117

Humour effect on memory and

attitude: moderating role of

product involvement

Hwiman Chung

New Mexico State University, Las Cruces

Xinshu Zhao

University of North Carolina at Chapel Hill

This study examined the moderating effects of product involvement on the effects of
humour on memory and attitude towards the advertisement by using multi-year survey
(1992 to 1997) of responses to commercials shown during the Super Bowl. Positive and
significant relationships between humorous advertisements on memory and attitude were
found through multiple regression analysis. Furthermore, results show that humorous
advertisements are more effective in low-involvement products in terms of memory and
attitude towards the advertisement.

Due in part to the popularity of using humorous advertising
campaigns (according to Weinberger and Spotts, 24.4% of prime-time
television advertising in the USA is intended to be humorous), the
advertising scholars have studied the effects of humorous advertising
campaigns on advertising effectiveness (e.g. Markiewicz 1974; Cantor
& Venus 1980; Belch & Belch 1983; Duncan et al. 1983; Gelb & Pickett
1983; Sutherland & Middleton 1983; Madden & Weinberger 1984).
Sternthal and Craig (1973) drew some tentative but useful conclusions
about the effects of humour on advertising by reviewing the early
literature on humour in general, and Gelb and Pickett (1983) and
Spotts et al. (1997) provide some theoretical discussions of how
humorous advertising may affect consumers. These discussions
consider the use of humorous messages, which can create some
positive (favourable) attitudes towards the advertised brand through a
transfer of effect created by the ad to the brand. This transfer of

International Journal of Advertising, 22, pp. 117–144
© 2003 Advertising Association
Published by the World Advertising Research Center, Farm Road, Henley-on-Thames,
Oxon RG9 1EJ, UK

Chung.qxd 28/01/03 13:45 Page 117

effect has been proven by researchers in consumer behaviour (Ray &
Batra 1983; Holbrook & O’Shaughnessy 1984; Mitchell 1986).

In terms of advertising effectiveness, numerous studies have

suggested that advertising liking could contribute to an advertise-
ment’s effectiveness in terms of recall, brand preference or persuasion
(Du Plessis 1994; Hollis 1995). As Du Plessis (1994) and Walker and
Dubitsky (1994) reported, commercial liking (or attitude towards the
ad) relates positively to advertising recall. One theoretical background
for this relationship is that likeable or well-liked advertisements can
affect an individual’s information processing by creating positive
arousal, increasing the memory of the advertised material, and
creating more favourable judgements of the advertisement message
(Edell & Burke 1986; Aaker & Myers 1987). Our purpose extends
work in this research stream by considering the issues of product
involvement. The primary focus of previous studies of the effects of
humorous advertisements has been on attitude towards the
advertisement and memory. In this study, we include product
involvement as a moderating variable to provide insight into the
differences of humorous ads on subjects’ attitudes towards the ad and
memory. The purpose of the study is to add to the body of knowledge
regarding the effects of humorous advertisements on cognitive and
affective aspects of

advertising effectiveness and product

involvement. It is usually agreed among advertising practitioners that
we should not use humorous advertising for high-involvement prod-
ucts because it may cause effects opposite to those we intended. It is
important for advertising practitioners to understand what exact
effects humorous messages have, compared with non-humorous
advertising, because often the advertising objective is to get high recall
for an advertised brand by increasing the amount of attention. If,
indeed, humorous appeal is more effective in terms of grabbing
attention, high recall and message comprehension, it will be much
easier for advertising practitioners to develop an advertising message.
As some researchers argue that advertising studies which use
laboratory settings are weak in their ‘generalisability’ (see, for example,
Zhao 1997), this study also tries to find the effects of humour on
memory and attitude in a natural television-watching environment.

118

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 118

BACKGROUND

How different message appeal in an advertisement relates to the
effectiveness of that advertisement is a long-standing and unsolved
question. Academic studies report inconsistent results on the
effectiveness of humour in ads, but the absence of systematic
empirical results contrasts with humour’s widespread use (Markiewicz
1974) and the intuitive belief of advertisers that humour in ads
enhances persuasion (Madden & Weinberger 1984). Considerable
anecdotal evidence suggests that humorous advertisements can be
effective in selling products in many diverse product categories such as
soft drinks, cars and insurance (Markiewicz 1974). In addition,
research in advertising has investigated the effects of humorous
advertisements on many other response variables such as memory,
advertising liking, brand attitude and purchase intention. Even though
there has been considerable research, the findings fail to show a
systematic effect of humour on recall, recognition and ad liking.
Furthermore, few studies have focused on the differences of humour
effects across product categories. A rule of thumb among advertising
practitioners is to avoid humorous advertising for high-involvement
products because the results may be counterproductive. Because the
advertising objective is often to get high brand recall through high
attention, it is important that advertising practitioners understand the
exact effects of humour appeal compared with non-humorous
advertising. If humour is indeed more effective at grabbing attention,
supporting high recall and aiding message comprehension, advertising
practitioners may be wise to add a few laughs to their advertising
messages.

Researches about effects of humour
In 1973, Sternthal and Craig drew some tentative but useful conclu-
sions about the effects of humour in advertising by reviewing the early
literature on humour in general. Even though the literature they
reviewed is small, and not specific to advertising, their conclusions
about the often positive effects of humour eased the way for future
studies on the effects of humour. After Sternthal and Craig’s study,
Murphy and colleagues (1979) studied the effect of TV programme
types on the recall of humorous TV commercials. They found that the
programme environments within which humorous ads appear affect
the performance of both ads and items in tests of unaided recall. They

MODERATING ROLE OF PRODUCT INVOLVEMENT

119

Chung.qxd 28/01/03 13:45 Page 119

1

Footnote.

found that overall ad recall is much higher for humorous ads than for
non-humorous ads. Unlike Murphy et al.’s study, which was done in a
laboratory setting, Cantor and Venus (1980) tested the effect of
humour in radio ads on memorability and persuasiveness in a quasi-
natural setting. The results of their study also support the general
conclusions drawn by Sternthal and Craig.

Most advertising practitioners already used humorous advertise-

ments to promote products and services. According to Markiewicz
(1974), humorous advertising on TV and radio accounted for as much
as 42% of the total. A survey of executives in leading agencies
revealed that 90% of the respondents believed that humour enhances
advertising effects (Madden & Weinberger 1984). Further, it was
estimated that 24% of prime-time television advertising in the USA
used humorous messages (Weinberger & Spotts 1989).

Madden and Weinberger (1982) studied the effects of humour on

attention levels, but, unlike previous studies, they used magazine
advertisements to test the effects of humour. They also tested whether
the potential heightening of attention is moderated by audience
factors such as race and gender. They found that humorous
advertisements outperformed normal ads on each recall category.
Gelb and Pickett (1983) tried to find out whether humour in an ad
influenced cognitive components (e.g. ad liking/disliking, attitude
towards ad, attitude towards brand, and purchase intention of
advertised product), as well as attention and recall. They found a
relationship between the perception of humour in an ad and a positive
attitude towards the ad, although the direction of causal flow is
unknown. They also found a positive relationship between attitude
towards brand and perceived humour. However, the perception of
humour in an ad was not related to purchase intention. Belch and
Belch (1983) found similar results. They found that humorous
messages are evaluated more favourably by the audience than serious
messages, and they produce more positive perceptions of advertiser
credibility, more favourable attitudes towards the ad, and more
favourable cognitive responses. However, attitude towards using the
advertised product (in this case the product was the services of
Federal Express) and purchase intention were not affected differently
by serious vis-à-vis humorous messages.

Lammers and colleagues (1983) also tried to understand the

persuasive effects of humour by using trace consolidation theory.
They hypothesised that a more humorous message would increase

120

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 120

1

Footnote.

persuasion-related measures (cognitive responses and attitude) in the
long run. It was found that when cognitive response measures were
taken immediately after subjects were exposed to ad materials, there
was little difference between the serious and humorous ads. However,
when cognitive response measures were delayed, the humorous appeal
produced more cognitive responses than the serious appeal. It was also
found that most of the increased cognitive activity came in the form
of pro-argumentation. They concluded that humorous appeal may be
more effective than serious appeal because humour, in the long run,
stimulates more favourable cognitive responses. Sutherland and
Middleton (1983) also tried to expand the effects of humour to
include message credibility as well as recall. They found, however, that
although humour can attract audience attention, there is no difference
between straight and humorous appeals in terms of recall of the
advertising message. Moreover, they found that straight messages are
more likely to be judged as credible than humorous messages and that
straight messages have more authority than humorous messages. Thus
Sutherland and Middleton’s study produced totally different results
compared with previous studies of recall and credibility.

Duncan and colleagues (1983) re-examined the effects of humour

on advertising comprehension by focusing on type of humour
measurement (manipulated vs. perceived) and humour location in the
advertisement. Their results also confirmed the results of previous
studies about the effects of humour on advertising comprehension.
Duncan and Nelson (1985) also found that humour can increase
attention paid to an ad, improve advertising liking, reduce irritation
experienced from the commercial and increase product liking. Just as
in previous studies, however, humour did not have any influence on
purchase intention. They concluded that humorous ads seem to be
more appropriate for generating awareness than for generating
persuasion or purchase intention.

Recently, advertising scholars used different approaches to study

the effects of humour by focusing on the role of moderating or
mediating variables on the effects of humour, such as advertising
repetition, prior exposure to ad messages and audience size. Zhang
and Zinkhan (1991) studied the effects of humour in ads in relation to
ad repetition and size of audience. They found that humorous ads
tend to produce higher levels of perceived humour, positive brand
attitude and brand information recall. However, ad repetition has no
influence on perceived humour and overall effectiveness of

MODERATING ROLE OF PRODUCT INVOLVEMENT

121

Chung.qxd 28/01/03 13:45 Page 121

1

Footnote.

advertising. Further, Zhang (1996) studied the effects of humour in
print ads using ‘need for cognition’ as a mediating variable and found
that the effect of humour is moderated by individual differences in
need for cognition. Chattopadhyay and Basu’s study (1990) found that
the effect of humour on consumer attitude and choice behaviour was
moderated by the message recipient’s prior evaluation of the
advertised brand. Therefore, when prior brand evaluation is positive,
humorous ads are more effective than non-humorous ads and vice
versa.

In sum, previous research has failed to prove consistently superior

persuasive effects of humorous ads over non-humorous ads. The
absence of empirical results contrasts with humour’s widespread use
in many different products (Markiewicz 1974) and the intuitive belief
of advertising practitioners that humour in ads enhances persuasion
(Madden & Weinberger 1984). Most studies measuring the effects of
humorous ads on recall and comprehension suggest that findings are
mixed; that is, some found positive effects and others found negative
effects. However, most studies of source credibility and liking of
source found that humorous ads have a positive influence. Finally,
several studies found that humorous ads do not have a positive impact
on choice behaviour, such as purchase intention.

THEORETICAL BACKGROUND

Cognitive and affective effects of humour
Humour’s effects on the cognitive process have usually been measured
in terms of memory and comprehension. In advertising research, the
emphasis has been on memory rather than on comprehension
(Du Plessis 1994). Advertising researchers have identified recall and
recognition as processes that access memory traces of commercial
messages. Although the recall and recognition to measure advertising
effectiveness is a long-standing debate (see Du Plessis (1994) for a
review), the fundamental difference between the two is that recall is
measured by asking subjects to specify the stimulus without aid,
whereas recognition is measured by asking subjects to identify whether
they have seen or heard the stimulus before. Krugman (1986) argues
that recall and recognition measures are different in nature and
suggests that the advertising industry has failed to make the
distinction.

122

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 122

1

Footnote.

There is no simple way to decide which method is most useful. In

some situations, advertising research requires either recall or
recognition measures; in other cases, both recall and recognition are
required. The threshold theory posits that recall and recognition
measure the same memory but that recognition requires a lower
threshold of familiarity (Kintsch 1970). However, according to the
dual-process hypothesis (Anderson & Bower 1972), recall consists of
two steps – memory search and recognition. In this sense, recognition
is a sub-process of recall. To recall items, a subject generates possible
candidates for recall during the search process and then selects items
through recognition. It is therefore a logical explanation that
recognition is less sensitive than recall and understandable that
recognition scores are substantially higher than recall scores. Thus to
gain higher recall, a stronger encoding process and more frequent
exposure is needed.

Humour’s effects on recall and recognition may be explained by

operant conditioning theory. As Nord and Peter (1980) explain,
operant conditioning occurs when the probability that an individual
will emit one or more behaviours is altered by changing the events or
consequences that follow the particular behaviour. Unlike
information-processing theory, operant conditioning views humour as
a reward for listening to the advertising message (Phillips 1968).
Therefore, a humorous advertisement could be better understood and
recalled than a similar non-humorous advertisement because humour
was a positive reinforcement. This better memory may also be
explained by the positive impact of emotional arousal (effect) to
memory. Ambler and Burne (1999) posit that if consumers are
emotionally aroused while watching commercials, those commercials
are more likely to be recalled by them. Thus they argue that advertising
with high affective components is more likely to be remembered by
consumers. In this sense, it is possible that consumers can be
emotionally aroused through watching humorous advertisements and
this emotional arousal in turn affects consumers’ memory over
advertisements. Another possible rationale for the effects of humour
is Helson’s adaptation-level theory (1959), which deals with the
capacity of a stimulus to attract attention. Each stimulus that an
individual encounters becomes associated with an adaptation or
reference level. Thus attention is attracted when the individual
perceives the focal stimulus to be plainly different from its reference
stimuli. In this case, humour specific to an advertising context or

MODERATING ROLE OF PRODUCT INVOLVEMENT

123

Chung.qxd 28/01/03 13:45 Page 123

1

Footnote.

perceived as exceptional will be noticed because, in general, unique
advertisements are learned and recalled better than non-humorous
commercials. Therefore, the first hypothesis regarding the humour
effect is as follows:

H1:

Degree of perceived humour in an advertisement will be
positively related to the unaided and aided recall.

As indicators of advertising effectiveness, Attitude towards the ad

(hereafter Aad), Attitude towards the brand (hereafter Ab) and
Purchase Intention (hereafter PI) are usually examined. Many studies
have reported Aad as a mediator of advertising effects on Ab and PI
(Mitchell & Olson 1981; Lutz 1985; MacKenzie et al. 1986; Holbrook
& Batra 1987). In 1981, Mitchell and Olson first introduced the notion
that consumers’ choice behaviour is likely to be influenced by attitude
towards the advertising stimulus. Mitchell and Olson (1981) proposed,
and found empirical support for, the mediational effects of attitude
towards the ad. They suggested that Aad should be considered as
distinct from beliefs and brand attitudes. Using a classical conditioning
approach, they reasoned that the pairing of an unknown brand name
(unconditioned stimulus) with a highly valenced visual (conditioned)
stimulus probably causes the transference of affect from ad to brand.
Researchers have since shown that Aad, which is defined as an
affective construct representing feelings of

favourability/

unfavourability towards the advertising itself, mediates the effects of
advertising content on Ab and consumers’ Acb (Attitude towards
choice behaviour) (Mitchell & Olson 1981; Shimp 1981; Lutz 1985;
MacKenzie et al. 1986; MacKenzie & Lutz 1989). This mediating role
of Aad has been found continuously in many other consumer studies
(Belch & Belch 1983; Gelb & Pickett 1983; Park & Mittal 1985;
Zinkhan & Zinkhan 1985; Park & Young 1986; Zhang 1996).

Recently, however, some studies have found the reverse relationship

between Aad and Ab (see e.g. Madden & Ajzen 1991). That is, in some
cases, consumers’ prior attitudes towards the brand also influence
positively or negatively their attitudes towards the advertisement of
that brand (Machleit & Wilson 1988). Thus, in a familiar brand,
attitude towards the advertisement will be influenced by consumers’
prior attitude towards the brand.

In the advertising area, the Advertising Research Foundation (ARF)

Copy Research Validation project has emphasised the role of ‘liking’ a
commercial as an important evaluative measurement (Haley &

124

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 124

1

Footnote.

Baldinger 1991). The basic question relating to commercial liking is
whether likeable advertising is inherently more effective than less
likeable advertising. In broader terms, there are two primary rationales
to explain how ad liking might contribute to advertising effectiveness.
The first has to do with cognitive processing. If consumers like the
advertising they are more likely to notice and pay attention to the ads
and more likely to assimilate and respond to the advertising message.
The second rationale has to do with affective response. According to
Lutz’s (1985) affect transfer model, if consumers experience positive
feelings towards the advertising, they will associate those feelings with
the advertiser or the advertised brand. Thus the more the ad is liked,
the more positive feelings are created towards the brand. As seen in
previous studies, several advertising scholars have found that
perceived humour in an advertisement has an impact on the message
receiver’s attitude towards the ad (Belch & Belch 1983; Gelb & Pickett
1983). That is, the more humour the receiver perceives in the
advertisement, the more favourable attitude towards the ad the
receiver has. This finding is also confirmed by Chung and Zhao
(2000). Thus the second hypothesis is suggested:

H2:

Degree of perceived humour in an advertisement will be
associated positively with the attitude towards the ad.

Moderating role of product involvement
In the advertising research area, involvement has a long history. First,
Krugman drew the involvement issue to the forefront of advertising
research. Applying learning theory, Krugman (1965, 1977) found that
people remembered better those ads which were presented first and
last. Krugman (1965) argued that advertising actually had low levels of
involvement. He also operationalised the involvement as the number
of ‘bridging experiences’, namely connections or personal references
per minute that the viewer made between his own life and the
advertisement. Since Krugman’s seminal argument about television
advertising, the construct of involvement has emerged as an
important factor in studying advertising effectiveness (Wright 1973;
Krugman 1977; Rothschild 1979; Petty & Cacioppo 1981a, 1981b;
Petty et al. 1981; Petty et al. 1983; Greenwald & Leavitt 1984). In these
studies, involvement usually refers to: personal relevance to the
message and product (Petty & Cacioppo 1981; Engel & Blackwell
1982; Greenwald & Leavitt 1984); arousal, interest, or drive evoked by

MODERATING ROLE OF PRODUCT INVOLVEMENT

125

Chung.qxd 28/01/03 13:45 Page 125

1

Footnote.

a specific stimulus (Park & Mittal 1985); a person’s activation level
(Cohen 1982); and goal-directed arousal capacity (Park & Mittal 1985;
Park & Young 1986).

The variables proposed as the antecedents of involvement may be

divided into three categories. The first relates to the characteristics of
the person, the second relates to the physical characteristics of the
stimulus. Thus involvement will be different according to the types of
media or content of the communication. The third category relates to
the situation. For example, the person’s involvement will be different if
he or she watches the advertising when planning to buy that product.
These three categories are usually used for ascertaining involvement.
Among these proposed antecedents, the second and third categories
were based on the assumptions that involvement is activated by
external stimulus (Taylor & Joseph 1984).

Although involvement has been recognised as an interaction

between individual and external stimuli, product involvement has been
defined as ‘salience or relevance of a product rather than an
individual’s interest in a product’ (Salmon 1986). Recently, researchers
divided product involvement into two distinct types. The first type is
situational involvement

, which reflects product involvement that occurs

only in specific situations. The second type is enduring involvement, which
represents an ongoing concern with a product that transcends
situational influences (Houston & Rothschild 1978; Rothschild 1979).
All these constructs are focused mainly on the external stimulus rather
than on an individual’s general interest in a product.

Analysing individuals’ common interest in a product is very

important in the sense that marketers and advertisers need some
baselines to segment markets according to consumers’ product
involvement. In this sense, the construct ‘product involvement’ has a
meaning that may be used for the majority of consumers. Therefore,
the term ‘product involvement’ used in the business area has a very
different meaning compared with those constructs that are focused
mainly on relations between individual and specific external stimuli.

‘Product involvement’ is often used interchangeably with ‘perceived

product involvement’ in the marketing literature (Kapferer & Laurent
1985). The meaning and definition of ‘product involvement’ differ
across researchers. For example, Cushing and Douglas-Tate (1985)
defined ‘product involvement’ as ‘how the product fits into that
person’s life’ (p. 243). To them, product involvement is a sort of
degree of importance to a person. To Zaichkowsky (1985), product

126

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 126

1

Footnote.

involvement is referred to as the relevance that individuals perceive in
the product’s values according to their own interests and needs.
Similarly, Tyebjee (1979) describes product involvement as strength of
belief about the product class, but others characterise involvement in
the product class as the relevance or salience of a product class to
receivers (Mitchell 1979; Greenwald & Leavitt 1984; Zaichkowsky
1985).

The Elaboration Likelihood Model (hereafter ELM: Petty &

Cacioppo 1981a, 1986a, 1986b) posits that persuasion can occur via
two routes – the central and peripheral routes. The central route
requires a person’s cognitive elaboration of advertising message
(persuasive message in their studies), and the peripheral route occurs
in the absence of cognitive elaboration for those persuasive
arguments. According to ELM, a person’s processing of information
differs by his or her level of involvement. When consumers have high
MAO (Motivation, Ability and Opportunity) to process communi-
cation, they are willing or able to exert a lot of cognitive processing
effort, which is called high-elaboration likelihood.

On the contrary, when MAO is low, consumers are neither willing

nor able to exert a lot of effort. However, a person’s elaboration
likelihood is also influenced by situational variables such as product
type. That is, a high-involvement product situation would enhance a
person’s motivation for issue-relevant thinking and increase a person’s
‘elaboration likelihood’, so the central route to persuasion will
probably be induced. A low-involvement product situation would
probably create low consumer motivation to process information,
which leads to greater possibility of a peripheral route to persuasion.
Therefore, we expect that a humorous message in an advertisement
will work as a peripheral cue so that it is more effective for a low-
involvement rather than a high-involvement product. That is, a
consumer is less motivated to process information for a low-
involvement product and is thus more likely to form an attitude
towards the ad based on peripheral cues such as a humorous message
that we expect to function as a peripheral cue. Conversely, the
humorous advertisement is less likely to affect consumers with a high-
involvement product since consumers are more motivated to expend
cognitive processing effort for high-involvement products.

Thus the following hypotheses are suggested:

MODERATING ROLE OF PRODUCT INVOLVEMENT

127

Chung.qxd 28/01/03 13:45 Page 127

1

Footnote.

H3:

Humorous ads will be more effective in a low-involvement
product than in a high-involvement product in terms of
attitude towards the advertisement.

H4:

Humorous ads will be more effective in a low-involvement
product than in a high-involvement product in terms of
memory of advertised brand.

METHOD

Telephone survey
The data on attitude towards the ad and memory were collected via a
telephone interview and then aggregated across respondents.
Telephone interviews were conducted in Chapel Hill, North Carolina
from 1992 to 1996, and in Minneapolis, Minnesota in 1997. Three
telephone interview sessions were conducted from Monday through
Wednesday evenings following the Super Bowl games. Graduate and
undergraduate students enrolled in research classes conducted
telephone surveys of local residents. Random-digit dialling was used
to include unlisted numbers. The interviewers asked for the person
who had the next birthday in the household. If a call yielded no
answer, the number was redialled at least three times before being
discarded. No respondents knew beforehand that we would be
conducting the interviews after the games, so the viewing situation
was completely natural.

Measurements
Wells and colleagues (1992) state that advertising plays several
different roles: the marketing role, the communication role, the
economic role and the societal role. However, these different roles are
all based on the function of providing information for different
purposes. Advertising imparts information that triggers consumer
needs, provides information for buyers and helps consumers make
wise decisions based on comparing product features. All purchase
decisions involve the memory of alternative brands. Therefore, a
rational, conscious brand choice decision depends on memories of the
brands. For this reason, all market or advertising research involves
some sort of memory test because memory is one of the most

128

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 128

1

Footnote.

important measurements for advertising effectiveness. Further,
Attitude towards the ad (Aad) is a frequently employed measure of
advertising effectiveness. Many studies have explored the role of Aad
as a mediator of advertising’s Attitude towards the brand (Ab) (e.g.
Mitchell & Olson 1981; Lutz et al. 1983; Lutz 1985; Edell & Burke
1987; Holbrook & Batra 1987). And, the hierarchy-of-effects para-
digm (e.g. Lavidge & Steiner 1961; Preston 1982) has also explored
Aad’s role in the purchase process.

Following the traditional measurement for advertising effectiveness,

this study also measures memory and attitude towards the
advertisement as two important dependent variables.

MEMORY

The unit of analysis in this study is each brand advertised during the
games. Memory was measured by unaided recall and aided recall.
Following an instruction, interviewers asked each respondent whether
he or she had watched the Super Bowl game, and which part. Those
who watched any part of the game were asked to list all
advertisements they remembered seeing during the game. Two coders
coded the response separately, which had been recorded verbatim
during the interviews. The two sets of results were in agreement in all
but one case (more than 99%). The unaided recall rates were then
calculated by dividing the number of respondents who recalled the
brand (Rü) by the number of respondents who watched the segment
(s) in which the brand was advertised (W¾).

After the unaided recall measure, aided recall was measured.

Respondents were given a list of brand names and asked if they
remembered seeing an advertisement for that brand during the game.
The aided recall rate was then calculated by dividing the number of
respondents who said they remembered seeing an advertisement for
each brand (Gü) by the number of respondents who watched the
segment (s) in which the brand was advertised (W¾). Although the
memory measure used here is not exactly the same as the day-after
recall (DAR) used by advertising agencies or research companies, it has
been validated in a series of previous studies (Zhao 1997; Chung &
Zhao 2000).

ATTITUDE TOWARDS THE AD

The second dependent variable used in this study was attitude towards
the ad. In experimental design, researchers have used several

MODERATING ROLE OF PRODUCT INVOLVEMENT

129

Chung.qxd 28/01/03 13:45 Page 129

1

Footnote.

dimensions to measure attitude towards the ad. For example, several
studies have used a four-item index: good–bad, like–dislike, irritating–
not irritating and uninteresting–interesting (Mitchell & Olson 1981;
Gardner 1985; Mitchell 1986). Batra and Ahtola (1991) support the
argument that attitude towards the ad is not only one-dimensional.
Instead, they found two dimensions of attitude: hedonic and
utilitarian. However, other studies have used only a one-item index
such as like–dislike or favourable–unfavourable (Burke & Edell 1986;
Edell & Burke 1987).

For this study, we focused on the degree of advertising likeness.

Our purpose was to find whether or not the perceived humour affects
advertising liking. Therefore, it was unnecessary to measure other
dimensions of attitude in this study. Attitude towards the ad was
measured by asking those respondents who remembered seeing an ad
how likeable or dislikeable they thought the ad was. Nine-point Likert
scales ranging from 1 – ‘it was one of the best’ – to 9 – ‘it was one of
the worst’ – were used. To facilitate interpretation (to obtain the same
scale with unaided and aided recall), all the scores were linearly
transformed to 0–100; here 100 represents the most likeable and 0
represents the least likeable. Those scored were then averaged across
respondents for each brand for each year.

Independent variable

DEGREE OF HUMOUR

An independent variable for this study was perceived humour in an
advertisement. Some studies have found that different types of
humour have different effects in terms of attention and memory (see
e.g. Madden 1982; Speck 1991). For instance, Speck (1991) divided
humorous messages into five different categories – comic wit,
sentimental humour, satire, sentimental comedy and full comedy – and
found that the effects of humour ranged from strongly positive for
full comedy to an essentially null effect for sentimental humour. Some
studies used five- or six-item semantic questions to measure the degree
of humour in an advertisement. However, in this study humour was
treated and measured as a unitary form since we focused on only
comic wit type of humour which is used most frequently in
advertising.

To measure the degree of perceived humour in each advertisement

aired during the Super Bowl game, undergraduate students in a large
southern university were used. Even though it has been found that

130

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 130

1

Footnote.

younger people are more likely than older people to rate advertise-
ments as humorous and that younger people’s categories of humour
are different from those of older people, students were used for
methodological convenience since it was impossible to ask each
respondent who watched the game and advertisement to rate the
degree of perceived humour in each ad.

All students watched the taped advertisements through a big screen

in a computer laboratory room and rated the degree of perceived
humour in an advertisement with a nine-point Likert scale (1 is least
humorous and 9 is most humorous). In addition, to obtain the same
scale as other variables, all degrees of perceived humour were linearly
transformed to a 0–100 scale, where 100 represents the most
humorous and 0 represents the least humorous. Those scores were
then averaged across respondents for each brand for each year.

Moderating variable

PRODUCT INVOLVEMENT

For this study, the Foote, Cone and Belding (FCB) grid for 60
common products (Ratchford 1987) was used as a guideline to divide
the products advertised during the game into two different categories:
high-involvement product and low-involvement product. Two
graduate students were trained to categorise the products according to
an FCB grid for 60 common products. Each worked independently
and categorised all products advertised during the game according to
the same FCB grid for 60 common products. After finishing
categorising the products, they exchanged their work and checked
whether there were differences. In fact, almost all the products
advertised during the Super Bowl game are within the range of 60
common products or similar products to those of 60 common
products used by Ratchford (1987). Therefore, no problem exists with
regard to categorising products according to the FCB grid.

Control variables
There are several other variables that can explain significant amounts
of variance in the dependent variables, such as ad frequency. A brand
that places more ads should be more likely to influence the dependent
variables (in particular memory), and this is also more likely to
truncate the effects of humour in ads. Even though several scholars
have found that attitude towards the ad decreased with increased

MODERATING ROLE OF PRODUCT INVOLVEMENT

131

Chung.qxd 28/01/03 13:45 Page 131

1

Footnote.

advertising exposure (high frequency) (Messmer 1979; Burke & Edell
1986; Machleit & Wilson 1988), ‘mere exposure’ theory posits that
people are likely to give higher attitude ratings to the repeated exposed
stimuli. Further, higher frequency is likely to be associated with better
memory. Ad frequency is therefore controlled in the regression
analysis as a continuous variable. We also considered the variable ‘year’
as another confounding variable, since we recognised that when we
pooled the data there was a chance that differences between years
could confound the effects of humour in each ad during the Super
Bowl game. We therefore created five dummy variables, year 1993
through year 1997 (1992 serves as a comparison).

RESULTS

Data screening
The data were analysed through Systat version 9.0 and were screened
before the analysis began. The results found no severe univariate
outliers. Though descriptive statistics found the unaided recall to be
highly skewed, it is natural and reasonable to expect a highly skewed
recall score (Jin & Zhao 1999; Chung & Zhao 2000). Since regression
analysis is very sensitive to outlying cases, all the statistics for finding
outlying and influencing cases were worked out. Apart from unaided
recall, there were no significant outlying and influencing cases in terms
of leverage, studentised deleted residual, and Cook’s distance (see
Appendix for details). In terms of unaided recall, one case was identi-
fied as a highly influencing case. However, because the percentage of
Cook’s distance for this case (17.8%) belongs to moderate range (10 to
50%: Neter et al. 1999), this case was not deleted from the dataset.

Descriptive statistics for data
Table 1 shows the descriptive statistics for the dependent variables –
memory and attitude towards the ad – independent variable,
moderating variable and control variable. The highest score for
unaided recall was 87% and the lowest was 0, and the mean was
6.54%. For aided recall, the highest brand had 78.31% and the lowest
brand had 8.61%, with a mean of 29%. The least favourable brands
had an attitude score of 35% and the most favourable brands had an
attitude score of 92%. For the dependent variable, the least humorous

132

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 132

commercial had 6 points and the most humorous commercial had 92
points in terms of degree of humour.

Hypotheses tests

HYPOTHESES 1

The first hypothesis proposes a positive relationship between a
humorous advertisement and subjects’ memory of an advertised
brand. Using degree of humour as an independent variable, simple
regression was done to check the relationship between memory and
humour. As hypothesised, humour and memory have a positive
relationship (Table 2).

Simple regression shows the significant regression model ( p <

0.001), and regression coefficients were both significant ( p < 0.001)
and positive for unaided and aided recall. Total variations that can be
explained by an independent variable were 22% for unaided recall and
26% for aided recall. Humour effect on memory was further
investigated including control variable such as frequency. In this
analysis, we also included dummy coded year variables since there is a
possibility that a different year has a different degree of humour.

MODERATING ROLE OF PRODUCT INVOLVEMENT

133

TABLE 1

DESCRIPTIVE STATISTIC FOR VARIABLES

Min

Max

Mean

SD

Median

Skewness

Kurtosis

Attitude towards

the ad

35.00

91.57

58.45

8.98

57.54

0.279

0.958

Unaided recall

0.00

87.00

6.54

16.42

0.00

3.336

11.168

Aided recall

8.61

78.31

29.00

14.92

24.42

1.172

1.028

Ad frequency

1.00

7.00

1.61

1.12

1.00

Humour

6.64

91.01

50.45

21.47

47.27

0.142

–1.041

TABLE 2

HUMOUR EFFECTS ON UNAIDED AND AIDED RECALL

Unaided recall

Aided recall

Coefficient

Beta

Coefficient

Beta

Constant

–11.632

10.931***

Humour

0.360***

0.471***

0.358***

0.515***

Total R²

0.222

0.266

Adjusted R²

0.219

0.263

*p < 0.05; **p < 0.01; ***p < 0.001

Chung.qxd 28/01/03 13:45 Page 133

1

Footnote.

Humour including those control variables was also regressed on both
unaided and aided recall (Table 3). As expected, degree of humour in
a commercial related positively to both unaided and aided recall above
and beyond the control variables.

Year variable (1992 was used as a comparison year) shows no

significant effect on dependent variable, which means no significant
differences among different years. Also, as expected, ad frequency has
a positive relationship with memory and ad frequency itself explains
30% and 27% of total variations for unaided and aided recall, respec-
tively. Humour effect, when control variables were included, became
smaller (10% for unaided recall and 15% for aided recall), but still had
significant and positive coefficients for memory. Therefore, the first
hypothesis suggesting a positive relationship between memory and
humour was supported.

HYPOTHESES 2

The second hypothesis also proposes a positive relationship between
humour and attitude towards the advertisement. Following simple
regression of humour on attitude, multiple regression including
control variables was also run (Table 4). As shown in the table,
humour itself explains almost 9% of variation in attitude towards the
advertisement and had a significant positive coefficient for attitude

134

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

TABLE 3

HUMOUR EFFECTS ON UNAIDED AND AIDED RECALL
INCLUDING CONTROL VARIABLES

Unaided recall

Aided recall

Constant

–18.707

6.015**

1993

0.419

0.164

1.188

0.030

1994

3.742

0.089

–0.879

–0.023

1995

–0.307

–0.007

–0.002

–0.001

1996

2.348

0.050

1.253

0.030

1997

2.063

0.049

–0.557

–0.015

Frequency

6.593***

0.451***

5.415***

0.408***

Humour

0.261***

0.341***

0.280***

0.404***

R² by year

0.009

0.002

R² by frequency

0.298***

0.271***

R² by humour

0.106***

0.149***

Total R²

0.414

0.422

Adjusted R²

0.400

0.408

*p < 0.05; **p < 0.01; ***p < 0.001

Chung.qxd 28/01/03 13:45 Page 134

1

Footnote.

towards the advertisement. Unlike for memory, ad frequency explains
only 1% of variation but it was significant. Therefore, Hypothesis 2,
suggesting positive relations between attitude and humorous message,
was also supported by the data.

HYPOTHESES 3 AND 4

Hypotheses 3 and 4 predict the possible moderating effect by product
involvement. In this study, product involvement was not measured
directly from surveyed individuals. Instead, categorical product
involvement was used. That is, advertised products were categorised
into high-involvement and low-involvement products based on a
frequently used FCB grid. Further, to test hypotheses 3 and 4, the
amount of effect of humour on memory and attitude was compared
for high- and low-product involvement. Table 5 shows the amount of
total variations explained by humour across high- and low-product
involvement. As shown in the table, humour explains many more
variations in low-product involvement. In terms of unaided recall
there are 11.1% differences between high and low-product
involvement, and for aided recall there are almost 13% differences. For
attitude towards the ad, a humorous message explains 10% more
variations in low-involvement products.

MODERATING ROLE OF PRODUCT INVOLVEMENT

135

TABLE 4

HUMOUR EFFECTS ON ATTITUDE TOWARD THE AD
INCLUDING CONTROL VARIABLES

Unaided recall

Constant

51.119***

1993

1.815

0.076

1994

–0.512

–0.022

1995

0.557

0.022

1996

–0.223

–0.009

1997

1.070

0.047

Frequency

0.267

0.033

Humour

0.128***

0.305***

R² by year

0.005

R² by frequency

0.014*

R² by humour

0.085***

Total R²

0.105

Adjusted R²

0.084

*p < 0.05; **p < 0.01; ***p < 0.001

Chung.qxd 28/01/03 13:45 Page 135

1

Footnote.

CONCLUSION AND DISCUSSION

This study examined the relationship between a humorous advertise-
ment and memory and attitude, and the role of product involvement
in this relationship. Overall, strong positive relationships were found
between a humorous advertisement and memory of advertised brand
and attitude towards the advertisement. Further, it was found that
those positive relationships were much stronger within low-
involvement products than within high-involvement products.

All the research hypotheses were supported in our data. The simple

and multiple regressions for hypotheses 1 and 2 show that humour in
a television commercial does appear to have some positive effects on
unaided and aided recall and attitude towards the ad. Our findings
imply that most humorous advertisements during the 1992 to 1997
Super Bowl games did a good job in these respects.

As the advertising environment has become increasingly crowded,

humour in advertising appears to be playing a larger role in helping ads
stand out. In addition, most previous studies were done in forced
exposure environments using print ads (two researchers used
television ads in forced exposure); therefore, the subjects’ attention
levels could have been higher than in a natural television viewing
situation. The fact that we allowed attention to vary naturally in our
study may partially explain the larger effects we observed. We might
therefore infer that the ‘believers’ in humour advertisements are right.
Indeed, despite the contradicting opinions of other researchers and
advertisers, it can be a powerful tool for attention-grabbing in some
sense since research has suggested that enhanced attention leads to
more extensive processing, which in turn leads to higher memory
(Petty & Cacioppo 1985).

These findings have some practical implications for advertising

practitioners. First of all, using humorous advertisements will be more
effective in the highly cluttered environment of broadcasting, in

136

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

TABLE 5

MODERATING EFFECTS OF PRODUCT INVOLVEMENT

Unaided recall

Aided recall

Attitude towards ad

R² by humour²

R² by humour

R² by humour

High involvement

3.5***

4.1***

2.1***

Low involvement

14.6***

16.9***

10.3***

*p < 0.05; **p < 0.01; ***p < 0.001

Chung.qxd 28/01/03 13:45 Page 136

1

Footnote.

particular television, to win higher attention from the audience and to
further increase memory of the brand and a favourable attitude
towards the advertisement. Humorous advertisements worked well for
gaining favourable attitudes from the audience, and worked well even
for increasing memory of the advertised brand. Scholars have found
that affective reaction to stimulus will increase attitude both towards
the ad and towards the brand. Even though this study did not measure
attitude towards the brand, it seems possible that humorous
advertisements can elicit a favourable attitude towards the advertise-
ment and further increase favourable attitude towards the advertised
brand (Batra & Ray 1985; Edell & Burke 1986). A second implication
is related to the role of product involvement. Findings suggest that
even though the effects of humour were statistically significant in both
high- and low-involvement products, the effects of humour in high-
involvement products were small and marginal.

As expected, humorous messages work very well in low-

involvement products. This phenomenon may be explained fully by
ELM; that is, humorous messages can serve as a peripheral cue and
work better only in a low-involvement situation. Therefore, advertising
practitioners should be very cautious about using humorous
advertisements for high-involvement or high-risk products. In some
sense they can be distracting elements for those who have high-
product involvement.

Limitations
Several limitations should be considered in interpreting the findings of
this study. First, this study did not consider the effects of news
coverage on brand recall and recognition. Jin and Zhao (1999) suggest
that news coverage can explain more than 70% of the variance for
brand recall and 60% of the variance for brand recognition. That is,
brand recall and recognition are influenced heavily by news coverage
of the advertised brand. In this context we should consider the
characteristics of the Super Bowl game. Because the Super Bowl is the
most visible advertising event and has tremendous media coverage,
Super Bowl advertising also draws special media attention. Therefore,
Super Bowl advertising will have wide media coverage before it is
aired; thus the higher recall and recognition score may be due in part
to the media coverage received. Second, in this study humour was
treated and measured as a unitary form. However, as Speck (1991)
pointed out, different types of humour may have a different humour-

MODERATING ROLE OF PRODUCT INVOLVEMENT

137

Chung.qxd 28/01/03 13:45 Page 137

1

Footnote.

type effect. Third, recall and recognition rates may have been affected
by ads aired either before or after the Super Bowl game. Some national
advertisers use new advertisements only for the Super Bowl, while
others rerun their famous or favourite advertisements during the
Super Bowl. Fourth, brand familiarity may be an important variable
that can influence dependent measures. More familiar brands are more
likely to be remembered. Fifth, methodological concerns may exist
because we used university students to rate the humorous advertising,
while recall and recognition and ad liking were measured among city
residents. It was found that younger people are more likely than older
people to rate advertisements as humorous and that younger people’s
categories of humour are different from those of older people. These
differences may work against our positive findings between humorous
advertisements and memory and attitude.

APPENDIX: SCREENING DATA

To see whether there are outlying and/or heavily influencing data,
leverage, studentised deleted residual, and Cook’s distance were used
to evaluate the outlying and influencing cases in the data.

First, the average leverage was calculated for all three variables.

Average leverage for unaided recall, aided recall, and attitude towards
the ad was 0.0705 (11/156), so it was considered larger in terms of
leverage if the leverage of case exceeds twice the average leverage.
Therefore, several cases were identified as having high leverage.
However, those cases had smaller leverage than 0.5 (usually considered
as a cut point; see Neter et al. 1999). So in terms of leverage no case
was identified as an outlying case.

Second, the studentised deleted residual (defined as ti = di/s{di})

was calculated to see whether there was any outlying case. Bonferroni-
corrected critical value at

α = 0.05 was 2.998. In unaided recall, 6 cases

were identified as having bigger score in terms of studentised deleted
residual, but in other variables, no case was identified as having bigger
score.

Finally, the Cook’s distance was calculated, and the corresponding

percentile to the value of Cook’s score was used to see whether there
was a heavily influencing case. As a general rule of thumb, the case
that had more than 50% of Cook percentile was used as a major
influencing case (see Neter et al. 1999). All the cases identified as

138

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 138

1

Footnote.

outlying case in terms of leverage and studentised deleted residual
were not major influencing cases in terms of Cook. All the cases have
Cook’s percentile less than 50%, the largest percentile among cases
was only 17.8% (Cook’s distance 0.202). Therefore, no one case was
deleted for final analysis.

REFERENCES

Aaker, D.A. & Myers, J.G. (1987) Advertising Management. Englewood Cliffs, NJ:

Prentice-Hall.

Ambler, T. & Burne, T. (1999) The impact of affect on memory of advertising.

Journal of Advertising Research

, 39 (March/April), pp. 25–34.

Anderson, J.R. & Bower, G.H. (1972) Recognition and retrieval process in free

recall. Psychological Review, 79 (March), pp. 97–123.

Batra, R. & Ahtola, O.T. (1991) The measurement and role of utilitarian and

hedonic attitudes. Marketing Letters, 2(2), pp. 159–170.

Batra, R. & Ray, M. (1985) How advertising works at contact. In Psychological

Processes and Advertising Effects

, L.F. Alwitt & A.A. Mitchell (eds). Hillsdale, NJ:

Lawrence Erlbaum Associates, pp. 13–44.

Belch, G.E. & Belch, M.A. (1983) An investigation of the effects of repetition on

cognitive and affective reactions to humorous and serious television
commercials. Advances in Consumer Research, 11, pp. 4–10.

Burke, M.C. & Edell, J.A. (1986) Ad reactions over time: capturing changes in the

real world. Journal of Consumer Research, 13 (June), pp. 114–118.

Cantor, J. & Venus, P. (1980) The effect of humor on recall of a radio

advertisement. Journal of Broadcasting, 24(1), pp. 13–22.

Chattopadhyay, A. & Basu, K. (1990) Humor in advertising: the moderating role

of prior brand evaluation. Journal of Marketing Research, 27 (November),
pp. 466–476.

Chung, H. & Zhao, X. (2000) Does humor really matter?: Some evidence from

super bowl advertising. Paper presented in the Advertising Research Division
of the Association for Education in Journalism and Mass Communication,
Phoenix, AZ, 9–12 August.

Cohen, J.B. (1983) Involvement and you: 1000 great ideas. In Advances in Consumer

Research

, Vol. 10, ?.? Wilkie, R.P. Bagozzi & A.M. Tybout (eds). Ann Arbor, MI:

Association for Consumer Research, pp. 325–328.

Cushing, P. & Douglas-Tate, M. (1985) The effect of people/product relationship

on advertising processing. In Psychological Processes and Advertising Effects: Theory,
Research, and Applications

, L.F. Alwitt, & A.A. Mitchell (eds), Hillsdale, NJ:

Lawrence Erlbaum Associates, pp. 241–259.

Duncan, C.P. (1979) Humor in advertising: a behavioral perspective. Journal of the

Academy of Marketing Science

, 7(4), pp. 285–306.

Duncan, C.P. & Nelson, J.E. (1985) Effects of humor in a radio advertising

experiment. Journal of Advertising, 14(2), pp. 33–40.

MODERATING ROLE OF PRODUCT INVOLVEMENT

139

Chung.qxd 28/01/03 13:45 Page 139

1

Footnote.

Duncan, C.P., Nelson, J.E. & Frontczak, N.T. (1983) The effect of humor on

advertising comprehension. Advances in Consumer Research, 11, pp. 432–437.

Du Plessis, E. (1994) Recognition versus recall. Journal of Advertising Research, 34(3),

pp. 75–91.

Edell, J.A. & Burke, M.C. (1987) The power of feelings in understanding

advertising effects. Journal of Consumer Research, 14 (December), pp. 421–433.

Engel, J. & Blackwell, R.D. (1982) Consumer Behavior. New York: Dryden Press.
Fletcher, A.D. & Bowers, T.A. (1991) Fundamentals of Advertising Research (4th edn).

Belmont, CA: Wadsworth.

Gardner, M.P. (1985) Mood states and consumer behavior: a critical review. Journal

of Consumer Research

, 12 (December), pp. 281–300.

Gelb, B.D. & Pickett, C.M. (1983) Attitude-toward-the-ad: links to humor and to

advertising effectiveness. Journal of Advertising, 12(2), pp. 34–42.

Gelb, B.D. & Zinkhan, G.M. (1986) Humor and advertising effectiveness after

repeated exposures to a radio commercial. Journal of Advertising, 15(2),
pp. 15–20.

Greenwald, A.G. & Leavitt, C. (1984) Audience involvement in advertising: four

levels. Journal of Consumer Research, 11 (June), pp. 581–592.

Haley, R.I. & Baldinger, A.L. (1991) The ARF copy research validity project.

Journal of Advertising Research

, 31(2), pp. 11–32.

Helson, H. (1959) Adaptation-level theory. In Psychology: A Study of a Science,

I. Sensory Perception and Physiological Formulations

. New York: McGraw-Hill.

Holbrook, M.B. & Batra, R. (1987) Assessing the role of emotions as mediators of

consumer response to advertising. Journal of Consumer Research, 14 (December),
pp. 404–420.

Holbrook, M.B. & O’Shaughnessy, J. (1984) The role of emotion in advertising.

Psychology and Marketing

, 1 (summer), pp. 45–64.

Hollis, N.S. (1995) Like it or not, liking is not enough. Journal of Advertising

Research

, 35(5), pp. 7–16.

Houston, M.J. & Rothschild, M.L. (1978) Conceptual and methodological

perspectives on involvement. In 1978 Educators’ Proceedings, S.C. Jain (ed.).
Chicago, IL: American Marketing Association, pp. 184–187.

Jin, H. & Zhao, X. (1999) Effects of media coverage on advertised brand recall

and recognition. In Proceedings of the 1999 Convention of the American Academy of
Advertising

, M.S. Roberts (ed.), pp. 54–61.

Kapferer, J. & Laurent, G. (1985) Consumers’ involvement profile: new empirical

results. In Advances in Consumer Research, Vol. 12, E.C. Hirschman & M.B.
Holbrook (eds). Provo, UT: Association for Consumer Research, pp. 290–295.

Kintsch, W. (1970) Models for free recall and recognition. In Models of Human

Memory

, D.A. Norman (ed.). New York: Academy Press, pp. 333–373.

Krugman, H.E. (1965) The impact of television advertising: learning without

involvement. Public Opinion Quarterly, 29, pp. 349–394.

Krugman, H.E. (1977) Memory without recall, exposure without perception.

Journal of Advertising Research

, 17(4), pp. 7–12

Krugman, H.E. (1986) Low recall and high recognition of advertising. Journal of

Advertising Research

, 26(2), pp. 79–86.

140

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 140

1

Footnote.

Lammers, H.B., Leibowitz, L., Seymour, G.E. & Hennessey, J.E. (1983) Humor

and cognitive responses to advertising stimuli: a trace consolidation approach.
Journal of Business Research

, 11, pp. 173–185.

Lavidge, R.J. & Steiner, G.A. (1961) A model for predictive measurements of

advertising effectiveness. Journal of Marketing, 25 (October), pp. 59–62.

Lutz, R.J. (1985) Affective and cognitive antecedents of attitude toward the ad: a

conceptual framework. In Psychological Processes and Advertising Effects: Theory,
Research, and Applications

, L.F. Alwitt, & A.A. Mitchell (eds). Hillsdale, NJ:

Lawrence Erlbaum Associates, pp. 45–56.

Lutz, R.J., MacKenzie, S.B. & Belch, G.E. (1983) Attitude toward the ad as a

mediator of advertising effectiveness: determinants and consequences. In
Advances in Consumer Research

, Vol. 10, R.P. Bagozzi & A.M. Tybout (eds). Ann

Arbor, MI: Association for Consumer Research, pp. 532–539.

MacKenzie, S.B. & Lutz, R.J. (1989) An empirical examination of the structural

antecedents of attitude toward the ad in an advertising pretesting context.
Journal of Marketing

, 53 (April), pp. 48–56.

MacKenzie, S.B., Lutz, R.J. & Belch, G.E. (1986) The role of attitude toward the

ad as mediator of advertising effectiveness: a test of competing explanations.
Journal of Marketing Research

, 23 (May), pp. 28–35.

Machleit, K.A. & Wilson, R.D. (1988) Emotional feelings and attitude toward the

advertisement: the roles of brand familiarity and repetition. Journal of
Advertising

, 17(3), pp. 27–35.

Madden, T.J. & Ajzen, I. (1991) Affective cues in persuasion: an assessment of

causal mediation. Marketing Letters, 2(4), pp. 359–366.

Madden, T.J. & Weinberger, M.G. (1982) The effects of humor on attention in

magazine advertising. Journal of Advertising, 11(3), pp. 8–14.

Madden, T.J. & Weinberger, M.G. (1984) Humor in advertising: a practitioner view.

Journal of Advertising Research

, 24(4), pp. 23–29.

Markiewicz, D. (1974) Effects of humor on persuasion. Sociometry, 37(3),

pp. 407–422.

Messmer, D.J. (1979) Repetition and attitudinal discrepancy effects on the affective

response to television advertising. Journal of Business Research, 7(1), pp. 75–93.

Mitchell, A.A. (1979) Involvement: a potentially important mediator of consumer

behavior. In Advances in Consumer Research, Vol. 6, W.H. Wilkie (ed.). Ann Arbor,
MI: Association for Consumer Research, pp. 20–24.

Mitchell, A.A. (1986) The effect of verbal and visual components of

advertisements on brand attitudes and attitude toward the ad. Journal of
Consumer Research

, 13 (June), 12–24.

Mitchell, A.A. & Olson, J.C. (1981) Are product attribute beliefs the only mediator

of advertising effects on brand attitude. Journal of Consumer Research, 18
(August), pp. 318–332.

Murphy, J.H., Cunningham, I.C. & Wilcox, G.B. (1979) The impact of program

environment on recall of humorous television commercials. Journal of
Advertising Research

, 18(2), pp. 17–21.

Neter, J., Kutner, M.H., Nachtsheim, C.J. & Wasserman, W. (1999) Applied Linear

Statistical Models

(3rd edn). New York: WCB/McGraw-Hill.

MODERATING ROLE OF PRODUCT INVOLVEMENT

141

Chung.qxd 28/01/03 13:45 Page 141

1

Footnote.

Nord, W.R. & Peter, J.P. (1980) A behavioral modification perspective on

marketing. Journal of Marketing, 44 (spring), pp. 36–47.

Park, C.W. & Mittal, B. (1985) A theory of involvement in consumer behavior:

problems and issues. In Research in Consumer Behavior, Vol. 1, J.N. Sheth (ed.).
Greenwich, CT: JAI Press.

Park, C.W. & Young, S.M. (1986) Consumer response to television commercials:

the impact of involvement and background music on brand attitude
formation. Journal of Marketing Research, 23 (February), pp. 11–24.

Petty, R.E. & Cacioppo, J.T. (1981a) Attitudes and Persuasion: Classic and Contemporary

Approaches

. Dubuque, IA: William C. Brown.

Petty, R.E. & Cacioppo, J.T. (1981b) Issue involvement as moderator of the

effects on attitude of advertising content and context. In Advances in Consumer
Research

, Vol. 8, K.B. Monroe (ed.). Ann Arbor, MI: Association for Consumer

Research, pp. 20–24.

Petty, R.E. & Cacioppo, J.T. (1985) Central and peripheral routes to persuasion:

the role of message repetition. In Psychological Processes and Advertising Effects,
L.F. Alwitt & A.A. Mitchell (eds). Hillsdale, NJ: Lawrence Erlbaum Associates,
pp. 91–111.

Petty, R.E. & Cacioppo, J.T. (1986a) Communication and Persuasion: Central and

Peripheral Routes to Attitude Change

. New York: Springer-Verlag.

Petty, R.E. & Cacioppo, J.T. (1986b) The elaboration likelihood model of

persuasion. In Advances in Experimental Social Psychology, Vol. 19, L. Berkowitz
(ed.). New York: Academic Press, pp. 123–205.

Petty, R.E., Cacioppo, J.T. & Goldman, R. (1981) Personal involvement as a

determinant of argument-based persuasion. Journal of Personality and Social
Psychology

, 41, pp. 847–855.

Petty, R.E., Cacioppo, J.T. & Schuman, D. (1983) Central and peripheral routes to

advertising effectiveness: the moderating role of involvement. Journal of
Consumer Research

, 10 (September), pp. 135–146.

Phillips, K. (1968) When a funny commercial is good, it’s great! Broadcasting,

74

(13), p. 26.

Preston, I. (1982) The association model of the advertising communication

process. Journal of Advertising, 11(2), pp. 3–15.

Ratchford, B.T. (1987) New insights about the FCB grid. Journal of Advertising

Research

, 27(4), pp. 24–38.

Ray, M.L. & Batra, R. (1983) Emotion and persuasion in advertising: what we do

and don’t know about affect. In Advances in Consumer Research, Vol. 10,
R.P. Bagozzi, & A.M. Tybout (eds). Ann Arbor, MI: Association for Consumer
Research, pp. 543–548.

Rothschild, M.L. (1979) Advertising strategies for high and low involvement

situations. In Attitude Research Plays for High Stakes, J. Maloney, & B. Silverman
(eds). Chicago, IL: American Marketing Association, pp. 74–93.

Salmon, C.T. (1986) Perspectives of involvement in consumer and communication

research. In Progress in Communication Sciences, Vol. 7, B. Dervin & M.J. Voight
(eds). Norwood, NJ: Ablex Publishing, pp. 243–268.

Shimp, T.A. (1981) Attitude toward the ad as a mediator of consumer brand

choice. Journal of Advertising, 10(2), pp. 9–15.

142

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 142

1

Footnote.

Speck, P.S. (1991) The humorous message taxonomy: a framework for the study of

humorous ads. In Current Issues & Research and Advertising, J.H. Leigh & C.R.
Martin (eds), The Division of Research Michigan Business School, The
University of Michigan, Ann Arbor, MI, pp. 1–44.

Spotts, H.E., Weinberger, M.G. & Parsons, A.L. (1997) Assessing the use and

impact of humor on advertising effectiveness: a contingency approach. Journal
of Advertising

, 26(3), pp. 17–32.

Sternthal, B. & Craig, C.S. (1973) Humor in advertising. Journal of Marketing, 37

(October), pp. 12–18.

Sutherland, J.C. & Middleton, L.A. (1983) The effect of humor on advertising

credibility and recall. In Proceedings of the 1983 Convention of the American Academy
of Advertising

, D.W. Jugenheimer (ed.), pp. 17–21.

Taylor, M.B. & Joseph, W.J. (1984) Measuring consumer involvement in products.

Psychology and Marketing

, 1(2), pp. 65–77.

Tyebjee, T.T. (1979) Response time, conflict, and involvement in brand choice.

Journal of Consumer Research

, 6, pp. 259–304.

Walker, D. & Dubitsky, T.M. (1994) Why liking matters. Journal of Advertising

Research

, 34(3), pp. 9–18.

Weinberger, M.G. (1992) The impact of humor in advertising: a review. Journal of

Advertising

, 21(4), pp. 35–59.

Weinberger, M.G. & Spotts, H.E. (1989) Humor in U.S. versus U.K.

TV advertising. Journal of Advertising, 18(2), pp. 39–44.

Wells, W.D., Burnet, J. & Moriarty, S.E. (1992) Advertising: Principles and Practice.

Englewood Cliffs, NJ: Prentice-Hall.

Wright, P. (1973) The cognitive processes mediating acceptance of advertising.

Journal of Marketing Research

, 10 (February), pp. 53–62.

Zaichkowsky, J.L. (1985) Measuring the involvement construct. Journal of Consumer

Research

, 12 (December), pp. 341–352.

Zhang, Y. & Zinkhan, G.M. (1991) Humor in television advertising: the effects of

repetition and social setting. Advances in Consumer Research, 18, pp. 813–818.

Zhang, Y. (1996) Responses to humorous advertising: the moderating effect of

need for cognition. Journal of Advertising, 25(1), pp. 15–32.

Zhao, X. (1997) Clutter and serial order redefined and retested. Journal of

Advertising Research

, 37 (September/October), pp. 57–73.

Zinkhan, G.M. & Gelb, B.D. (1987) Humor and advertising effectiveness

reexamined. Journal of Advertising, 16(1), pp. 66–68.

Zinkhan, G.M. & Zinkhan, F.C. (1985) Response profiles and choice behavior: an

application to financial services advertising. Journal of Advertising, 14(3),
pp. 39–66.

MODERATING ROLE OF PRODUCT INVOLVEMENT

143

Chung.qxd 28/01/03 13:45 Page 143

1

Footnote.

ABOUT THE AUTHORS

Hwiman Chung

received his Ph.D. from the University of North

Carolina at Chapel Hill. He has worked as an ad agency account
executive, handling diverse clients ranging from a cosmetic company
to an airline. His major interest is in consumer behaviour in the new
media environment, with a particular focus on structure and design
issues in www advertising.

Xinshu Zhao

received his Ph.D. from University of Wisconsin,

Madison. His areas of research interest are Super Bowl advertising,
public opinion, and consumer behaviour in the new media
environment. His research papers have appeared in many industry
journals.

144

INTERNATIONAL JOURNAL OF ADVERTISING, 2003, 22(1)

Chung.qxd 28/01/03 13:45 Page 144