed at low involvement goods significantly
outperforms that of high involvement goods but in
commentary only. No significant differences were
found in advertising attitudes. Therefore, H
5b
is sup-
ported, but H
5a
is rejected. Further examination of
Figure 6, however, shows no support for the bulk of
literature findings suggesting that humor’sgreatest
appeal is for emotionally motivated products.
Therefore, neither H
6a
nor H
6b
are supported. Finally,
the ratio of humor compared to non-humor is shown
to be proportionally higher for yellow goods and
proportionally lower for red goods. Therefore, both
H
7a
and H
7b
are supported.
CONCLUSIONS
These results answer the original four research ques-
tions in the following way. First, humor does outper-
form other forms of entertaining social content, but
only in attitudes. The reverse is found in the case of
engagement. This coupled with the disparity of results
between H
3a
and H
3b
and H
5a
and H
5b
answers the
second question in that many of the effects of humor
examined in traditional advertising settings may not be
Figure 2
Performance of Humorous and Non-Humorous Social Videos
Amazement
Creative
Fun
Ad Attitude (mean score)
1
4.85
n= 2911 social videos
Circle size
denotes number
of social vid eos.
Avg. Comment s per View
Avg.
0.091% 0. 099% 0. 107% 0.115%0.059% 0. 067% 0.075% 0. 083%
5.00
5.15
5.30
5.45
3.95
4.10
4.25
4.40
4.55
1) Bas ed on Like rt score (1 -7, 7= stro ngly agree)
* p<.05; ** p<.01; ***p<.001
Humor (61%)
Non-Humor (39%)
}t=5.8***
t=2.0*
t=2.5*
Inspiration
To ta l
Humor
t=3.4* *
t=2.0*
t=5.6***
To ta l
Non-
Humor
176 Journal of Marketing Theory and Practice

Figure 3
Performance of Social Videos by Humor Type
Ad Attitude (mean score)
1
4.85
n = 2911 social videos
Circle size
denotes number
of social vid eos.
Avg. Comment s per View
0.091% 0.099% 0. 107% 0.115%0.059% 0.067% 0.075% 0.083%
5.00
5.15
5.30
5.45
3.95
4.10
4.25
4.40
4.55
Comic Wit
(HMT1)
Satire
(HMT3)
Sentimental
Humor
(HMT2)
Full
Comedy
(HMT5)
To ta l
Non-
Humor
Sentimental
Comedy (HMT4)
t=4.6* **
1) Based on Likert score (1-7, 7= strongly agree)
* p<.05; ** p<.01; ***p<.001
t=2.4*
t=2.0*
t=4.3*** t=4.3***
Figure 4
Performance of Social Videos by Dominance
Ad Attitude (mean Likert score)1
4.85
n = 1777 s oc ial videos
Circle size
denotes number
of social vid eos.
Avg. Comments per View
0.091% 0. 099% 0. 107% 0.115%
0.059% 0.067%
0.079%
5.00
5.15
5.30
5.45
3.95
4.10
4.25
4.40
4.55
1) Based on Likert score (1-7, 7= strongly agree)
* p<.05; ** p<.01; ***p<.001
t=4.5***
Humor
Dominant
Image
Dominant
Information
Dominant
t=2.8*
Winter–Spring 2018 177

Figure 5
Performance of Social Videos by Humor Placement
Ad Attitude (mean Likert score)
1
4.85
n= 759 social videos
Circle size
denotes number
of social vid eos.
Avg. Comments per View
0.099% 0. 107% 0.115%0.059% 0.067% 0. 075% 0.083%
5.00
5.15
5.30
5.45
3.95
4.10
4.25
4.40
4.55
1) Based on Likert score (1-7, 7= strongly agree)
Initial
Embedded
Closing
Figure 6
Performance of Social Videos by Industry Type
Ad Attitude (mean Likert score)
1
4.85
Circle size
denotes number
of social vid eos.
Avg . Co mme nts p er Vi ew
0.091% 0.099% 0. 107% 0.115%0.059% 0.067% 0.075% 0. 083%
5.00
5.15
5.30
5.45
3.95
4.10
4.25
4.40
4.55
1) Based on Likert score (1-7, 7= strongly agree)
* p<.05; ** p<.01; ***p<.001
t=3.0**
Humor (61%)
Non-Humor (39%)
n= 2911 social videos
Red
Yellow
White
Blue
Red
t=2.8* *
Yellow
Blue White
t=3.3**
178 Journal of Marketing Theory and Practice

extendable to social engagement behaviors. This may
suggest that despite humor’s welcoming distraction
from an otherwise insignificant message, the distrac-
tion may actually dissuade social media participants
from imparting comments about a routine product
purchase. Third, observations made from Figures 2
through 5 suggest that humor intended by advertisers
generally mirrors the actual audience perceptions of
humor effectiveness. Some exceptions, however,
include the high attention given to the relatively low
performing satire form of humor (see Figure 3) and the
inordinately low attention given to the high perform-
ing inspirational form of non-humor (see Figure 2).
Finally, the level of humor interest does vary in some
cases over the span of audience involvement and moti-
vation associated with the advertised product. In parti-
cular, humor aimed at yellow and blue (low
involvement) goods outperforms that of red and
white (high involvement) goods, respectively, in gen-
erating comments.
MANAGERIAL IMPLICATIONS AND STUDY
LIMITATIONS
This study extends the research in humor by analyzing
its effectiveness in social media marketing. As brands
recognize the power of recasting their television com-
mercials and corporate videos on YouTube, marketers
can now examine the effectiveness of humor from
engagement statistics. Results of this research found
that commercials intended to be funny represent a
vast majority of the content garnering over 50,000
views. And since humor generally scores higher in
advertising attitudes than the average of high perform-
ing content involving serious entertainment, marketers
should not shy away from its usage.
A potential drawback of using humor, however, may
be its inherent one-way form of entertainment.
Examination of social video themes involving inspira-
tion or astonishment suggests that some forms of ser-
ious entertainment may be more conducive to
audience conversation. Content marketing practi-
tioners often refer to social content having to be
remarkable (Halligan and Shah 2010) or talk-worthy
(Cohen 2013; Fay et al. 2016) so as to incentivize
their targeted audiences to make their voices heard.
This may suggest that humor should be mixed with
more serious forms of entertainment over time so as
to permit more interactive conversations.
Some limitations are acknowledged in this study
regarding its universality. Since the analyzed content
was restricted to television commercial recasts hosted
on YouTube, audience experiences may differ from that
of print magazines and live television viewer.
Furthermore, studies of humor suggest that Speck’s
(1991) taxonomy be examined for situational modera-
tors like national culture. Another limitation relates to
methods used in classifying and scoring humor.
Although the number of comments generated serve as
an indication of engagement, the metric was not used
to determine if the comments were predominantly
good or bad. Much like that seen in politic commen-
tary, many comments are unintelligible rantings that
do not serve the brand well. Further research is encour-
aged that considers other comment-based metrics that
suggest overall favorability or disapproval of the mes-
sage. Finally, future research is encouraged to see if the
relative proportions of humor and serious entertain-
ment stay constant. The recent wave of heartfelt story-
telling, for example, suggests that the current
dominance of humor may not last.
REFERENCES
Alden, Dana L., and Wayne D. Hoyer (1993), “An
Examination of Cognitive Factors Related to
Humorousness in Television Advertising,”Journal of
Advertising, 22 (2), 29–37.
______, Ashesh Mukherjee, and Wayne D. Hoyer (2000), “The
Effects of Incongruity, Surprise and Positive Moderators
on Perceived Humor in Television Advertising,”Journal
of Advertising, 29 (2), 1–15.
Barry, James M. (2015), Social Content Marketing for
Entrepreneurs, New York: Business Expert Press.
______, and John Gironda (2017a), “Developing Social Selling
Influence: An Archetypal Examination of Content
Strategies and Influence Tactics,”Journal of Selling, 1 (18).
______, and ______ (2017b), “A Dyadic Examination of
Inspirational Factors Driving Social Media Influence,”
Journal of Marketing Theory and Practice,26(1–2).
______, and Sandra Graça (2013), “A Cross-Cultural Typology
of Advertising Humor,”2013 Cross Cultural Research
Conference. Antigua, Guatemala.
______, and D. Hale. (2014), “Humor Determinants and
Relevance in High Engagement Social TV Ads, in Ideas
in Marketing: Finding the New and Polishing the Old,K.
Kubacki ed., Proceedings of the 2013 AMS Annual
Conference. Springer.
Berger, Arthur Asa (1993) An Anatomy of Humor. New
Brunswick, NJ: Transaction Books.
Blattberg, Eric (2015). “The Demographics of YouTube,”
(available athttp://bit.ly/2s9Otot).
Winter–Spring 2018 179

Buijzen, Moniek, and Patti M. Valkenburg (2004),
“Developing a Typology of Humor in Audiovisual
Media,”Media Psychology, 6, 147–67.
Cline, Thomas W., and James J. Kellaris (2007), “The
Influence of Humor Strength and Humor-Message
Relatedenss on Ad Memorability,”Journal of Advertising,
36 (1), 55–67.
Cochran, Lucy (2004), “Product Involvement and Humour in
Advertising: An Australian Empirical Study,”Journal of
Asia Pacific Marketing, 3 (1), 68–88.
Cohen, Heidi (2013), “7 Experiential Content Marketing
(With Examples),”[(available at http://bit.ly/2f2qrlc).
Djambaska, Aneta, Ilijana Petrovska, and Elena Bundalevska
(2016), “Is Humor Advertising Always Effective?
Parameters for Effective Use of Humor in Advertising,”
Journal of Management Research,”8 (1), 1–19.
Eisend, Martin (2009), “A Meta-Analysis of Humor in
Advertising,”Journal of the Academy of Marketing Science,
37 (2), 191–203.
______, (2011), “How Humor in Advertising Works: A Meta-
Analytic Test of Alternative Methods,”Marketing Letters,
22 (2), 115–32.
Fay, Brad, Grame Hutton, and David Poltrack (2016),
“Driving Effectiveness with Talkworthy Advertising,”
[(available at http://bit.ly/2fcK8K5).
Frasco, Stephanie (2015). “Why Social Video Is Important
For Your Engagement,”(available at http://bit.ly/
2e5D2Tl).
Fitzgerald, Tony (2015). “Who’s Watching What on
YouTube,”(available at http://bit.ly/2s9HPOX).
Flaherty, Karen, Marc G. Weinberger, and Charles S. Gulas
(2004), “The Impact of Perceived Humor, Product
Type, and Humor Style in Radio Advertising,”
Journal of Current Issues and Research in Advertising,26
(1), 25–36.
Halligan, Brian, and Dharmesh Shah (2010). Inbound
Marketing: Getting Found Using Google, Social Media, and
Blogs,: Wiley.
Hatzithomas Leonidas, Christina Boutsouki, and Yorgos
Zotos (2009), “The Effects of Culture and Product
Type on the Use of Humor in Greek TV Advertising:
An Application of Speck’s Humorous Message
Taxonomy,”Journal of Current Issues and Research in
Advertising 31 (1), 43–61.
O’Neill, Megan (2016), “The 2016 Social Video Forecast,”
(available at http://bit.ly/28QBVHV).
Petty, Richard E., and John T. Cacioppo (1986). The
Elaboration Likelihood Model of Persuasion, New York:
Springer Verlag.
______, ______, and David Schumann (1983), “Central and
Peripheral Routes to Advertising Effectiveness: The
Moderating Role of Involvement,”Journal of Consumer
Research,10(4),135–46.
Rossiter, John. R., and Larry Percy (1997), Advertising
Communications and Promotion Management, Boston: Irwin/
McGraw Hill.
Ruedlinger, Ben (2012), Does Video Length Matter?”(avail-
able at http://wi.st/2cYLvwi).
Scott, Cliff, David M. Klein, and Jennings Bryant (1990),
‟Consumer Response to Humor in Advertising: A Series
of Field Studies Using Behavioral Observation,”Journal of
Consumer Research, 16 (4), 498–501.
Speck, Paul Surgi (1991), “The Humorous Message Taxonomy: A
Framework for the Study of Humorous Ads,”Current Issues
& Research in Advertising,13(1–2), 1–44.
Spotts, Harlan E., Marc G. Weinberger, and Amy L. Parsons
(1997), “Assessing the Use and Impact of Humor on
Advertising Effectiveness: A Contingency Approach,”
Journal of Advertising, 26 (3), 17–32.
Syndacast (2014), “Video Marketing Statistics & Trends
2015,”(available at http://bit.ly/1xHQOhO).
Toncar, Mark F. (2001), “The Use of Humour in Television
Advertising: Revisiting the US-UK Comparison,”
International Journal of Advertising, 20 (4), 521–39.
Unger, Lynette S. (1995), “Observations: A Cross-Cultural Study
on the Affect-Based Model of Humor in Advertising,”
Journal of Advertising Research, 35 (January/February), 66–71.
Weinberger, Marc G., and Charles S. Gulas (1992), “The
Impact of Humor in Advertising: A Review,”Journal of
Advertising, 21 (4), 35–59.
______, and Harlan E. Spotts, (1989), “Humor in U.S. versus
U.K. TV Commercials: A Comparison Guide,”Journal of
Advertising, 18 (2), 39–44.
______, ______, Leland Campbell, and Amy L. Parsons (1995),
“The Use and Effect of Humor in Different Advertising
Media,”Journal of Advertising Research, 35 (3), 44–57.
Zhang, Yong (1996), “The Effect of Humor in Advertising: An
Individual Difference Perspective,”Psychology &
Marketing, 13 (6), 531–45.
______, and George M. Zinkhan (2006). “Responses to
Humorous Ads: Does Audience Involvement Matter?”
Journal of Advertising, 35 (4), 113–27.
180 Journal of Marketing Theory and Practice