Journal of Advertising

ISSN: 0091-3367 (Print) 1557-7805 (Online) Journal homepage: https://www.tandfonline.com/loi/ujoa20

Attitude toward the AD as a Mediator of Consumer

Brand Choice

Terence A. Shimp

To cite this article: Terence A. Shimp (1981) Attitude toward the AD as a Mediator of Consumer
Brand Choice, Journal of Advertising, 10:2, 9-48, DOI: 10.1080/00913367.1981.10672756

To link to this article: https://doi.org/10.1080/00913367.1981.10672756

Published online: 30 May 2013.

Submit your article to this journal

Article views: 1703

View related articles

Citing articles: 70 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=ujoa20
ATTITUDE TOWARD THE AD AS A MEDIATOR
OF CONSUMER BRAND CHOICE
Philosophers of science (e.g., 7, 13) have recognized that
the identification of new and potentially rich concepts often
is a spur to knowledge advancement. This article will hope-
fully provide a step in this direction by conceptualizing an
advertising approach termed "attitude toward the advertise-
ment" or AIT A. The AIT A approach is contrasted with
the more common "attitude toward the brand" (AIT B)
approach. A theoretical case is developed showing how the
AIT A approach influences brand choice; empirical evi-
dence in support of this case is presented; and suggestions
for future research are offered.

THE ATT 8 VS. ATT A ADVERTISING

APPROACHES
The ATT B AdvertisiDI Approach
The AIT B approach attempts to influence brand choice
Terence A. Shimp, an associale professor of markelilll al the University by engendering favorable consumer attitudes toward the
of South Carolina, received a D.B.A. in markeling from the University advertised brand. This is done by structuring ads to influ-
of Maryland. His research interests have been focused primarily on exam- ence consumers' beliefs and evaluations regarding the
inillllhe consumer behavior and public policy ramificalions of adverlisilll
favorable consequences of consuming the brand. Specific
strategies and methods. In addition 10 previous publicatiens in Ihis jour-
nal, his research has appeared in the Journal of Marketing, Journal of product attributes and benefits are emphasized. If done
Consumer Research, Journal of Appli«l Psychology, Journal of Con- successfully, favorable attitude formation and an increased
sumer Affairs, Journal of Adwrtising Research, as well as in other jour- probability of trial or repeat purchase are likely outcomes
nals and various conference proceedings. of the AIT B advertising approach.
The AIT B approach to advertising has solid grounding
Communicalions regarding Ihis paper should be addressed 10 Prof.
Terence A. Shimp, College of Business Adminislration, University of
in the theory of behavioral action (1, 8). It also receives
South Carolina, Columbia, South Carolina 29208. considerable support from advertising practitioners. For
example, the widespread view that the primary purpose of
advertising is to solve consumer problems (e.g., 6) implicitly
ABSTRACf calls for an AIT B approach to advertising. This approach
This article addresses the issue of how advertising influences consumers' also follows from the argument that creative copywriting
brand choices. A concept termed the attitude toward the advertisement involves the appropriate matching of product attributes/
approach (A1T A) is introduced; a theoretical case is developed showilll
why A1T A is an importanl mediator of brand choice; empirical evidence
benefits with consumer needs/wants (19). There is, however,
in support of this case is presented; and suggestions for future research a major limitation to the effective utilization of the AIT B
are offered. approach; namely, its opportunity for use is restricted to
those instances where the advertised brand truly possesses
INTRODUCfION relative advantages over competitive offerings. When the
This article concentrates on the advertising undertaken brand is essential1y equivalent to competitive offerings, i.e.,
by marketers of consumer packaged goods. How brand a parity product, the AIT A approach offers a viable alter-
choicebehavioris influenced by such advertising is the issue native.
of primary concern. A vast literature of conventional wis-
doms related to this issue has accumulated, but true under- The ATT A AdvertisiDI Approach
standing, i.e., scientific knowledge, is far from complete. Here the advertising is not directed at specific product
Probably more is known about the process underlying attributes/benefits, and the objective is not to influence
brand choice behavior than is known about the specific consumers' beliefs toward the brand per se, The effort is
advertising methods and techniques which facilitate this directed instead at creating a favorable attitude toward the
process. advertisement in order to leave consumers with a positive

9
feeling after processing the ad. Underlying the use of such Appreciation is extended to Professor Richard Lutz of the
advertising is the tacit assumption that consumers are UCLA marketing faculty for bringing this point to the
hedonistically motivated by the desire to feel good. (In author's attention. Although this argument is technically
contrast, the ATT B approach is guided by the implicit correct, the lumping together of a brand's advertisement
assumption of the consumer as a rational, systematic deci- with "real" product features would represent a crude com-
sion maker.) A highly desirable feature of the ATT A bination, which taken to its logical conclusion would also
approach is that it demands minimal cognitive processing justify including word-of-mouth influence as an attitudinal
from consumers, since affective reactions are relatively component. The resolution offered here is that the separate
independent of cognition (31). identities of ATT Band ATT A be maintained in the same
Several well-known advertising practitioners have way that Fishbein has distinguished between the Aact and
recently presented arguments in support of the importance subjective norm components when modeling behavioral
of engendering favorable consumer attitudes toward ads intentions. Some justification already exists for including
(2, 11, 27). Other practitioners apparently share this per- ATT A as a third component when modeling consumer
spective, since numerous package good advertisers employ purchase intentions (17).
this approach, with soft drink advertisements perhaps per-
sonifying it (recall, for example, the Mean Joe Green ad THEORETICAL RATIONALE FOR THE
for Coca-Cola, Pepsi's ad with the little boy rollicking with
ATT A APPROACH
puppies, or 7-Up's use of boxer Sugar Ray Leonard warm-
ing up with his son). The discussion to this point has viewed attitude toward
the ad as an advertising approach. The flip side of this view
The Duality of AIT A is to examine ATT A from the consumer's perspective--not
There actually are two relatively distinct dimensions of only is ATT A an advertising approach, it also is an effect
attitude toward the ad, one cognitive and the other engendered in consumers. Examining AIT A from the con-
emotional. In the former case consumers form attitudes sumer's perspective will enable a clearer understanding of
toward advertisements by consciously processing execu- the functioning and potential effectiveness of such adver-
tional elements. For example, a consumer may like a par- tising. The processing of advertising messages under varying
ticular ad because it employs a humorous execution, like degrees of involvement furnishes the theoretical rationale
another ad due to its use of an attractive endorser, but dis- for this discussion.
like a third because it is perceived as too clever. Alterna-
tively, an attitude toward an ad may result merely because Advertising Processing Involvement
the ad evokes an emotional response, such as a feeling of
Involvement represents the amount of arousal or interest
love, joy, nostalgia, or sorrow, without any conscious pro-
evoked by a particular advertisement (16). The amount of
cessing of executional elements.
arousal or interest determines the degree of attention devoted
It is to be expected that the cognitive and emotional
to an advertisement and also influences the processing strat-
dimensions of ATT A are non-equivalent in their impact
egy. One processing strategy used by receivers is to actively
on consumers, due to the different underlying mechanisms,
process advertising information so as to form an overall
one a conscious process and the other non-volitional.
evaluation of the advertised brand. Mitchell, Russo, and
Research is needed to better understand these differential
Gardner (18) term this a "brand evaluation strategy." They
effects. This paper's focus is directed elsewhere, however.
also identify a "nonbrand strategy" whereby the receiver
The distinction has been introduced merely for concep-
devotes attention only to nonbrand advertising content (e.g.,
tual clarity. In keeping with this clarification, the remaining
music, scenery, characters).
discussion relates strictly to the emotional dimension of
These two factors, attention and processing strategy,
ATTA· provide a useful framework for delineating different
degrees of involvement (cf. 9, 18). "High involvement"
A Final Clarification
arises when advertising receivers are both attentive to an
Some readers may question whether this concept of an advertising message and employ a brand evaluation strategy.
attitude toward an ad is really independent of an attitude The absence of either of these processing conditions is tan-
toward the advertised brand. Indeed, it could be argued tamount to low involvement, for which there are two dis-
from a Fishbeinian-perspective (1, 8) that an advertise- tinct forms. The canonical form, widely accepted by prac-
ment for a brand is actually nothing more than another titioners and scholars alike (e.g., 10, 14, 23), is "attention-
component of the advertised brand; the consumer's evalu- limited" low involvement where receivers attend neither
ation of an advertisement would thus simply represent the brand-specific information nor any other information
another input into the attitude toward the brand per se. contained in the advertisement.

10
 "Strategy-limited" is an alternative form of low ATT A only. As a result of strategy-limited low involve-
involvement that is characterized by high attention to non- ment processing, the consumer would not have formed any
brand features and limited, if any, attention to brand-spe- semantic memory for brand information, and thus there
cific information (9, 18). This may be due to distraction or could not be any ATT B formation. However, an attitude
simply because the receiver is uninterested in the advertised toward the ad could be formed and stored in episodic
product and finds it easier to listen to the music or view memory in the fashion of Krugman's (15) "picture-image"
the scenery rather than avoid exposure altogether. memory. The memory of course need not be restricted
to pictures; it could include memory for sounds or even
Attitude Formation
feel. The point is that attitude would be limited to the ad
Consumers' processing of an advertisement or ad cam- itself, since semantic memory relevant to the advertised
paign for a brand enables the potentiality of four types of brand would not have been influenced.
attitude formation. The first would result from high in- No Attitude Formation. There would be no possibility
volvement information processing; the second from mod- for attitude formation toward either the brand or toward
erate involvement; the third from strategy-limited low in- the ad as a result of attention-limited processing.
volvement; and the fourth from attention-limited low in-
volvement. These are displayed in Figure 1 and discussed Summary
briefly below. Consumers' processing of advertisements under strategy-
limited low involvement conditions has provided a theo-
FIGURE I retical rationale for the plausibility of attitude toward the
ad (ATT A) as a distinct type of attitude formation. The
Four Types of Potential Attitude Formation discussion now turns to the relation between ATT A and
Resulting from Processing an Advertisement brand choice.

Non-Brand Information THE ROLE OF ATT A IN BRAND CHOICE

Processed?
Brand Information Figure 2 portrays three mechanisms by which brand
Processed? Yes No choice is influenced. The first path reflects a typical high
Yes I. Potential for II. Potential for involvement or learning model of brand choice. That is,
attitudes to be attitude toward consumers' beliefs and evaluations are influenced from
formed toward brand brand only processing an advertisement or ad campaign; attitude
and toward ad (ATT B only) toward the brand is engendered; and the choice of whether
(ATTB&ATT A) or not to purchase the brand is contingent on how favor-
able the attitude is. The third path illustrates a situation
No III. Potential for IV. No potential for where brand choice is independent of advertising efforts,
attitude toward attitude toward and is based instead on the desirability of in-store p-o-p
ad only either brand or (point-of-purchase) material.
(ATT A only) ad The second path is the one of current focus, as it con-
cerns the influence of AIT A on brand choice. To explicate
this mechanism it will be assumed that brand choice
ATT Band ATT A. The consumer's beliefs or evalua- involves a decision making process (for an alternate view,
tions may be influenced as a result of processing brand- see 21). Inherent in the decision making perspective is the
specific information, thereby enabling the formation of question of how consumers select a particular brand from
ATT B. Simultaneous processing of non-brand features among a set of alternatives. Various choice heuristics have
also would enable attitude formation toward the ad itself, been proposed (4, 29, 30). The premise in most of these is
i.e., ATT A. that information and beliefs are integrated (e.g., conjunc-
ATT B only. Formation of an attitude toward the brand tively or in a compensatory fashion), and consumer choice
is unlikely, since it is doubtful that consumers would pro- becomes a matter of selecting that alternative which satisfies
cess just brand features while disregarding non-brand some tacit objective function.
information. The exception, however, would be a situation Affect-referral is a choice rule that makes a different
where a consumer is highly motivated to obtain specific assumption about choice behavior. According to this heur-
brand information (e.g., price) and upon, say, chance expo- istic, consumer choice may simply involve the elicitation
sure to a commercial for a particular brand in a product from memory of an overall evaluation or affect for each
class of interest, the consumer is vigilant for price infor- alternative rather than examining beliefs about attributes
mation while disregarding anything else. (29). This is most likely for choice behavior where con-

11
 FIGURE 2

Three Alternative Brand Choice Mechanisms

Attitude toward
~
Consumer beliefs
and evaluations
..
~
the brand
(ATT B)
(1) ~
..
~
I Advertising
message or
( Attitude
Transfer Brand
Choice
campaign
\ /

(2) .. Attitude toward

the ad
(ATT A)
I \
(3)
In-store
I p-o-p material
\ J

sumers prefer to engage in minimal processing (3). An attractively, or advertised in the context of other pleasant
important issue related to the affect-referral heuristic con- stimuli (music, scenery, characters, etc.), The mediated
cerns the manner by which overall evaluation or affect for stimulus consequences of these connotative responses rep-
alternatives is formed. The consumer's product usage resent unconditioned stimuli (DCS). Denotative responses
experience is one basis. Advertising processing experience also result from advertising exposure. The stimulus conse-
provides an alternative basis in those instances where the quences of the denotative responses stored in active memory
consumer has never tried a particular brand. That is, in serve as the to-be conditioned stimuli. The pairing of a
the absence of an alternative source for forming an evalua- connotative response (e.g., a feeling of joy or nostalgia)
tion of a brand, the consumer may simply transfer his feel- with a denotative response (e.g., "this is a brand of tooth-
ing for the ad to the brand. Learning theory offers a theo- paste I've never heard of before") will, if done properly,
retical rationale for this transference process. The follow- lead to a conditioned emotional affect toward the advertised
ing perspective is illustrative: brand.
The upshot is that consumers' connotative responses to
At the most general level, we learn to like (or have)
advertisements for untried brands provide one input, per-
favorable attitudes toward objects we associate with
haps the major one, into their overall global affect toward
'good' things, and we acquire unfavorable feelings
toward objects we associate with 'bad' things and, thus, likelihood of choosing an untried brand. There-
(8:217).
fore, the quality of advertising execution (i.e., quality in
the sense of generating favorable connotative responses or
How a consumer's attitude toward an advertisement might AIT N is a critical factor in determining consumers' choice
be associated with the advertised brand becomes apparent behavior. Nord and Peter's discussion of behavioral modi-
by substituting advertised brand for "objects" and by con- fication theory arrives at a similar conclusion regarding
sidering the advertisement as the "good" or "bad" thing. the conditioning effects of advertising (20, especially pp.
37-38).
A Classical Conditioning Perspective
Attitude conditioning offers a more elaborate explana- EMPIRICAL SUPPORT
tion of how ATT A is transferred to ATT B, thus facilitat- Empirical evidence in support of the role performed by
ing the use of affect-referral choice procedures (24, 25). attitude toward the ad is accumulating. Research dealing
Evaluative connotative responses, such as pleasant feelings, with mood formation offers indirect evidence, while three
are one result of consumers being exposed to products that recent studies performed in advertising contexts provide
are described appealingly in advertisements, portrayed direct insight.

12
Evidence from Mood Research (24, 25) by designing an experiment which manipulated
Research by social psychologists has demonstrated a rela- both the visual emphasis and the explicitness of verbal claims
tion among a person's mood, memory, and subsequent for two hypothetical brands of beer. The results revealed
behavior. One explanation is provided in the "accessibility" that the visual content was as capable as the verbal content
and "cognitive loop" hypotheses (12), which purport that in increasing subjects' brand attitudes. One implication of
a person who is in a good mood is more likely to retrieve pos- this finding, though somewhat conjectural in light of the
itive than negative material from memory; in turn, this research procedures, is that the subjects assigned to the
improved access to positive materials affects mood-relevant "high visual emphasis" advertising conditions formed posi-
behavior. tive attitudes toward the ad, which through an affective
The significance of this is that AIT A essentially is tanta- transference process led to positive attitudes toward the
mount to mood formation. An advertisement can place a brand.
consumer in a good mood, which, in accordance with the Mitchell and Olson Research. This experiment provides
accessibility and cognitive loop hypotheses, could later in- concrete evidence to support the mediational role per-
fluence brand choice. Consider the following hypothetical formed by ATT A. Mitchell and Olson (17) manipulated
scenario illustrating the process. advertising content by constructing four different ads for a
A consumer is particularly fond of a television commer- hypothetical brand of facial tissues. One ad, for example,
cial for a brand of wine she has never purchased. A good consisted entirely of a picture of a kitten, assumed to con-
feeling (positive mood) is evoked every time she views the vey the notion of softness, with no verbal content. Another
ad. When doing her routine grocery shopping, she recog- ad contained only verbal content that described the facial
nized the need to purchase wine in anticipation of a forth- tissue's softness.
coming party. Upon encountering the advertised wine on Subjects were exposed on multiple occasions to all four
display, a positive mood (or some portion of it) is activated ads, followed by response to standard attitude and pur-
from previously processing the commercial. Since purchasing chase intention scales and a measure of their attitudes
wine is a mood-relevant behavior of sorts (because the toward the advertisements per se. Results of various statis-
party setting at which it will be served is conducive to feeling tical tests revealed that subjects' brand attitudes and pur-
good), there is a greater than chance probability that this chase intentions were largely determined by their affect for
hypothetical consumer will select the advertised wine rather the ads themselves, rather than being solely dependent on
than some alternative brand. (Additional discussion of the their beliefs toward product attributes, as Fishbein-type
relevance of mood research to advertising will be pursued logic (8) would predict.
in a later section.) Shimp and Yokum Research. Where the Mitchell and
Olson experiment assessed the effect of ATT A on the ante-
Direct Evidence from Advertising Experiments
cedents to behavior, Shimp and Yokum (26) investigated
Three recent advertising experiments provide direct in- the effect on purchase behavior. Two experiments were per-
sight into the role perfomed by attitude toward the ad as a formed under simulated marketing conditions. Subjects
mediator of consumer brand behavior. had an opportunity to make repeated purchase selections
Rossiter and Percy Research. These researchers have of hypothetical brands of cola dispensed in cups from a
formulated in various writings (22, 24, 25) a "dual loop" vending machine. Choice was influenced by exposing sub-
theory of advertising effects. This dual loop consists of a jects to one of three advertisements, two of which were
verbal loop and a visual loop. Advertising's effect in the deceptive and one which was not. The formulation of the
former loop is achieved by influencing consumers' beliefs colas also was manipulated, thereby structuring experi-
regarding the advertised brand. The verbal loop, in other ments where two factors, advertising content and product
words, reflects what this paper has termed the AIT B adver- formulation, were manipulated.
tising approach. In addition to these manipulations, measures of subjects'
In contrast, the visual loop involves a visual imagery attitudes toward the advertisements also were obtained. A
mechanism, whereby advertising's effect, though not pre- procedure developed for professional copy testing by Wells,
cludingthe possibilityof belief impact, is obtained primarily by Leavitt, and McConville (28) was used. Subjects rated how
influencingconsumers' imagination of doing that which is vis- well each of eight words described the experimental ad to
ually portrayed in the advertisement. In other words, accord- which they were exposed. As factor analysis had supported
ing to the visual loop notion, consumers' attitudes can be the adequacy of a one factor solution, each subject's atti-
created or altered without the aid of advertising copy and tude toward the ad was calculated by summing his ratings
without any verbal belief processing occurring. The visual on the eight descriptive words.
imagery serves as an unconditioned stimulus which is trans- This sum score was treated as a covariate in analysis of
ferred to the product via a classical conditioning process. covariance tests. In each experiment two analyses were per-
The dual loop notion was tested by Rossiter and Percy formed, one with actual purchase behavior as the depen-

13
dent variable and another in which taste ratings served as ders a good mood, then this good mood, when and if re-
the criterion. Results from both analyses and in both experi- initiated by in-store p-o-p cues, should increase the proba-
ments indicated clearly that subjects' attitudes toward the bility of purchasing the advertised brand that originally
experimental ads were a more important determinant of instigated the mood.
their purchase behavior and their taste ratings than were An important unanswered issue remains. Past mood
the main experimental manipulations--the more favorable research has been performed in research settings where good
their evaluations of the advertisement, the more positive mood was initiated immediately prior to measurement of
their taste ratings and the greater their frequency of pur- relevant criterion variables, e.g., donating money. Untested,
chasing the advertised brand. however, is what happens in a situation such as advertising
where mood is instigated, say, several days prior to the time
RESEARCH SUGGESTIONS where the consumer when shopping is confronted with the
Additional research is needed to provide a more complete decision of whether to buy the advertised brand that pre-
understanding of the functioning and effectiveness of the viously had initiated a positive mood. Will in-store stimuli
attitude toward the ad advertising approach. re-instigate the mood? Will the previously positive mood,
if re-instigated, facilitate retrieval of positive material from
Experimental Manipulation memory? These and other queries need to be tested to deter-
Experiments by Mitchell and Olson (17) and Shimp and mine whether the indirect evidence from mood research per-
Yokum (26) measured subjects' attitudes toward the experi- formed in non-advertising contexts is truly relevant to ad-
mental ads and then treated the measure as a statistical vertising.
covariate. Though both experiments detected highly signi-
ficant effects for the covariate, cause and effect inferences The Joint Influence of ATT A and ATT B on Consumer
are precluded, since in both studies the experimental proce- Choice Behavior
dures were strictly correlational rather than experimental Advertising information processing permits the poten-
with regard to the attitude toward the ad covariate (cf. 5). tiality for attitude formation toward both the advertised
Future research will need to experimentally manipulate the brand and also toward the advertisement itself. These
levels of ATT A in order to test for causal relations. attitudes have a theoretical range from positive to negative
with a neutral midpoint. With the possibility of three gene-
Conditions Conducive to ATT A Effectiveness ral attitudes (positive, neutral, negative) toward the adver-
Mitchell and Olson (17) have proposed, but have not tised brand and three toward the advertisement, a total of
empirically established, three conditions under which nine attitude combinations can be conceptualized. This is
ATT A should play a particularly important mediational shown in Figure 3.
role: (I) for ads that arouse a strong affective reaction
due to their visual content; (2) when the advertising copy FIGURE 3
is written so as to evoke positive feelings; and (3) when Nine Theoretical Attitude Combinations
strategy-limited low involvement characterizes the infor- Attitude toward the advertisement
mation processing strategy. (ATT A)
Testing of each of these conditions is needed. Research Attitude toward the
procedures would be relatively straightforward. With regard brand (ATT B) Positive (+) Neutral (0) Negative (-)
to the first condition, for example, a large sample of com-
Positive (+ ) 1 (++) 2 (+0) 3 (+-)
mercials could be selected and a panel of judges would
evaluate their visual content in terms of potential for Neutral (0) 4 (0+) 5 (00) 6 (0-)
generating affective reactions. With reliable interjudge
ratings, it then would be possible to test predictions con- Negative (-) 7 (-+) 8 (-0) 9 (--)
cerning which ads should represent important mediators of
attitudes toward the brand and which should not. A variety of important research issues emanate from this
portrayal of joint attitudes. For example, although it
Mood Research in Advertising Contexts
would be expected that the cell 1 (+ +) combination would
The basic conclusion from mood research is that people in have the highest probability of influencing choice behavior
a good mood tend to see the positive side of things and and cell 9 (--) the lowest probability, predictions become
respond more positively than others to a variety of stimuli. less clear away from these extremes. Consider for illustra-
A tendency exists for people to respond to or remember the tion the cell 4 (0+), cell 5 (00), and cell 6 (0-) possibilities.
quality of what they have experienced rather than the indi- All three combinations reflect the absence of attitude for-
vidual stimuli. These conclusions have obvious implica- mation toward the advertised brand, but they vary with
tions for advertising. In particular, if an advertisement engen- regard to attitudes toward the ad.

14
 The question arises as to which combination would have 4. Bettman, James R. and Michel A. Zins. "Constructive Processes in
the highest probability of affecting choice behavior. A rea- Consumer Choice," Journal of Consumer Research, Vol. 4 (September,
sonable prediction, founded on the theoretical arguments 1977), pp. 75-85.
5. Carlsmith, J. MerriIl, Phoebe C. Ellsworth, and Elliott Aronson. Methods
explicated previously, would hold that the cell 4 (0+) of Research in Social Psychology. Reading, MA: Addison-Wesley, 1976.
combination would yield the highest probability of influ- 6. Dillon, Tom. "The Triumph of Creativity Over Communication,"
encing choice behavior. But what about the other possibil- Journal ofAdvertising, Vol. 4 (Summer, 1975), pp. 15-18.
ities? It could be argued, and many practitioners presum- 7. Dubin, Robert. Theory Building. New York: Free Press, 1969.
ably hold this view (cL2), that a negative response to an ad 8. Fishbein, Martin and leek Ajzen. Belief, Attitude, Intention, and
Behavior: An Introduction to Theory and Research. Reading, MA:
is superior to none at all. This of course is a debatable Addison-Wesley, 1975.
proposition that necessitates scientifically rigorous testing. 9. Gardner, Meryl P., Andrew A. Mitchell, and J. Edward Russo. "Chrono-
A variety of additional research questions are suggested metric Analysis: An Introduction and an Application to Low Involvement
by the structure in Figure 3. Only upon a thorough research Perception of Advertisements," in H. Keith Hunt, (ed.), Advances in
program, free of proprietary interests, will enlightenment Consumer Research, Vol. 5, Ann Arbor: Association for Consumer
Research, 1978, pp. 581-589.
of the true role of consumers' affect for advertisements be 10. Houston, Michael 1. and Michael J. Rothschild. "Conceptual and
known. This indeed is an issue of considerable importance, Methodological Foundations on Involvement," in Subhash C. Jain
since advertising practice, presumably based on reasonable (ed.), Research Frontiers in Marketing: Dialogues and Directions, Chicago:
understanding of which advertising techniques are effective, American Marketing Association, pp. 184-187.
is a step ahead of scientific knowledge. 11. "In 'Era of Emotions,' Researchers Should Count Feelings, Not
Numbers," Marketing News, Vol. 14 (September 19, 1980), pp. 1,10, 11
CONCLUSION (quotes from speech presented by Lawrence M. Light).
12. Isen, Alice M., Margaret Clark, Thomas E. Shalker, and Lynn Karp.
Theoretical arguments and empirical evidence were pre- "Affect, Accessibility of Material in Memory, and Behavior: A Cognitive
sented in support of the claim that attitude toward the ad Loop?" Journal of Personality and Social Psychology, Vol. 36 (January,
(ATT A) is an important mediator of consumers' choice be- 1978), pp. 1-12.
havior. Information processing under strategy-limited low 13. Kaplan, Abraham. The Conduct of Inquiry: Methodology for Behav-
involvement conditions provided an explanation of how at- ioral Science. New York: Intext Educational Publishers, Chandler Pub-
lishing Company, 1964.
tude formation toward the ad isrelatively distinct from attitude 14. Krugman, Herbert E. "The Impact of Television Advertising: Learning
formation toward the advertised brand. A classical condi- Without Involvement," Public Opinion Quarterly, Vol. 29 (Fall, 1965),
tioning explanation provided the theoretical rationale for pp. 349-356.
the transference of affect from the ad to the brand. The 15. Krugman, Herbert E. "Memory Without Recall, Exposure Without
Perception," Journal of Advertising Research, Vol. 17 (August, 1977),
notion of an affect referral choice heuristic afforded an
pp.7-12.
explanation of how attitude toward a brand founded only 16. Mitchell, Andrew A. "Involvement: A Potentially Important Mediator
an attitude toward the advertisement(s) for that brand pro- of Consumer Behavior," in William Wilkie (ed.), Advances in Consumer
vides consumers with a sufficient basis for trial purchase Research, Vol. 6, Ann Arbor: Association for Consumer Research, 1979,
of the advertised brand. pp.191-196.
17. Mitchell, Andrew A. and Jerry C. Olson. "Are Product Beliefs the
In addition to the theoretical rationale, empirical evi-
Only Mediator of Attitude Formation and Change? An Examination of
dence was presented demonstrating the effectiveness of the the Theoretical Logic of the Fishbein Model," Carnegie-Mellon University/
ATT A advertising approach. The most convincing empiri- Graduate School of Industrial Administration Working Paper #60-79-80,
cal evidence stems from observation of actual advertising 1979.
practices, where the emphasis on creating favorable attitudes 18. Mitchell, Andrew A., J. Edward Russo, and Meryl Gardner. "Strategy-
Induced Low Involvement Processing of Advertising Messages," Carnegie-
toward the ad are prevalent if not dominant. The observa-
Mellon University manuscript (July, 1980).
tion of advertising practices is corroborated by several 19. Murray, Thomas D. "Better Ads--From Wheels," Advertising Age
research undertakings. Three experiments have offered pro- (July 7, 1980), pp. 43ff.
vocative evidence that consumers' brand specific attitudes, 20. Nord, Walter R. and Paul Peter. "A Behavior Modification Perspec-
purchase intentions, and indeed purchase behavior are tive on Marketing," Journal of Marketing, Vol. 44 (Spring, 1980), pp.
36-47.
influenced greatly by their attitudes toward the ads which
21. Olshavsky, Richard W. and Donald H. Granbois. "Consumer Decision
engendered these effects. Making--Fact or Fiction?" Journal of Consumer Research, Vol. 6, (Sep-
tember, 1979), pp. 93-100.
REFERENCES 22. Percy, Larry and John R. Rossiter. Advertising Strategy: A Communi-
1. Ajzen, leek and Martin Fishbein. Understanding Attitudes and Predict- cation Theory Approach. New York: Praeger Publishers, 1980.
ing Social Behavior. Englewood Cliffs, NJ: Prentice-Hall, Inc., 1980. 23. Robertson, Thomas S. "Low-Commitment Consumer Behavior,"
2. Bartos, Rena. "Do We Really Have To Make Them Mad To Sell Them?" Journal ofAdvertising Research, Vol. 16 (April, 1976), pp. 19-24.
Paper presented at the 26th Annual Conference of the Advertising 24. Rossiter, John R. and Larry Percy. "Visual Imaging Ability As a
Research Foundation, March, 1980. Mediator of Advertising Response," in H. Keith Hunt (ed.), Advances in
3. Bettman, James R. An Information Processing Theory of Consumer Consumer Research, Vol. 5. Ann Arbor: Association for Consumer
Choice. Reading, MA: Addison-Wesley, 1979. Research, 1978, pp, 621-629.
continued on page 48

15
worthy features of the book, including material from Adver- Attitude Toward the Brand as a Mediator of Consumer
tising Age's 1977 annual issue on international ad agencies Brand Choice continued from page 15.
(Appendix A: Advertising Agencies Around the World) 25. Rossiter, John R. and Larry Percy. "Attitude Change Through Visual
and from Starch INRA Hooper's advertising expenditures Imagery in Advertising," Journal of Advertising, Vol. 9:2 (1980), pp. 10-
(Appendix B: International Advertising Expenditures and 16.
Exchange Rates). Appendix B includes more than Starch 26. Shimp, Terence A. and J. Thomas Yokum. "The Influence of Decep-
INRA Hooper data, however. Information from specific tive Advertising on Repeat Purchase Behavior," University of South Car-
olina Working Paper #86 (March, 1980).
nations is included, with information taken from various 27. Tauber, Edward M. "Six Questionable Assumptions in Ad Research,"
local sources. To a media enthusiast this would be the most Marketing News, Vol. 14 (September 19,1980), p. 9.
valuable part of the book for he or she could judge the 28. Wells, William D., Clark Leavitt, and Maureen McConville. "A Re-
relative importance of advertising media from country to action Profile for TV Commercials," Journal of Advertising Research,
country. Again, the material is presented in a very uneven Vol. II (1971),pp.II-17.
29. Wright, Peter L. "Consumer Choice Strategies: Simplifying Vs. Opti-
manner, but perhaps that is because advertising differs so mizing," Journal of Marketing Research, Vol. II (February, 1975), pp.
much from country to country. 60-67.
I have neglected the first part of the book. Let me correct 30. Wright, Peter L. and Frederic Barbour. "The Relevance of Decision
that now. It is called "An Overview of International Adver- Process Models in Structuring Persuasive Messages," Communication
tising and Marketing," and it consists of 21 chapters Research, Vol. 2 (July, 1975), pp. 246-259.
31. Zajonc, R. B. "Feeling and Thinking: Preferences Need No Inferen-
grouped into four categories: (1) Management and Plann-
ces," American Psychologist, Vol. 35 (February, 1980), pp. 151-175.
ing, (2) Advertising and Public Relations, (3) Media, and
(4) Research. Some of the articles are serious thought-
pieces, such as Stu Britt's "Standardization of Marketing Program-Commercial Separators in Children's Television:
for the International Market" from the Columbia Journal Do They Help a Child Tell the Difference Between Bugs
of World Business, while others are merely descriptive fluff, Bunny and Quick Rabbit? continued from page 25.
such as "International Advertising Grows," from Adver-
24. "Separator Device Guides Enacted by TV Code Board to Aid Sub-
tisingAge's special issue on The New World of Advertising.
scribers," Code News, National Association of Broadcasters, Vol. 12 (May
Others are specialized studies of specific products, such as 1979), pp. 1-2.
Green-Cunningham-Cunningham's "The Effectiveness of 25. "The Older They Get, the Less They Believe TV Ads, Says Study of
Standardized Global Advertising" from the Journal of Kids," Marketing News (April 8, 1977), p. 6.
Advertising, which examined soft drinks and toothpaste as 26. Ward, Scott. "Children's Reactions to Commercials," Journal of
perceived by university students of four nations. And there Advertising Research, Vol. 12 (April 1972), pp. 37-45.
27. Ward, Scott. "Consumer Socialization," Journal of Consumer Re-
are some public relations pieces as well, such as John S. W. search, Vol. I (September 1974), pp. 1-14.
Wasley's "The International Advertising Association." 28. Ward, Scott and D. B. Wackman. "Children's Information Process-
In looking back over what I have just written, I seem to ing of Television Advertising," in Peter Clarke, ed., New Models for
have emphasized the negative. That is misleading. The pos- Communication Research. Beverly Hills, CA: Sage Publishing Co., 1973.
29. Ward, Scott, D. B. Wackman and Ellen Wartella. How Children
itives greatly outweight the negatives. The book is excellent--
Learn to Buy. Beverly Hills: Sage Publishing Co., 1977.
a first-rate reference source to keep on your desk until it is 30. Wartella, Ellen and L. S. Ettema. "A Cognitive Developmental Study
revised again. It fills an important gap in the literature, of Children's Attention to Television Commercials," Communications
and for the most part it is well-written and well-organized. Research, Vol. I (1975), pp. 69-88.
It is like eating Chinese food, though--you get hungry for
more and more the deeper you get into it. And by the time
you get through, you realize the herculean task it must
have been to put this project together, and your admiration
for the editors and contributors grows and grows.
One final note. The facts contained in the book get out
of date rather quickly, so this reference book will lose its
value quickly. Therefore, the third edition should come
out rather soon. I hope there is a timetable for revision
that is rather short-run. And I hope the next revision will
not take another 15 years so that the reviewer of the third
edition will not have to paraphrase the Aunt Jemima Syrup
slogan again. Congratulations to all on a job well done.

DONALD W. HENDON
Creighton University

48