Top Banner
8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 1/15 This article was downloaded by: [University of Groningen] On: 26 November 2014, At: 07:48 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-4 Mortimer Street, London W1T 3JH, UK Journal of Advertising Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/ujoa20 The Influence of Humor Strength and Humor—Message Relatedness on Ad Memorability: A Dual Process Model Thomas W. Cline Ph.D. a  & James J. Kellaris Ph.D. b a  Alex G. McKenna School of Business, Economics, and Government, St. Vincent College b  College of Business, University of Cincinnati Published online: 04 Mar 2013. To cite this article: Thomas W. Cline Ph.D. & James J. Kellaris Ph.D. (2007) The Influence of Humor Strength and Humor—Message Relatedness on Ad Memorability: A Dual Process Model, Journal of Advertising, 36:1, 55-67, DOI: 10.2753/JOA0091-3367360104 To link to this article: http://dx.doi.org/10.2753/JOA0091-3367360104 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed b Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with pr sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, c expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection n relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduct redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-cond
15

Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

Jun 02, 2018

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 1/15

This article was downloaded by: [University of Groningen]On: 26 November 2014, At: 07:48Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-4Mortimer Street, London W1T 3JH, UK

Journal of AdvertisingPublication details, including instructions for authors and subscription information:

http://www.tandfonline.com/loi/ujoa20

The Influence of Humor Strength and Humor—MessageRelatedness on Ad Memorability: A Dual Process ModelThomas W. Cline Ph.D.

a & James J. Kellaris Ph.D.

b

a Alex G. McKenna School of Business, Economics, and Government, St. Vincent College

b College of Business, University of Cincinnati

Published online: 04 Mar 2013.

To cite this article: Thomas W. Cline Ph.D. & James J. Kellaris Ph.D. (2007) The Influence of Humor Strength and Humor—Message

Relatedness on Ad Memorability: A Dual Process Model, Journal of Advertising, 36:1, 55-67, DOI: 10.2753/JOA0091-3367360104

To link to this article: http://dx.doi.org/10.2753/JOA0091-3367360104

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in thepublications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations orwarranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed bTaylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with prsources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, cexpenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection n relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproductredistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-cond

Page 2: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 2/15

 Journal of Advertising, vol. 36, no. 1 (Spring 2007), pp. 55–67.

© 2007 American Academy of Advertising. All rights reserved.

ISSN 0091-3367 / 2007 $9.50 + 0.00.

DOI 10.2753/JOA0091-3367360104

Following the early research on humor in advertising (e.g.,

Duncan 1979; Speck 1987; Sternthal and Craig 1973), numer-

ous studies have attempted to explain the relationship between

humor and ad outcomes (e.g., Alden and Hoyer 1993; Alden,

Mukherjee, and Hoyer 2000; Chattopadhyay and Basu 1990;Cline, Altsech, and Kellaris 2003; Krishnan and Chakravarti

2003; Lee and Mason 1999; Spotts, Weinberger, and Parsons

1997; Weinberger and Campbell 1991). Nevertheless, a lack of

systematic empirical results contrasts with humor’s widespread

use and an intuitive sense among advertising practitioners

that humor enhances ad persuasion (Madden and Weinberger

1984). Accordingly, the goal of our research is to provide a

clearer picture of humor’s role in advertising by examining

important stimulus factors and personality traits that shape

the effects of humorous appeals on consumers’ recall, as well

as the processes underlying such effects.

HUMOR AND ATTENTION

Although some of what is currently known about humor’s ef-

fects on advertising response may be equivocal, a strong case

can be made for humor’s impact on attention. The majority

of studies conducted in industry as well as in laboratory set-

tings bear this out. Madden and Weinberger (1982) find that

humorous magazine ads outperform nonhumorous ads on three

common surrogates for attention. Weinberger et al. (1995)

find evidence that humor is directly linked to attention and

recognition. Spotts, Weinberger, and Parsons (1997) show

that humor enhances initial attention, aids brand recall, and

holds attention. In a laboratory setting, Speck (1987) finds

that humorous ads outperform nonhumorous ads on sustained

attention, even after controlling for initial attention.

HUMOR STRENGTHAlthough some prior studies examined effects of humor’s pres-

ence (versus absence) in an ad (e.g., Lee and Mason 1999), the

impact of humor should depend in part on its strength (e.g., 

Alden, Mukherjee, and Hoyer 2000). Our use of the term humor

 strength refers to the extent of humor elicitation provoked by an

ad (Wyer and Collins 1992). Humor strength can be thought

of as “how funny” an ad is, rather than merely referring to

whether or not an ad employs humor (e.g., Elpers, Mukherjee,

and Hoyer 2004). Hence, in the present research, we vary the

strength of humor rather than its mere presence/absence.

HUMOR RELATEDNESS

Weinberger and Gulas (1992) argue that controlling for the

relatedness of humor makes experimental findings unanimous

in their support of humor’s positive impact on attention. In

fact, the relatedness of humor to the product or message may

also be a strong predictor of the success of an ad. Madden

(1982) finds that a radio commercial with product-related

humor is perceived as more interesting than one in which the

humor is unrelated to the product. Weinberger and Campbell

(1991) define related humor as being linked to the product and

the fabric of the commercial. Results of their study show that

related humor offers recall advantages over unrelated humor

for high-involvement/feeling goods. Weinberger et al. (1995)

THE INFLUENCE OF HUMOR STRENGTH AND HUMOR–MESSAGE

RELATEDNESS ON AD MEMORABILITY

A Dual Process Model

Thomas W. Cline and James J. Kellaris

ABSTRACT: This research examines contingencies that shape the effects of humorous appeals on consumers’ recall ofads, as well as the processes underlying such effects. Results of experimentation show that ads are more memorable when

humor is both strong and related to the message, and this interaction is mediated by attention and mood. Stronger humorappeals also induce higher recall among individuals with a high “need for humor” (NFH).

Thomas W. Cline (Ph.D., University of Cincinnati) is an associate

professor of marketing, Alex G. McKenna School of Business, Eco-

nomics, and Government, St. Vincent College.

 James J. Kellaris (Ph.D., Georgia State University) is a professor of

marketing, College of Business, University of Cincinnati.

The authors thank Karen Machleit and Steve Posavac for their helpful

comments and suggestions on earlier versions of this work. This research

was supported in part by a CoB Research Fellowship at the University of

Cincinnati. A portion of the work was conducted while the second author

was visiting at Bond University, Queensland, Australia.

Page 3: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 3/15

56  The Journal of Advertising 

focus specifically on thematic relatedness and find a positive ef-

fect for related humor. Spotts, Weinberger, and Parsons (1997)

use Speck’s (1991) typology to categorize humor relatedness.

They suggest that the most effective type of relatedness to use

in an ad depends on the product category. Speck (1987, 1991)

outlines a broad typology of humor that incorporates the re-

latedness of humor in an ad on three levels: (1) pragmatic, (2)thematic, and (3) structural. Pragmatic relatedness refers to

the hierarchy of the humor–message relationship (i.e., humor-

dominant or nonhumorous). Semantic (thematic) relatedness

describes the relationship between the humor and the product.

Structural (syntactic) relatedness refers to the syntactic place-

ment of the humor within the ad, that is, whether the humor

is meaningful to the message.

In one of the few laboratory studies to address the issue

of humor relatedness to brand claims, Krishnan and Chakra-

varti (2003) draw on Alba, Hutchison, and Lynch (1991) and

Meyers-Levy (1991) to define humor relevance as the degree of

pertinence of the execution to the brand claims. This definition

dovetails with Heckler and Childers’s (1992) conceptualiza-tion of relevancy as material pertaining directly to the meaning

of the primary message (e.g., brand claims). Krishnan and

Chakravarti (2003) demonstrate that the relevancy of humor

to the brand claims can positively influence memory for brand

claims. They argue that humorous executions strongly linked

to some or all other ad components (e.g., brand claims) produce

facilitation effects.

Kellaris, Cox, and Cox (1993) provide a conceptual frame-

work for understanding humor’s attention-gaining value

on advertising outcomes. They find that attention-gaining

background music enhances message reception when the

music evokes message-congruent (as opposed to incongruent)

thoughts. By analogy, humor may operate similarly. Kellaris,

Cox, and Cox’s (1993) notion of congruity corresponds with

Heckler and Childers’s (1992) concept of relevancy. Thus, it

seems likely that humor’s attention-gaining mechanisms (i.e.,

humor strength) translate into positive effects for recall, so

long as the humor is linked to the brand claims. In contrast,

incidental humor (i.e., humor that is unrelated to the brand

claims) could actually inhibit brand-claims recall. In the latter

case, all that is recalled is the funny part of the ad. In sum-

mary of our expectations, we anticipate that humor–message

relatedness will moderate the influence of humor strength

such that:

H1: The impact of humor strength on message claims recall

will be more positive when humor–message relatedness is high

than when it is low.

H2: The joint impact of humor strength and humor–message

relatedness on message claims recall will be mediated by

 attention to the ad.

MEDIATING ROLE OF MOOD

Research shows that a positive mood can facilitate recall (Isen

1987) and lead to greater receptiveness to persuasive commu-

nications (Galizio and Hendrick 1972). Bower and colleagues

(Bower, Gilligan, and Monteiro 1981) provide insight into the

processes through which mood states can bias cognitive activ-

ity. They demonstrate that people better remember material

that is congruent with their mood states at the time of encod-

ing (“mood-congruent hypothesis”). These mood-congruent

memory effects are echoed in Isen (1978), who found that

people in a positive mood during retrieval recalled more posi-

tive traits, whereas those in a negative mood recalled more

negative traits. Isen (1989) goes beyond the notion of mood-

congruent message elaboration, arguing that positive mood

states may broaden cognitive organization (i.e., they encourage

people to categorize a wider range of stimuli together) and,

consequently, promote more integrated and flexible thinking.

This cognitive flexibility, in turn, may enhance the processing

of verbal content in an ad. Thus, it is possible that positivemoods increase elaboration and subsequent recall of the mes-

sage. Isen (1989) also suggests that positive affect may promote

the use of heuristics or reliance on peripheral cues.

Humor is an ad feature that is likely to induce good moods.

Littmann (1983) cites joy as one of the principal effects of

humor. Wicker et al. (1981) find a number of emotion-related

scales to be significantly correlated with perceived funniness.

Olson and Roese (1995) find support for the contention that

perceivers use their own reactions to humorous stimuli (e.g.,

mirth), together with information about the environment, to

infer the emotion-eliciting qualities of humor. Thus, if par-

ticipants express joy, they are likely to infer funniness. O’quin

and Aronoff (1981) find that humor lessens self-reported ten-sion and increases enjoyment of a task. Smith (1993) argues

that humor in an ad can act to enhance the consumer’s mood,

and that this mood influences how individuals process the ad.

Madden, Allen, and Twible (1988) distinguish between affec-

tive reaction and cognitive evaluation in advertising response

data. They find that a humorous ad generates significantly

more nonevaluative positive affect than its nonhumorous

counterpart. Machleit and Wilson (1988) provide additional

support for this affective/evaluative dichotomy; they use the

term “emotional feelings” to describe nonevaluative, affective

responses. In the present study, we follow Bruner and Hensel

(1996, p. 435) in using the label “mood” to measure the non-

evaluative affect that has been induced by an object in a person

(see also Batra and Ray 1986). Our four-item, seven-point se-

mantic differential scale is borrowed from Yi (1990). We chose

the term “mood” to specifically avoid evaluative connotations

associated with “affect.”

To the extent that humor is related to the ad, it may be

important to understanding the brand claims. Thus, related

Page 4: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 4/15

 Spring 2007  57

humor, which induces positive mood states, may facilitate brand

claims recall via systematic, or central route processing (Cline,

Altsech, and Kellaris 2003). In summary, we anticipate that

H3: The joint impact of humor strength and humor–message

relatedness on message claims recall will be mediated by

mood.

INDIVIDUAL DIFFERENCES IN

NEED FOR HUMOR 

Individual differences are likely to affect consumers’ receptiv-

ity to humorous stimuli (Zhang 1996). Failure to account for

such differences may help explain differences in past studies

(Duncan and Nelson 1985; Duncan, Nelson, and Frontczak

1984). Gelb and Zinkhan (1985) point out that because what

is funny to one person may not be to another, individual dif-

ferences in perceptions of humor should be considered. More

recently, Elpers, Mukherjee, and Hoyer (2004) performed a

moment-to-moment analysis of humor in television advertis-ing, demonstrating that moment-to-moment surprise and

humor drive individual perceptions of humor.

We propose that need for humor (NFH) should capture

important individual differences that shape responses to hu-

mor. NFH is a personality trait that refers to one’s tendency

to generate and seek out humor (Cline, Machleit, and Kellaris

1998). We expect NFH to moderate the effects of humor

strength on ad memorability. Based on differential motiva-

tion to process humorous stimuli, individuals high in NFH

are expected to “seek out” and attend to humorous stimuli

more readily than those who are low in NFH. Thus, high-

NFH individuals should be prone to recognize and respond

to different levels of humor in an ad. In contrast, people lowin NFH are less likely to acknowledge or respond to humor

of any strength (Dixon et al. 1989). Analogous to “need for

cognition,” NFH should distinguish individuals inclined to

process ad information based on the humor in the ad from

those likely to process ads primarily on the basis of nonhumor-

ous elements. Therefore, NFH should moderate the effects of

humor strength such that:

H4: The impact of humor strength on message claims recall will

be significantly more positive among individuals with high levels

of NFH than among those with low levels of NFH.

STUDY

Overview

The goals of this study are to test the hypotheses that (1) humor’s

strength interacts with humor–message relatedness to influence

claims recall, (2) attention and positive mood generated by hu-

mor mediate the impact of humor’s relatedness on recall of mes-

sage claims, and (3) individual differences in NFH moderate the

influence of humor strength on claims recall. Thus, an experi-

ment crosses humor strength (lower versus higher), humor–mes-

sage relatedness (lower versus higher), and NFH (higher versus

lower) in a 2 × 2 × 2 between-subjects factorial design with a

no-humor control condition. We manipulate humor strengthand humor–message relatedness and measure NFH. The primary

dependent variable is message claims recall.

Participants

Two hundred fifty-three students enrolled in upper-level

undergraduate marketing courses at a large, Midwestern uni-

versity participated in the study for course credit. Ages ranged

from 18 to 48 (median = 21); 51% were female. Participants

completed the experiment independently during a single ses-

sion and received academic credit for participating.

Stimulus Materials and Pretests

Humor Strength

To manipulate humor strength, we developed three versions

of a simulated print ad (higher humor strength, lower humor

strength, and no humor control). A format was patterned

after prior research using print ads (e.g., Arias-Bolzmann,

Chakraborty, and Mowen 2000;  Cline, Altsech, and Kel-

laris  2003; Edell and Staelin 1983; Heckler and Childers

1992; Houston, Childers, and Heckler 1987; Krishnan and

Chakravarti 2003). Each ad contained a brand name, a head-

line, a product picture, two cartoon figures, brand claims,

and a tag line. Following Krishnan and Chakravarti (2003),the headline was used as the primary manipulation of humor

strength. Following Lee and Mason (1999), cartoon illustra-

tions reinforced the humor manipulation. The size, position,

and fundamental meaning of the cartoon figures were held

constant across conditions.

An extensive program of pretesting was conducted to select

the product category, brand name, and brand claims, and to

test the humor-strength manipulation. The first two pretests

(n = 54) asked participants to rate products on two separate di-

mensions: personal relevancy and humor expectancy. Based on

midscale ratings for both humor-expectancy rating ( M = 4.28

on a seven-point scale) and personal relevancy ( M = 3.94 on a

seven-point scale), coffee was selected as the product category.

Coffee is representative of a low-risk, convenience product

category in which the use of humor is both frequent and

believed to be effective. Weinberger and Campbell (1991)

find the incidence of ad humor to be highest in the low in-

volvement/feeling cell of the FCB (Foote, Cone, and Belding)

Matrix, a cell made up of “personal pleasures,” including coffee

Page 5: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 5/15

58  The Journal of Advertising 

(Vaughn 1980, 1986). In addition, Weinberger et al. (1995)

find that the category comprising coffee in their Product Color

Matrix had the highest incidence of humor for magazine ads

and uniformly positive humor effects on three Starch scores:

“Noted” (initial attention), “Seen-Associated” (aided-brand

recall), and “Read Most” (held attention).

To control for prior knowledge and familiarity, little-knownbrand names for coffee were selected from a sample of 50

actual coffee-producing firms. A third pretest (n = 42) was

conducted to identify unfamiliar brand names that were neutral

with respect to their “signaling” of humor. Participants rated

12 brand names for humor expectancy on seven-point scales.

Because students rated the Wachusett brand name relatively

low on humor expectancy ( M = 2.91 on a seven-point scale),

and because no participant recognized the name Wachusett, it

was selected as the coffee brand name. Hundreds of real brand

claims were also sampled from coffee advertisements on the

Internet. A subset of these claims was chosen on the basis of a

fourth pretest (n = 31), which asked students to rate the ap-

propriateness of each claim to the product category.For each condition of humor strength, a “one-liner” was

created to constitute the headline. One-liners are a form of

comic wit used frequently in humorous ads (Speck 1987, 1991).

One-liners were chosen over more involved jokes to keep the

headlines simple. Also, one-liners (e.g., puns) involve an incon-

gruity-resolution process (versus arousal-safety or disparagement

processes). Using this humor process is consistent with Wyer

and Collins’s (1992) theory of humor elicitation. Finally, Spotts,

Weinberger, and Parsons (1997) find that more than 80% of

humorous magazine ads use incongruity (the humor process

comprising comic wit). Manipulating humor strength in the

form of one-liners (similar to Cline, Altsech, and Kellaris 2003;

Krishnan and Chakravarti 2003) affords the best opportunity to

control for potential confounds such as humor process, length,

complexity, and evoked emotions (Lammers et al. 1983).

A final pretest (n = 75) evaluated perceived humor for each

of the three humor-strength conditions using completed ver-

sions of the mock-up ads in a between-subjects test. The results

indicate that the treatments were perceived as differentially

humorous, F(1, 74) = 8.53, p < .001. In addition, the lower

humor strength ad ( M = 3.33 on a seven-point scale) was

perceived as more humorous than the no-humor control ad

( M = 2.62 on a seven-point scale), t = 1.85, df  = 51, p < .036

(one-tailed), and the higher humor strength ad ( M = 4.36 on

a seven-point scale) was perceived as more humorous than thelower humor strength ad ( M = 3.33 on a seven-point scale),

t = 2.07, df  = 43, p < .023 (one-tailed).

Humor–Message Relatedness

To manipulate humor–message relatedness, the tag line and

claims either refer specifically to or do not refer to the humor

in the headline. Speck (1987, 1991) identifies three underlying

humor dimensions: (1) incongruity-resolution—level of respondent

surprise, (2) dispositional humor —identification with or detach-

ment from the humor’s victim, and (3) arousal-safety—the degree

of effected relief. Speck describes five  combinational humor

types: (1) comic wit —incongruity-resolution humor; (2)  senti-

mental humor —arousal-safety; (3) satire—incongruity-resolutionand dispositional humor; (4) sentimental comedy—incongruity-

resolution arousal-safety humor; and (5) full comedy—incongru-

ity-resolution, dispositional, and arousal-safety humor. Speck

uses semiotic theory to define three levels of humor–message

relatedness: (1) pragmatic —is the ad fundamentally humorous

(humor-dominant) or generally nonhumorous? (2) semantic —the

thematic relationship of the humor and the message, that is,

is the humor related to the product, its use, its name, or its

benefits? (3) syntactic —the structural relationship between the

humor and the message, that is, is the message separate from or

part of the joke work? Finally, Speck draws on the ELM (Elabora-

tion Likelihood Model) to describe three fundamental processing

strategies: (1) central route—issue-relevant thinking , (2) peripheralroute—nonissue-relevant thinking , and (3) humor-dominant —a

special case of peripheral processing where the message is gen-

erally structured as a joke. Taken together, Speck’s taxonomy

allows for 20 humor-dominant ad forms—5 (humor type) × 2

(semantic relatedness) × 2 (syntactic relatedness).

In the present research, we deliberately selected semanti-

cally related comic wit for all humor conditions and manipu-

lated the structural (thematic) relatedness of the humor for the

higher (versus lower) humor–message relatedness conditions.

Specifically, the headline and message claims are semantically

related, that is, the claims “delicate, earthy flavor” and “natural,

distinct flavor” relate to the joke in the headline. However, only

in the conditions with higher humor–message relatedness is

the message syntactically (structurally) related to the humor.

Here, the play on words in the headline is carried through to

the tag line, and hence the loop in the joke work is closed.

The efficacy of the humor–message relatedness manipula-

tion was evaluated in a pretest (n = 26). Participants rated the

relatedness of the headline to the claims and tag line on a four-

item measure developed by Krishnan and Chakravarti (2003).

Results indicate that the humor–message relatedness manipu-

lation was successful for both low and high levels of humor

strength. The advertising stimuli are provided in Figure 1.

Measures

Message claims recall was prompted by asking participants,

“What major claims did the ad make? Please list as many as

you can remember. What other details about the ad do you

recall?” Participants’ recall of the message claims was com-

puted on the basis of a set of a priori rules. Two independent

raters, blind to the purposes of the study, coded the claims

Page 6: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 6/15

 Spring 2007  59

FIGURE 1Experimental Stimuli

recall measure; disparities were resolved via discussion. Because

the interrater agreement (90.9%) and the correlation between

the two rater’s scores (r = .935; n = 253, p < .001) were both

high, the two scores were averaged to form a claims recall

index, which served as the dependent measure. The proposed

mediator, attention to the ad, was measured by a three-item,

seven-point Likert-type scale intended to determine the

amount of attention devoted to the written message in the

ad (Muehling, Stoltman, and Grossbart 1990). The proposed

mediator, mood, was measured by a four-item, seven-point

semantic differential scale developed to measure the feeling

that has been induced in a person by an object (Yi 1990). In

addition, cognitive responses (e.g., humor comprehension and

elaboration) were assessed via a thought protocol task (e.g.,

Madden, Allen, and Twible 1988).

NFH was measured using a scale developed by Cline,

Altsech, and Kellaris (2003). Construct validity for the NFH

scale was assessed following the procedures recommended by

Control Humor Strength Lower/Humor–Message

Relatedness Lower

Humor Strength Lower/Humor–Message

Relatedness Higher

Humor Strength Higher/Humor–Message

Relatedness Lower

Humor Strength Higher/Humor–Message

Relatedness Higher

Page 7: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 7/15

60  The Journal of Advertising 

Gerbing and Anderson (1988). A median split was used to

separate the sample into low and high groups for each dimension

(e.g., Srull, Lichtenstein, and Rothbart 1985).

A manipulation check for humor strength asked participants

to record their reactions to the ad on a six-item, seven-point

semantic differential scale (humorous/not humorous, funny/not

funny, amusing/not amusing, playful/not playful, not dull/dull,and not boring/boring) adapted from Alden, Mukherjee, and

Hoyer (2000) and Chattopadhyay and Basu (1990). Further-

more, because asking individuals whether they perceive a mes-

sage to be funny or amusing might vary significantly between

individuals, a second measure was employed. Using a one-item,

seven-point semantic differential scale, participants were asked

to what extent the advertiser was attempting  to be humorous (i.e.,

a more objective measure of humor). This scale is suggested by

Duncan and Nelson (1985).

A manipulation check for humor–message relatedness asked

participants to record their reactions to the ad on a four-item

scale (related well/related poorly, were consistent/were inconsis-

tent, fit well/fit poorly, corresponded well/corresponded poorly)taken from Krishnan and Chakravarti (2003). Finally, questions

assessing argument strength, product familiarity, and brand

familiarity were included as confounding checks.

Procedure

Participants were told that the study was concerned with their

reactions to a magazine advertisement prototype. Each partici-

pant received a questionnaire booklet containing a stimulus

ad. After about 10 minutes, participants were asked to place

the questionnaire on the floor and continue with a filler task

(e.g., Heckler and Childers 1992) to clear their short-term

memories. Subsequently, message claims recall was assessedvia a second questionnaire, which also contained manipulation

and confounding checks, as well as a final question pertaining

to the purpose of the study. At the conclusion of the study,

participants were debriefed.

RESULTS

Manipulation and Confounding Checks

Humor Strength

Both subjective and objective measures of humor were used to

verify the effectiveness of the humor manipulation. Two-factoranalyses of variance (ANOVAs) were performed on (1) a six-

item measure of perceived humor (Cronbach’s α = .971) and

(2) a one-item measure of attempted humor with the humor

strength and humor–message relatedness as the factors. Only

the anticipated main effect for humor was significant for both

perceived humor, F(1, 211) = 19.38, p < .001, ω2 = .080, and

attempted humor, F(1, 211) = 35.25, p < .001, ω2 = .139. No

other main or interactive effects were observed. In addition,

one-way ANOVAs with post hoc multiple comparisons indicate

that the three levels of humor (control, lower humor, and higher

humor) were perceived as differentially humorous and received

significantly different ratings for attempted humor ( p < .001).

Humor–Message Relatedness

A two-factor ANOVA was performed on the perceived hu-

mor–message relatedness dependent variable with manipulated

levels of humor and humor–message relatedness as the fac-

tors. Results indicated a significant main effect for perceived

humor–message relatedness, F(1, 205) = 63.62,  p < .001,

ω2 = .231, and a nonsignificant main effect for humor, F(1,

205) = .236, p = .628.

Measures assessing claim strength, product familiarity, and

brand familiarity were included as confounding checks. The

data were analyzed in a series of ANOVAs with manipulations

of humor strength and humor–message relatedness as the fac-tors. Claim strength was not confounded with humor strength,

F(1, 205) = .096, p = .901, humor–message relatedness, F(1,

205) = .093, p = .760, or the interaction of humor strength

and humor–message relatedness, F(1, 205) = 2.35, p = .127.

Similarly, product familiarity was not confounded with the

manipulations or their interaction (all p values >.513). Brand

familiarity was not confounded with the manipulations or their

interaction (all p values >.554). In summary, results suggest

that the stimulus ads manipulated the humor and relatedness

constructs successfully and independently, and the treatments

were not confounded with other traits likely to provide alter-

native explanations.

The Impact of Humor Strength and

Humor–Message Relatedness on Recall

H1 proposed that claims recall would be more positively

influenced by humor strength when humor–message relat-

edness is high than when it is low. ANOVA shows a main

effect of humor–message relatedness, F(1, 211) = 6.25,

 p = .013, ω2 = .024, with higher humor–message related-

ness ( M = 1.45) engendering greater claims recall than lower

humor–message relatedness ( M = 1.08), t = 2.51, df  = 190,

 p = .031. An interaction between humor strength and hu-

mor–message relatedness was also observed, F(1, 211) = 4.34, p = .038, ω2 = .02. When humor strength is higher (versus

lower) people recall more ad claims under conditions of

higher ( M = 1.66) versus lower humor–message relatedness

( M = .01), t = 3.26, df  = 101, p = .001 (one-tailed). Thus, H1

is supported. In addition, when humor–message relatedness

is higher (versus lower) people recall more ad claims under

Page 8: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 8/15

 Spring 2007  61

conditions of higher ( M = 1.66) versus lower humor strength

( M = 1.20), t = 1.95, df  = 103,  p = .028 (one-tailed). The

higher humor strength/high humor–message relatedness

condition ( M = 1.66) produces greater recall than the control

group ( M = 1.24), t = 1.85, df  = 95,  p = .034 (one-tailed).

Claims recall for the control group does not differ significantly

from any other treatment condition. Figure 2 illustrates the

interaction. Table 1 provides descriptive statistics, and Table

2 provides the ANOVA.

Attention and Mood as Mediators

 Attention

To test the hypothesis that the joint impact of humor strength

and humor–message relatedness on claims recall is mediated

by attention to the ad (H2), we used a procedure similar to

that described by Sobel (1982). First, the multiplicative in-

dependent variable, humor strength × humor–message relat-

edness, is positively correlated with the dependent variable,

claims recall, r = .131,  p = .038. Second, the independent

variable is positively correlated with the proposed mediator,

attention to the ad (r = .200, p = .001). Third, the mediator

is positively correlated with the brand claims recall dependent

variable, r = .147,  p = .019. Finally, when the dependentvariable, claims recall, is regressed on both the mediator

(attention to the ad) and the independent variable (humor

strength × humor–message relatedness), the relationship be-

tween the independent variable and the dependent variable

attenuates (β = .108, t = 1.69, p = .092), thereby providing

support for a partial mediation model and for H2.

 Mood 

H3 predicted that the joint impact of humor strength and

humor–message relatedness on ad claims recall would be medi-

ated by mood. Results from H2 indicate that the independent

variable, humor strength × humor–message relatedness, is

positively correlated with the dependent variable, claims recall.

Second, the independent variable is positively correlated with

the proposed mediator, mood, r = .548, p < .001. Third, mood

is positively correlated with claims recall, r = .216, p = .001.

Finally, when the dependent variable is regressed on both the

independent and mediator variables, the relationship between

the independent variable and the dependent variable becomes

nonsignificant (β = .018, t = .244, p = .808), suggesting full

mediation and support for H3.

Concurrent Tests of Mediation

To explain further the process by which mood mediates the

joint effects of humor strength and humor–message relatedness

on claims recall, additional analyses were conducted. Inde-

pendent support for H2 and H3 suggests that both attention

to the ad and mood may mediate the joint impact of humor

strength and humor–message relatedness on claims recall.

What is not clear from the separate analyses, however, is the

nature of the relationship between attention and mood. Thus,

the dependent variable, claims recall, was regressed on both

mediators (attention and mood) and the independent factor(humor strength × humor–message relatedness). Results of

this model indicate that mood accounts for most of the vari-

ance in claims recall. In addition, the results imply that mood

jointly mediates the relationship between attention and recall,

as well as the relationship between the independent variable

and recall. Specifically, the coefficient for mood is significant,

FIGURE 2Humor Strength × Humor–Message Relevancy on

Recall

TABLE 1

Descriptive Statistics

  Claims recall

  Mean n

Low humor 

  Low humor–message relatedness 1.156 52  High humor–message relatedness 1.204 50

  Total 1.179 102

High humor 

  Low humor–message relatedness 1.009 54

  High humor–message relatedness 1.661 56

  Total 1.341 110

Total 

  Low humor–message relatedness 1.082 107

  High humor–message relatedness 1.448 105

  Total 1.263 212.9

.7

Page 9: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 9/15

62  The Journal of Advertising 

proposed by Ping (1996). The anticipated link between humor

strength × humor–message relatedness and mood is observed

(γ21 = .054, t = 9.95, df = 3,  p < .01), and a significant linkbetween humor strength × humor–message relatedness and

attention is observed (γ11

= .026, t = 3.11, df = 3,  p < .05).

In addition, attention appears to influence mood (β21

= .151,

t = 3.33, df = 3, p < .05) and elaboration (β31

= .377, t = 4.54,

df = 3, p < .01). In contrast, mood is not significantly related

to elaboration (β32

= –.195, t = 1.83, df = 3, n.s.). Finally,

mood positively influences claims recall (β42

= .256, t = 3.62,

df  = 3,  p < .05) and elaboration positively influences claims

recall (β43

= .117, t = 2.47, df = 3, p < .05).

In addition to validating the proposed nomological net-

work, the model fit the data exceptionally well. The χ2 value

was not significant (χ2 = .51, df = 3,  p = .92). Furthermore,

the goodness-of-fit index (GFI = .99), the adjusted goodness-

of-fit index (AGFI = .98), and the root mean square residual

(RMSR = .062) provide evidence that the model fit the data

well. No standardized residuals fell outside the range of –1.0 to

1.0. Thus, the path model appears to explain the relationships

between the interaction of humor strength and humor–message

relatedness, attention to the ad, mood, elaboration generated

from the ad, and message claims recall. Humor strength and

humor–message relatedness influence claims recall via a dual

 process—attention and mood. Moreover, mood mediates—both

directly and indirectly—the joint effects of humor strength

and humor–message relatedness on claims recall. In contrast,

attention appears to work through mood and elaboration toinfluence claims recall.

NFH as a Moderator

H4 predicted that NFH would moderate the impact of humor

strength such that its effects on claims recall would be more

positive for people with higher levels of NFH than for people

βmood

= .186, t = 2.48,  p = .014, whereas the coefficients for

humor strength × humor–message relatedness and attention

are both nonsignificant, βhum×rel = .012, t = .161,  p = .872,β

attention = .084, t = 1.20, p = .194.

Next, we examined the possibility that mood may operate

through elaboration. The correlation between mood and total

thoughts generated from the ad is nonsignificant, however,

r  = –.027, p = .672. Therefore, mood is unlikely to produce

effects on message claims recall via elaboration. Moreover, in a

separate model, when claims recall was regressed on both mood

and elaboration, both predictors produced significant coeffi-

cients and offered no indications of collinearity (βmood

 = .220,

t = 3.62, p < .001; βelaboration

= .166, t = 2.72, p = .007). This

suggests that mood and elaboration may work independently

to affect claims recall.

Finally, elaboration was tested as a mediator to the relation-

ship between attention to the ad and claims recall. Pairwise

correlations among elaboration, attention, and claims recall are

significant , r attention,

 

recall

 = .137 ( p = .030), r attention,

 

elaboration

 = .255

( p < .001), r elaboration,

 

recall = .160 ( p < .011). When claims recall

is regressed on both attention and elaboration, however, the

association between attention and claims recall becomes nonsig-

nificant (β = .103, t = 1.60, p = .111). This suggests that elabo-

ration mediates the relationship between attention and claims

recall. In summary of these findings, it appears that whereas

mood and elaboration produce independent effects on claims

recall, attention works through both mood and elaboration to

influence claims recall. In both cases, the interaction of humorstrength and humor–message relatedness is likely to stimulate

the process. To test these relationships in a nomological net-

work, a path model with reliabilities (Figure 3) was developed.

The humor strength × humor–message relatedness served as

the exogenous variable and attention, mood, elaboration, and

claims recall were specified as endogenous variables.

To test the interaction hypothesis, we used the methodology

TABLE 2

Claims Recall: Humor Strength × Humor–Message Relatedness

  Experimental method

  Sum of Mean

  squares df   square F   Significance

Main effects  Combined 8.31 2.00 4.15 3.75 .03

  Humor strength 1.22 1.00 1.22 1.10 .30

  Humor–message relatedness 6.92 1.00 6.92 6.25 .01

Two-way interactions

  Humor strength × humor–message relatedness 4.81 1.00 4.81 4.34 .04

Model 13.12 3.00 4.37 3.95 .01

Residual 230.16 208.00 1.11

Total 243.28 211 1.15

 Note: ANOVA = analysis of variance.

Page 10: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 10/15

 Spring 2007  63

with lower levels of NFH.. An ANOVA was performed on

claims recall, with humor strength and NFH as the factors.

The anticipated interaction between humor strength and NFH

was observed, F(1, 195) = 4.37, p = .038,ω2 = .02. The results

indicate that when humor strength is higher (versus lower),

those higher in NFH recalled more ad claims ( M = 1.59) than

those lower in NFH ( M = 1.07), t = 2.48, df  = 97,  p = .008

(one-tailed). Thus, H4 is supported. Figure 4 illustrates this

interaction. Table 3 provides descriptive statistics, and Table 4

provides the ANOVA.

Discussion

As anticipated, the data demonstrate that humor’s attention-

gaining mechanisms (i.e., humor strength) may translate into

positive effects for memory. Specifically, it appears that humor

strength and humor–message relatedness jointly influence

participants’ recall of advertising claims. The results show

that when humor strength is higher (versus lower), partici-pants recall more ad claims when the humor is relevant to the

claims. The data also show that when the humor is related

to the message, participants recall more advertising claims

when humor strength is higher than when it is lower. When

the humor is both stronger and related to the message, the

humorous ad outperforms the control ad with respect to mes-

sage claims recall. These results are important because they

bring into clearer relief the value of using strong, relevant

humor, as opposed to stronger, incidental humor. It appears

that strong humor is not its own virtue; it must be connected

to the brand claims to facilitate recall. In contrast, weaker

humor, whether related to the message or not, does not aid

brand claims recall.

The data also provide additional insight into person-by-

situation influences on advertising responses. For example,

an individual’s tendency to generate humor (NFH) appears

to moderate humor strength effects such that the impact of

humor strength on claims recall is significantly more posi-

tive for people with high levels of NFH than for people with

low levels of NFH. When humor strength is higher (versus

lower), individuals higher in NFH recall more ad claims than

those lower in NFH. In addition, those higher (versus lower)

in NFH recall more ad claims when humor strength is higher

than when it is lower.

The data also support the notion that attention and mood

operate via dual processes to mediate the joint impact of hu-mor strength and humor–message relatedness on claims recall.

Having first established the importance of strong, structurally

related humor, this dual process model helps explain why the

strong, relevant humor leads to higher brand claims recall.

These results are interesting because they suggest that humor

may trigger both cognitive (attention) and affective (mood)

routes to memorability.

FIGURE 3Path Model with Reliabilities

* p < .05.

** p < .01.

Page 11: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 11/15

64  The Journal of Advertising 

GENERAL DISCUSSION

The goal of this research was to test the generative mechanisms

by which humor–message relatedness influences brand claims

recall. In addition, we introduced the personality variable

“need for humor” (NFH) as a moderator of humor’s effects on

advertising recall and investigated the mediating roles of at-

tention, mood, and elaboration in a nomological network.

This research makes a number of theoretical contributions.

First, by examining the joint effects of humor strength and

humor–message relatedness on advertising recall, this research

provides evidence that message-relevant stimuli may produce

consistent effects across various attention-gaining contexts. For

example, the findings suggest that humor–message relatedness

interacts with humor’s attention-gaining properties in a man-

ner that is analogous to the interaction between music–message

congruity and music’s attention-gaining properties (Kellaris,

Cox, and Cox 1993). Thus, the present research provides a

foundation for suggesting that attention-gaining message

appeals, in general, may engender better recall when they are

structurally related to the message theme. Second, by exam-

ining the psychological mechanisms that underlie consumer

responses to humorous stimuli (e.g., attention, mood, and

elaboration), as well as the theoretical principles that activate

these mechanisms (i.e., humor–message relatedness), this re-search leads to better explanations for why humorous effects

obtain or fail to obtain. Attention and mood, for example, are

shown as dual mediators of the impact of humor strength and

humor–message relatedness on recall for the message claims.

Third, NFH, like other individual difference variables, explains

an additional source of variation in advertising outcomes

through its role as a moderator. The NFH construct broadens

FIGURE 4Humor Strength × NFH (Need for Humor) on Recall

TABLE 3

Descriptive Statistics

  Recall

  Mean n

Low humor 

  Lower NFH 1.21 49  Higher NFH 1.10 48

  Total 1.16 97

High humor 

  Lower NFH 1.07 48

  Higher NFH 1.59 51

  Total 1.34 99

Total 

  Lower NFH 1.14 104

  Higher NFH 1.38 92

  Total 1.25 196

 Note: NFH = need for humor.

our knowledge of the consumer’s sense of humor in general,

and under what conditions it influences responses to humor-

ous communication.

The incremental contributions of the present research are

twofold. First, whereas prior research has shown a positive re-

lationship between humor and attention, and between related

humor and recall, previous studies have not explicitly investi-

gated the joint interplay of humor strength and humor–message

relatedness. The present study specifically addresses this im-

portant interaction and finds that with respect to claims recall,

neither humor strength nor humor–message relatedness is its

own virtue. Relatively stronger humor must be related to the

brand or message to engender higher claims recall. Second,

previous research has not investigated the generative mecha-

nisms that follow from the interplay of humor strength and

humor–message relatedness. The present research specifically

evaluates the influence of attention and mood in a nomological

network of variables. We find that attention and mood oper-

ate in partnership, jointly mediating the interactive effects of

humor strength and humor–message relatedness on claims

recall. The path model uncovers this dual process. Mood directly

mediates humor’s influence on claims recall. In contrast, atten-

tion operates indirectly through both mood and elaboration to

influence claims recall. Thus, both variables help explain why 

the interaction between humor strength and humor–messagerelatedness influences claims recall.

The present findings also have practical implications in

terms of guidelines for advertisers who use humorous ads

and imply that the effects of message-congruent humor may

generalize to other attention-gaining appeals, such as music or

sexual attraction. On the basis of humor’s prevalence and the

belief in its universal effectiveness, NFH may be useful as a

.7

.9

.5

Page 12: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 12/15

 Spring 2007  65

segmentation tool. Clearly, advertisers cannot administer NFH

scales to members of their target audiences; however, marketresearch can identify media that draw groups of people with

high levels of NFH (e.g., Mad Magazine readers and Late Show

with David Letterman watchers). In addition, this information

can be used to determine which product categories or brands

tend to be popular with specific media users (e.g., Mad Maga-

zine readers may tend to be heavy users of B-movies or video

games). Thus, NFH may be helpful both in media selection

and in targeting audiences for specific products.

LIMITATIONS AND FUTURE RESEARCH

The results of this research are necessarily qualified by the

limitations of the studies’ designs, each of which suggests an

opportunity for additional research. First, the experimenta-

tion used mock advertisements in a controlled setting. Thus,

the extent to which the findings are limited by the artificial

context and the representativeness of the ads and product is

unknown. Second, both studies use comic wit, manipulated in

the headline of a print ad, to convey humor that is independent

of the brand name, thus limiting generality to other types of

humor. This choice, however, allows for control over the for-

mat of the ads across conditions and thus avoids confounding

other ad characteristics with humor (Krishnan and Chakravarti

2003). Furthermore, it enables clear identification of the locus

of effects to specific ad components (e.g., brand claims). Third,whereas all ads were designed such that the humor was related

to the brand claims (i.e., thematically related), the structural

relatedness of the humor was manipulated via the tag line.

Thus, future research should examine the structural location

of the humor in an ad. Fourth, despite extensive pretesting, it

could be argued that humor is more appropriate in a broadcast

medium (e.g., Alden and Hoyer 1993; Chattopadhyay and

Basu 1990) or that other humor dimensions (e.g., disposi-

tional humor; Speck 1987, 1991) or humor structures (e.g.,message-dominant/image-oriented; Spotts, Weinberger, and

Parsons 1997) may be more effective in a print context. The

use of incongruous humor, however, is justified on the grounds

that it appears to be the most pervasive humor in magazine ads

and for nondurable items such as coffee (Spotts, Weinberger,

and Parsons 1997).

The results of the experimentation are intuitively appealing.

Each year, advertisers spend millions of dollars on humorous

ads. Some contain strong, relevant messages that engender

recall. Recently, several brands have generated remarkable

brand awareness with strong, related humor. Geico’s gecko and

Aflac’s duck appear to be deliberate attempts at linking the

source of the humor with the brand name, and hence increas-

ing recall. Orbit gum links its “dirty mouth” humor with its

primary attribute, clean-tasting gum. Classic successes like the

Eveready’s Energizer Bunny were designed to link the humor

with the claims. The message is indirect but clear—like the

bunny, the batteries keep going and going. In contrast, it ap-

pears that millions of dollars may be wasted on advertisements

that make no connection between the humor and the claim,

subsequently creating brand confusion or inhibiting recall.

CONCLUSION

The pervasive use of humor in advertising attests to the wide-spread belief that humor enhances the effectiveness of ads. The

research reported here examines contingencies that shape the

effects of humorous appeals on an important outcome of ad-

vertising—consumers’ recall of ad claims—as well as processes

through which and boundary conditions within which such

effects operate.  In so doing, this research seeks to elucidate

when and how humor contributes to remembering ads.

TABLE 4

Recall: By Humor Strength by NFH

  Experimental method

  Sum of Mean

  squares df   square F   Significance

 Main effects  Combined 3.74 2 1.87 1.70 .19

  Humor strength 1.14 1 1.14 1.04 .31

  NFH 2.27 1 2.27 2.06 .15

Two-way interactions

  Humor strength × NFH 4.82 1 4.82 4.37 .04

 Model   8.57 3 2.86 2.59 .05

Residual   211.93 192 1.10

Total   220.50 195 1.13

 Note: ANOVA = analysis of variance; NFH = need for humor.

Page 13: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 13/15

66  The Journal of Advertising 

REFERENCES

Alba, Joseph W., J. Wesley Hutchison, and John G. Lynch (1991),“Memory and Decision Making,” in Handbook of Consumer Research, Thomas S. Robertson and Harold H. Kassarjian,eds., Englewood Cliffs, NJ: Prentice-Hall, 1–49.

Alden, Dana L., and Wayne D. Hoyer (1993), “An Examination

of Cognitive Factors Related to Humorousness in TelevisionAdvertising,” Journal of Advertising, 22 (2), 29–37.

———, Ashesh Mukherjee, and Wayne D. Hoyer (2000), “TheEffects of Incongruity, Surprise, and Positive Moderatorson Perceived Humor in Television Advertising,” Journal of Advertising, 29 (2), 1–15.

Arias-Bolzmann, Leopoldo, Goutam Chakraborty, and John C.Mowen (2000), “Effects of Absurdity in Advertising: TheModerating Role of Product Category Attitude and theMediating Role of Cognitive Responses,” Journal of Adver-tising, 29 (1), 35–49.

Batra, Rajeev, and Michael L. Ray (1986), “Affective ResponsesMediating Acceptance of Advertising,” Journal of Consumer Research, 13 (September), 234–249.

Bower, Geoffrey H., Stephen G. Gilligan, and Kenneth P. Mon-teiro (1981), “Selectivity of Learning Caused by AffectiveStates,” in  Affect and Social Cognition,   Susan Fiske andMargaret Clark, eds., Hillsdale, NJ: Lawrence Erlbaum,291–331.

Bruner, Gordon C., II, and Paul J. Hensel (1996), Marketing ScalesHandbook, vol. 2, Chicago: American Marketing Association.

Chattopadhyay, Amitava, and Kunal Basu (1990), “Humor in Ad-vertising: The Moderating Role of Prior Brand Evaluation,” Journal of Marketing Research, 27 (November), 466–476.

Cline, Thomas W., Moses B. Altsech, and James J. Kellaris(2003), “When Does Humor Enhance or Inhibit Ad Re-sponses? The Moderating Role of Need for Levity,” Journalof Advertising, 32 (3), 31–46.

———, Karen A. Machleit, and James J. Kellaris (1998), “IsThere a Need for Levity?” in  Proceedings of the Society forConsumer Psychology 1998 Winter Conference,  Margaret C.Campbell and Karen A. Machleit, eds., Washington, DC:American Psychological Association, 155–157.

Dixon, Paul N., William Willingham, Donald A. Strano, andCynthia K. Chandler (1989), “Sense of Humor as a MediatorDuring Incidental Learning of Humor-Related Material,” Psychological Reports, 64, 851–855.

Duncan, Calvin P. (1979), “Humor in Advertising: A BehavioralPerspective,” Journal of the Academy of Marketing Science, 7(4), 285–306.

———, and James E. Nelson (1985), “Effects of Humor in aRadio Advertising Experiment,” Journal of Advertising, 14

(2), 33–40.———, ———, and Nancy T. Frontczak (1984), “The Effect

of Humor on Advertising Comprehension,” in Advances inConsumer Research, Thomas C. Kinnear, ed., Provo, UT: As-sociation for Consumer Research, 432–437.

Edell, Julia A., and Richard Staelin (1983), “The InformationProcessing of Pictures in Print Advertisements,” Journal ofConsumer Research, 10 (1), 45–61.

Elpers, Josephine, Ashesh Mukherjee, and Wayne Hoyer (2004),“Humor in Television Advertising: A Moment-to-MomentAnalysis,” Journal of Consumer Research, 31 (3), 592–598.

Galizio, Mark, and Clyde Hendrick (1972), “Effect of Musical Ac-companiment on Attitude: The Guitar as a Prop for Persua-sion,” Journal of Applied Social Psychology, 2 (4), 350–359.

Gelb, Betsy, and George Zinkhan (1985), “The Effect of Repeti-

tion on Humor in a Radio Advertising Study,”  Journal of Advertising, 14 (4), 13–20, 68.

Gerbing, David W., and James C. Anderson (1988), “An UpdatedParadigm for Scale Development Incorporating Unidimen-sionality and Its Assessment,” Journal of Marketing Research, 25 (May), 186–192.

Heckler, Susan E., and Terry L. Childers (1992), “The Role ofExpectancy and Relevancy in Memory for Verbal and VisualInformation: What Is Incongruency?”  Journal of Consumer Research, 18 (4), 475–492.

Houston, Michael J., Terry L. Childers, and Susan E. Heckler(1987), “Picture-Word Consistency and the ElaborativeProcessing of Advertisements,” Journal of Marketing Research, 24 (December), 359–369.

Isen, Alice M. (1987), “Positive Affect, Cognitive Processes andSocial Behavior,” in Advances in Experimental Social Psychol-ogy, Lawrence Berkowitz, ed., New York: Academic Press,203–253.

———. (1989), “Some Ways in Which Affect Influences Cogni-tive Processes: Implications for Advertising and ConsumerBehavior,” in Advertising and Consumer Psychology, Alice M.Tybout and Patricia Cafferata, eds., New York: LexingtonBooks, 91–117.

Kellaris, James J., Anthony D. Cox, and Dena Cox (1993), “TheEffect of Background Music on Ad Processing: A Contin-gency Explanation,”  Journal of Marketing,  57 (October),114–125.

Krishnan, H. Shanker, and Dipankar Chakravarti (2003), “A

Process Analysis of the Effects of Humorous AdvertisingExecutions on Brand Claims Memory,”  Journal of Consumer Psychology, 13 (3), 230–245.

Lammers, H. Bruce, Laura Liebowitz, George E. Seymour,and Judith E. Hennessey (1983), “Humor and CognitiveResponses to Advertising Stimuli: A Trace ConsolidationApproach,” Journal of Business Research, 11 (2), 173–185.

Lee, Yih L., and Charlotte Mason (1999), “Responses to Informa-tion Incongruency in Advertising: The Role of Expectancy,Relevancy, and Humor,”  Journal of Consumer Research, 26(2), 156–169.

Littmann, Jane R. (1983), “A New Formulation of Humor,”   Advances in Descriptive Psychology, 3, 183–207.

Machleit, Karen A., and R. Dale Wilson (1988), “Emotional

Feelings and Attitude Toward the Advertisement: The Rolesof Brand Familiarity and Repetition,” Journal of Advertising, 17 (Fall), 27–35.

Madden, Thomas J. (1982), “Humor in Advertising: Applicationsof a Hierarchy-of-Effects Paradigm,” Ph.D. diss., Universityof Massachusetts–Amherst.

———, Chris T. Allen, and Jacquelyn L. Twible (1988), “At-titude Toward the Ad: An Assessment of Diverse Measure-

Page 14: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 14/15

 Spring 2007  67

ment Indices Under Different Processing ‘Sets,’” Journal of Marketing Research, 25 (3), 242–252.

———, and Marc G. Weinberger (1982), “The Effects of Humoron Attention in Magazine Advertising,” Journal of Advertis-ing, 11 (3), 8–14.

———, and ——— (1984), “Humor in Advertising: A Practi-tioner View,” Journal of Advertising Research, 24 (4), 23–29.

Meyers-Levy, Joan (1991), “Elaborating on Elaboration: The Dis-tinction Between Relational and Item-Specific Elaboration,” Journal of Consumer Research, 18 (3), 358–367.

Muehling, Darrel D., Jeffrey J. Stoltman, and Sanford Gross-bart (1990), “The Impact of Comparative Advertising onLevels of Message Involvement,” Journal of Advertising, 16(3), 41–50.

Olson, James M., and Neal J. Roese (1995), “The Perceived Fun-niness of Humorous Stimuli,” Personality and Social PsychologyBulletin, 21 (9), 908–913.

O’quin, Karen, and James Aronoff (1981), “Humor as a Tech-nique of Social Influence,”  Social Psychology Quarterly, 44(4), 349–357.

Ping, Robert A., Jr. (1996), “Latent Variable Interaction and

Quadratic Effect Estimation: A Two-Step Technique UsingStructural Equation Analysis,”  Psychological Bulletin,  119(1), 166–175.

Smith, Stephen M. (1993), “Does Humor in Advertising EnhanceSystematic Processing?” in Advances in Consumer Research, vol.20, Leigh McAlister and Michael L. Rothschild, eds., Provo,UT: Association for Consumer Research, 155–158.

Sobel, Mark E. (1982), “Asymptotic Intervals for Indirect Effectsin Structural Equation Models,” in Sociological Methodology,S. Leinhart, ed., San Francisco: Jossey-Bass, 290–312.

Speck, Paul S. (1987), “On Humor and Humor in Advertising,”Ph.D. diss., Texas Tech University.

——— (1991), “The Humorous Ads,” in Current Issues and Re- search in Advertising, James H. Leigh and Claude R. Martin,

 Jr., eds., Ann Arbor: University of Michigan Press, 1–44.

Spotts, Harlan E., Marc G. Weinberger, and Amy L. Parsons(1997), “Assessing the Use and Impact of Humor on Ad-vertising Effectiveness: A Contingency Approach,” Journalof Advertising, 26 (3), 17–32.

Srull, Thomas K., Meryl Lichtenstein, and Myron Rothbart (1985),“Associative Storage and Retrieval Processes in Person Mem-ory,” Journal of Experimental Psychology, 11 (2), 316–345.

Sternthal, Brian, and C. Samuel Craig (1973), “Humor in Ad-vertising,” Journal of Marketing, 37 (4), 12–18.

Vaughn, Richard (1980), “How Advertising Works: A PlanningModel,” Journal of Advertising Research, 20 (5), 27–33.

——— (1986), “How Advertising Works: A Planning ModelRevisited,” Journal of Advertising Research, 26 (1), 57–66.

Weinberger, Marc G., and Leland Campbell (1991), “The Useand Impact of Humor in Radio Advertising,”  Journal of Advertising Research, 31 (December), 44–52.

———, and Charles S. Gulas (1992), “The Impact of Humorin Advertising: A Review,”  Journal of Advertising, 21 (4),36–59.

———, Harlan E. Spotts, Leland Campbell, and Amy L. Par-sons (1995), “The Use of Humor in Different Advertising

Media,”  Journal of Advertising Research,  35 (May/June),44–56.

Wicker, Frank W., Irene M. Thorelli, William L. Barron III, andMarguerite R. Ponder (1981), “Relationships Among Af-fective and Cognitive Factors in Humor,” Journal of Researchin Personality, 15, 350–370.

Wyer, Robert S., and James E. Collins (1992), “A Theory of Hu-mor Elicitation,” Psychological Review, 99 (4), 663–688.

Yi, Youjae (1990), “Cognitive and Affective Priming Effects ofthe Context for Print Advertisements,” Journal of Advertis-ing, 19 (2), 40–48.

Zhang, Yong (1996), “Responses to Humorous Advertising:The Moderating Effect of Need for Cognition,” Journal of Advertising, 15 (1), 15–32.

Page 15: Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

8/10/2019 Cline & Kellaris 2007 - The Influence of Humor Strength and Humor-Message Relatedness on Ad Memorability

http://slidepdf.com/reader/full/cline-kellaris-2007-the-influence-of-humor-strength-and-humor-message-relatedness 15/15