AM809001 Marketing in Digital Age: A Creative Marketing Proposal

This marketing proposal critically evaluates an organization's existing offline and digital marketing communications strategies and develops a new or improved creative digital marketing communications campaign for a selected target market segment. It emphasizes the importance of consistency in brand messaging, particularly in the context of integrated marketing communications (IMC). The proposal explores the impact of brand familiarity on the effectiveness of consistent versus moderately consistent messages, drawing on theories from cognitive psychology and marketing. It suggests that for familiar brands, moderately consistent messages can enhance awareness and brand attitude, while for unfamiliar brands, highly consistent messages are more effective in building initial brand recall. The document also considers the cognitive effects of brand recall and brand associations, advocating for a balance between consistency and incongruity to foster consumer interest and engagement. This proposal is for the Auckland International Campus, AM809001 Marketing in a Digital Age course.

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/242347895
Revitalising brands through communication messages: The role of
brand familiarity
Article in European Journal of Marketing · February 2012
DOI: 10.1108/03090561211189220
CITATIONS
47
READS
1,550
3 authors:
Some of the authors of this publication are also working on these related projects:
IMC: effects on brand image and consumer behaviorView project
Call for papers: Special issues in Brand Management, Spanish Journal of marketingView project
Elena Delgado Ballester
University of Murcia
53PUBLICATIONS1,913CITATIONS
SEE PROFILE
MARÍA ÁNGELES NAVARRO-BAILÓN
University of Murcia
18PUBLICATIONS129CITATIONS
SEE PROFILE
Maria Sicilia
University of Murcia
56PUBLICATIONS832CITATIONS
SEE PROFILE
All content following this page was uploaded by Elena Delgado Ballester on 12 September 2014.
The user has requested enhancement of the downloaded file.

Revitalising brands through
communication messages: the role
of brand familiarity
Elena Delgado-Ballester,Angeles Navarro and Marı´a Sicilia
Marketing Department, University of Murcia, Murcia, Spain
Abstract
Purpose – From an integrated marketing communications perspective,this study aims to analyse
what level of consistency among brand messages is more effective in terms of customer-based brand
equity.In particular it aims to evaluate its impact on brand knowledge structure,and how brand
familiarity moderates this influence.
Design/methodology/approach – A sample of194 subjects participated in a between subjects
experiment. An integrated communication campaign composed of two different tools (advertising and
nonmonetary promotion) was assessed by individuals. Brand familiarity (familiar brand vs unfamiliar
brand)and consistency among messages (high vs moderate)were manipulated to testa set of
hypotheses.
Findings – Results show that the effectiveness of consistency among messages depends on brand
familiarity.For familiar brands moderately consistentmessages improve their awareness (recall),
enrich their network of associations,and generate more favourable responses and brand attitudes.
However, for unfamiliar brands, no significant differences are found between high and moderate levels
of consistency,except for brand recall,being higher when highly consistent messages are used.
Research limitations/implications – Limitations of the study are those typically associated with
the experimental methodology.Specifically,a single product category and only two communication
tools were used in the experiment which may limit the generalisability of the results.
Practical implications – For unfamiliar brands, brand managers should focus on consistent brand
messages to build awareness for these unknown brands.By contrast,for familiar brands the goal of
the communication strategy mustbe to revive the interestin them through moderate consistent
messages that can excite consumers and make them think again about these brands.
Originality/value – The originality of this study resides in incorporating the newest approach of
communication management(integrated marketing communication orIMC) to illustratehow
consistency among messages could be used to build the type ofbrand knowledge structure that
nurtures brand equity. Furthermore, compared to previous studies of IMC, which have addressed this
issue under a merely conceptualperspective,this paper offers empiricalevidences using a more
practical perspective and focusing on managing brand knowledge structures as a way for improving
brand image.
Keywords Integrated marketing communications, Consistency, Brand knowledge, Brand familiarity,
Information processing, Brand awareness
Paper type Research paper
Introduction
During the last decade,the concept of integrated marketing communications (hereafter
IMC) has made its way into the marketing and branding literature (Reid et al.,2005).
Recent voices (see Naik and Raman, 2003; Madhavaram et al., 2005) calls for IMC to help
firms in building and maintaining brand equity because “can help business move
forward in the highly competitive world of the twenty-first century” (Kitchen et al., 2004).
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/0309-0566.htm
Revitalising
brands
31
Received May 2009
Revised November 2009
January 2010
Accepted March 2010
European Journal of Marketing
Vol.46 No.1/2,2012
pp.31-51
q Emerald Group Publishing Limited
0309-0566
DOI 10.1108/03090561211189220

Centralto this conceptis the notion thatconsistency among allcommunication
messages is essentialto assemble the right brand-knowledge structure (Reid etal.,
2005).It refers to the existence of sharing common brand meaning among multip
means of communication (Keller,1996).In fact,a commonly held conception in the
branding literature is that building customer-based brand equity requires consis
brand meaning upheld by brand communications over time (Keller,1993).
However, from a schema-based approach, a growing body of research in field
cognitive psychology,social cognition,and marketing has proven the likely positive
communication effects derived from information incongruency (Alden etal.,2000;
Arias-Bolzmann et al., 2000). Being so, an interesting research question is what
consistency is more effective on building brand equity when developing an integ
communication campaign.
Brand familiarity may play an important role in this research question becaus
one of the most differentiating features among brands.In this sense there is ample
evidence that familiar brands have major communicative advantages over unfa
ones (Lange and Dahle´n, 2003).Drawing on the brand familiarity literature,and
differentConsistency Theories,this paperproposesthat the desirablelevel of
consistency depends on brand familiarity. Specifically, this paper provides answ
the following research questions:
. What levelof consistency among brand messages ismoreeffective when
building brand equity?
. Doesthe appropriatelevel of consistency depend on thelevel of brand
familiarity?
Theoretical framework
From a cognitive psychology perspective brands play an important role in deter
the effectiveness of marketing efforts due to the differential consumers’ reactio
presence of the brand (Keller, 1993). This occurs when consumers are familiar w
brand and hold a network of strong,favourable,and unique brand associations.In
such a case a successful brand management requires creating the desirable str
brand knowledge in consumers’minds.
Marketing communication programs play an important role in developing bra
knowledge because its objective is informing,persuading,and reminding consumers
aboutthe brand.Nevertheless,building brand knowledge is becoming harderas
current marketing communications are operating in an environment characteris
rising media costs, increasing communication options and audience fragmentat
response to this many organisations are considering how to improve the manag
of their communication programs using IMC (Reid et al., 2005). IMC advocates t
coordination of brand messages,from whatever source,must be consistent to ensure
brand image (Duncan and Moriarty,1998;Madhavaram etal.,2005).This idea of
consistency is also a commonly held conception in the branding literature,which
suggests that brand value is increased by consistent communication of the bran
time (see Keller,1993;Broniarczyk and Alba,1994).
From a strategic pointof view,consistency refers to the existence ofsharing
common brand meaning and contentamong multiplemeansof communication,
because the messages may be focused on a host of different tangible or intangi
aspects of the brand itself (Duncan and Moriarty,1998;Keller,2008).
EJM
46,1/2
32

Despite the basic recommendation that inconsistency should be avoided the truth is
that a growing body of research has observed some positive communication effects
derived from information incongruency.Several studies on advertising and consumer
behaviourliteratures have used Mandler’s (1982)schema incongruity processing
theory to analyse the effects of consistent vs inconsistent combinations of stimuli (see
De Pelsmacker et al.,2002;Lee and Thorson,2008;To¨rn and Dahle´n,2009).
While these studies have focused on the consistency,match or congruence among
two elements within a message (e.g. visual-verbal components, company-cause . . .), the
consistencybetweentwo differentcommunicationmessageshas not received
attention. Therefore, we use Mandler’s theory, other theoretical domains and empirical
findings from previous studies foranticipating the effects ofconsistency among
messages.Our position is that the desirable levelof consistency depends on brand
familiarity.
Research hypotheses
Brand familiarity reflects the extent of a consumer’s direct and indirect experience with
a brand (Alba and Hutchinson,1987).It capturesconsumers’brand knowledge
structures,being more limited and weaker for unfamiliar brands and stronger and
more sophisticated and accessible forfamiliarones (Campbelland Keller,2003).
Because of these knowledge differences,numerous studies have shown that familiar
brands have major advantages over unfamiliar brands in terms ofprocessing and
attitudes.Consumers develop less effortin processing information aboutfamiliar
brands,the information is more easily retrieved and stored,and these brands are
usually better liked (Kent and Allen,1994;Dahle´n and Lange,2004).
These same ideas can be drawn from recent studies on fluency (see Lee and Labroo,
2004;Labroo and Lee,2006;Novemsky et al.,2007) that have observed its effects on
awareness,affective judgment,and choice.As far as familiar brands are both more
perceptualand conceptualfluentthan unfamiliarbrands,becausethey can be
recognised and identified more easily and come to mind more readily, they enjoy more
cognitive and affective advantages.
Based on thepreviousreasoning,the currentresearch formulatesa set of
hypotheses to analyse the moderating role of brand familiarity on the effects derived
from a consistency-based integration strategy.We consider two levels of consistency
(high and moderate) in brand messages from different communication tools.From a
practicaland methodologicalperspectivewe havenot analysed a low levelof
consistency. From a practical point of view, a low consistency is not realistic because it
would represent a change in brand positioning too drastic and extreme so as to confuse
consumers,and weaken the brand equity that already exists.From a methodological
point of view,previous studies have mainly focused on the effects derived from high
and low levels of consistency while a moderate one has been ignored,being much
difficult to design.Therefore the focus on moderate and high levels of consistency
would be of more value for both academic and managerial purposes.
In particular, we delve into the effects on consumer knowledge structures because it
is important for brands to be recalled, to have a large number of associations (cognitive
effects),and to be favourably evaluated in terms ofbrand associations and brand
attitude (persuasive effects).
Revitalising
brands
33

Cognitive effects: brand recalland brand associations
At the heart of different consistency theories is the basic assumption that peopl
harmony in theirthoughts.Nevertheless,undera communication and persuasion
perspective,a mechanism frequently used to foster interest and motivation to proc
the messages is the use of a mismatch between stimuli. In this sense, previous
in advertising and consumer behaviour (see Meyers-Levy and Tybout,1989;Moore
et al., 2005; To¨rn and Dahle´n, 2009) have provided empirical evidences for the positiv
cognitive and persuasive effects of using inconsistent combinations of stimuli.
The theoreticalfoundations of these effects can be found in different theoretica
perspectives, although the schema congruity theory (Mandler, 1982) has been t
widely embraced.For example,Festinger’s (1957) cognitive dissonance theory posits
that inconsistency between two piecesof information producesdissonancethat
motivates people to seek ways of reducing it.Consistency theory (Hunt,1963)also
statesthat inconsistency causesarousaland that,depending on thedegreeof
inconsistency,the degree of arousalcan be either positive or negative.Unlike these
theories, the conceptual contribution of Mandler’s schema congruity theory is it
on how inconsistency is processed and resolved (Lee and Schumann,2004).
Based on Mandler (1982), when people are exposed to brand messages that
to their expectations there is little need to process information in-depth.This occurs
when there is a high consistency among messages thatcan be easily processed
employing few cognitiveresources.In contrast,the amountof elaboration and
processing thatpeople engage in willbe higher atmoderate levels ofconsistency
because people are motivated to resolve the lack of consistency. This resolution
be accomplishedby assimilating themoderateconsistentinformation without
prompting an important change in previous expectations.
In summary,a moderate consistency leads to a more extensive processing and
careful elaboration.As a result multiple routes for retrieval information are formed i
memory increasing the accessibility of the brand,which,in turn,enhances its recall
(Sjo¨din and To¨rn,2006).This effect is expected to occur mainly for familiar brands
because consumers have a betterand more sophisticated understanding oftheir
associations.This greaterknowledge comes to mind more readily because these
brands are more conceptually fluent (Labroo and Lee,2006)which,in turn,allows
consumers to better analyse new brand messages.As the new information received
does not completely fit with previous brand messages, cognitive processing incr
leading to the effects described above.
The opposite may occur for unfamiliar brands.The fact that they are perceptually
less fluentimplies thatonly highly consistentmessages willfavourbrand recall
because the same information serves as a cue for retrieving previous brand me
Being so,we propose that:
H1. Brand familiarity moderates the effectof brand messages consistency on
brand recall.
H1a.For familiarbrands,brand recallwill be higherat a moderate levelof
consistency than at a high level of consistency.
H1b.For unfamiliarbrands,brand recallwill be higherat a high level of
consistency than at a moderate level of consistency.
EJM
46,1/2
34

Furthermore, exposing consumers to moderately consistent brand messages is a more
interesting way to enrich the structure of associations for familiar brands because
people are more likely to engage in more extensive processing. However, for unfamiliar
brands, motivation to process information does not depend on the level of consistency
between messages because any incoming information will be perceived as novel and
relevant (Campbell and Keller,2003).Any information will be used to learn and form
an accurate impression of the brand and will be similarly processed due to the weaker
knowledge existing aboutthe brand.Being so,and compared to a high levelof
consistency,people have the same motivation to process the more diversified brand
associations in messages that are present in a moderate consistency level.Therefore,
there will be no differences in the number of brand associations evoked between high
and moderate levels of consistency.Based on this reasoning we hypothesize that:
H2. Brand familiarity moderates the effect of brand messages consistency on the
number of brand associations:
H2a.For familiar brands,the number of brand associations willbe higher at a
moderate level of consistency than at a high level of consistency.
H2b.For unfamiliar brands,there willbe no differences in the number of brand
associations between a moderate and a high level of consistency.
Persuasive effects: favourability of associations and brand attitude
Differentlevels ofconsistency may have also effects on the persuasiveness ofthe
communication campaign.Persuasiveness is related to the net favourableness of the
cognitive responses that people evoke as they elaborate on a message,as wellas to
brand attitudes(Meyers-Levy and Malaviya,1999).For example,Streufertand
Streufert’s (1978)generalincongruity adaptation levelhypothesized an “inverted U
shaped” curve reflecting the relationship between affect and incongruity. Furthermore,
the general prediction from schema theory is that messages that are incongruent are
evaluated more favourably than congruent or strongly incongruent messages as far as
people can successfully resolve the inconsistency (Sjo¨din and To¨rn,2006).There are
two main reasons that explain why solving inconsistencies produce positive responses
for the brand:
(1) The more thorough processing that occurs when there are inconsistencies leads
to enhanced confidence in resolution judgments, which is usually interpreted as
a greater linking of the brand (Lee,2000).This idea is related to the empirical
evidences provided by recent research in which emotions and positive affect
may arise not only as a result of the evaluation of information,but also from
how much effort is involved in resolving the incongruity (Orth and Holancova,
2003).
(2) Incongruentinformation producesa kind of entertainmentvalue,as it
challengesthe consumerto solve a puzzle.Solving thepuzzlethat the
incongruent stimulioffer should thus produce positive affect and a sense of
accomplishment (Peracchio and Meyers-Levy,1994).
Revitalising
brands
35

For these reasons we argue that solving inconsistencies creates a sense of enjo
and positiveaffectthat improvesbrand evaluation in termsmorefavourable
associations and brand attitude.
This effect can be accounted only in the context of familiar brands because, d
the factthatthey are perceptually more fluent,the effectof fluency on affective
judgment is not monotonic,and boredom may set it,leading to a decline in the brand
evaluation as the frequency of exposure arises (Machleit et al.,1993;Lee and Labroo,
2004).Therefore,the existence of some discrepancies can be seen as challenging a
interesting.Such situation is more likely to happen at a moderate level of consiste
among brand messages.
Unfamiliar brands do not suffer from consumer boredom.In this case,the most
important thing may be to facilitate fluent perception of these brands,irrespective of
the high or moderate consistency among brand messages because this fluency
more favourable attitudes (Novemsky et al., 2007). Hence, our next hypotheses
that:
H3. Brand familiarity moderates the effect of brand messages consistency on
favourability of brand associations.
H3a.For familiar brands,favourability of brand associations willbe higher at a
moderate level of consistency than at a high level of consistency.
H3b.For unfamiliar brands,there willbe no differences in the favourability of
brand associations between a moderate and a high level of consistency.
H4. Brand familiarity moderates the effectof brand messages consistency on
brand attitude.
H4a.For familiarbrands,the increase in brand attitude willbe higherat a
moderate level of consistency than at a high level of consistency.
H4b.For unfamiliar brands,there willbe no differences in the increase of brand
attitude between a moderate and a high level of consistency.
Method
Design of the main study
To test the hypotheses we used a 2 (consistency among messages: high, moder
(brand familiarity:familiar,unfamiliar)between subjects experimentaldesign.Soft
drinks were selected as the product category.
The level of consistency was manipulated through the design of two commun
tools (advertising and sales promotion). Advertising was used first to communic
set of associations of the brand schema and to create brand awareness and ima
Brand familiarity was manipulated by using two different brands.As we needed a
well-established brand schema for the familiar brand, a real one was selected (F
fictitiousbrand wasselected forthe unfamiliarbrand becausewe needed the
participants not to have a strong and sophisticated schema in memory (Sjo¨din and
To¨rn,2006).The fictitious brand was created using the same schema of associatio
evoked by thefamiliarbrand.This procedurewas followed to ensureinternal
consistency as wellas to have the same reference point from which to manipulate
consistency among brand messages.
EJM
46,1/2
36

Pretests
To design the levels of consistency several focus groups and pilot studies with different
samples of undergraduate students (see Figure 1) were conducted.We needed a brand
association schema from which the two levels of consistency were to be designed. For this
purpose, in a first focus group 25 subjects listed associations they held with the familiar
brand. The most commonly mentioned associations were “summer” (11 mentions), “study
trip” (8),“party” (4),“young people” (4),“enjoyment” (3) and “friends” (3).The fictitious
brand was created taking into account these associations. An advertising agency helped
us to select both the brand name and the logo for the fictitious brand.
The same agency also helped in designing two print ad versions (one version for
each brand)to communicatethe initial brand schemaand to facilitatebrand
identification (see Figure 2 the stimuli for the fictitious brand). Both versions included
the same background image and slogan.
An additionalstudy was conducted to testwhether the two printad versions
communicate the desirable associations. A total of 65 undergraduates were exposed to
one of the two ads (35 for Fanta and 30 for Fungo). They listed spontaneously all those
brand associations that came to their minds when they were exposed to the ad.Next,
they rated on a ten point-scale how well the ads communicated the ideas identified in
the first focus group.The results showed that the four most commonly mentioned
associations for each combination brand-ad were among those previously identified for
the familiar brand (see Table I).
When subjects were asked about those associations that represent the initial brand
schema we did not find significant differences between the two versions of the print ad,
except for “study trip” ( p ¼ 0.048) (see Table II). Therefore the selected fictitious brand
name was mostly coherent with the schema associated to the familiar one, meaning that
there are not significant differences in brand associations that could confound the results.
Figure 1.
Timing of the pretests
Revitalising
brands
37

The next step was to design the second stimulifor both brands (the promotion of a
specific activity) in order to manipulate the level of consistency among the mes
For this purpose, we needed to select two specific promotional activities that re
each levelof consistency (high and moderate).For the high levelof consistency the
activity had to communicatethe initialschema ofbrand associations described
previously.For the moderate levelof the promotionalactivity an alternative set of
associations has to be developed.
To make the design of the experiment simpler and to better control for the in
and attractiveness of the promotion,the reception delay of the premium and the way
people compete for a prize,the same type of promotion were kept constant across
Figure 2.
Advertising stimuli for the
unfamiliar brand
Associations Mentions
Fanta (n ¼ 35)
Summer 26
Party 26
Enjoyment 25
Sun/beach 20
Friends 13
Soft drink 14
Young people 13
Fungo (n ¼ 30)
Summer 24
Enjoyment 26
Young people 18
Party 20
Soft drink 14
Sun/beach 12
Happiness 7
Table I.
Most commonly
mentioned associations
for FANTA and FUNGO
EJM
46,1/2
38

conditions.We used sweepstakes because itis a very popular promotionaltoolto
enhancebrand image by associating the brand with an attractive prize thatis
meaningfully related tothe brand.In particular,the sweepsoffer peoplethe
opportunity to win a trip. We varied the destination of the trip with the aim of evoking
a different set of associations across the two consistency levels.
A second focus group (25 subjects) was used to elicit a list of possible destinations
that were coherentwith the initialbrand schema represented by Fanta.Next,in a
second pre-test,85 individuals assessed the perceived congruency between each
destination and the initial brand schema.The studies of Speed and Thompson (2000)
and Gwinner and Eaton (1999) were used to measure the level of congruency. The pair
composed ofFanta and the Caribbean as destination was found to be the most
consistent.The pair composed of Fanta and Tunisia as destination was the one that
best represents a moderate consistency between the image of Fanta and the one evoked
by Tunisia (see Table III).
We also checked that both brands did not have different congruency levels for each
destination.The results suggest that Fanta was as congruent with the Caribbean trip
as Fungo was (MFanta¼ 5.84;MFungo¼ 5.90 t93¼ 0.214 p ¼ 0.831).Similarly,Fanta
was as incongruent with the Tunisia trip as Fungo was (MFanta¼ 3.21; MFungo¼ 2.90
t82¼ 0.912 p ¼ 0.364).
Ad n Mean SD t
Summer Fanta 35 9.40 1.21 t(63)¼ 0.77
Fungo 30 9.17 1.20 p ¼ 0.44
Study trip Fanta 35 4.11 2.31 t(63)¼ 22.02
Fungo 30 5.47 3.08 p ¼ 0.05
Party Fanta 35 9.46 1.19 t(63)¼ 1.00
Fungo 30 9.10 1.64 p ¼ 0.31
Young people Fanta 35 9.34 0.83 t(63)¼ 0.96
Fungo 30 9.00 1.89 p ¼ 0.34
Enjoyment Fanta 35 9.66 0.48 t(63)¼ 0.97
Fungo 30 9.37 1.69 p ¼ 0.33
Friends Fanta 35 7.80 2.23 t(63)¼ 0.28
Fungo 30 7.63 2.56 p ¼ 0.78
Table II.
Mean value of
associations
communicated by ads
(T-test)
Destinations FANTA Mean Similarity (SD) FUNGO Mean Similarity (SD)
Greece Islands 4.37 (1.6) 4.43 (1.79)
The Caribbean 5.84 (1.3) 5.90 (1.62)
“Attractions parks tour” 2.63 (1.8) 3.17 (1.84)
Vatican 1.26 (0.58) 1.10 (0.29)
Mediterranean cruise 4.90 (1.8) 4.93 (1.82)
Tunisia 3.21 (1.6) 2.90 (1.35)
Centre of Europe 2.28 (1.5) 2.48 (1.50)
North of Europe 1.58 (1.1) 1.83 (1.03)
Table III.
Image based similarity
between brands and
different destinations
Revitalising
brands
39

The finalstep was to design the promotionalstimulifor both brands.Due to the
difficulty of exposing individuals to a promotion of a fictitious brand in a real con
they were exposed to a description of the specific promotional activity conducte
the brand (Coppetti et al., 2009; Cornwell et al., 2006). Figure 3 shows an exam
copy for the fictitious brand.
Sample and procedure
The experimenttook place in two separated sessions to avoid ceiling effects on
consumer’s memory.A sample of 194 undergraduate students was selected based o
their accessibility and the need of participating twice in the experiment.In the first
session,individuals were randomly exposed to one of the ads (familiar or unfamil
brand).After ad exposure,subjects wrote down all the thoughts they had while they
were exposed to this ad as well as their attitude towards the advertised brand.Two
days later individuals were exposed to the promotionalactivity for the same brand
they had been exposed to in the first session.For this second stimulus,individuals
were randomly exposed to a moderately or highly consistent promotionalstimulus
(Tunisia or Caribbean trip). After second exposure, individuals were requested t
a questionnaire.At the end of the second session studentswerethanked for
participating and were paid e5 each.
Measures
Four dependentvariables were assessed.Brand recallwas measured as the total
number ofvisualcharacteristics thatserve to identify the brand (e.g.name,logo,
slogan and so on) and that were correctly recalled by subjects (Edell and Keller,
Two independent judges counted these brand elements.Interjudge agreement was 98
per centand coding discrepancies were resolved through discussion.Brand recall
varies from 0 (for those individuals that did not correctly remember any charact
of the brand) to 5.
Number and favourability of brand associations were calculated using the tho
individuals had spontaneously reported after each exposure (Krishnan, 1996). B
so we can test how processing varies along the campaign.The total number of brand
associations was measured following the next formula: total number of associat
the first session plus new associations elicited in the second session. The total n
Figure 3.
Promotional stimuli
EJM
46,1/2
40

of brand associationsvariesfrom 4 to 16.To measurefavourability ofbrand
associations,each association was classified as positive,negative or neutral.This
categorisation was also done by the same two independent judges with an agreement
of 97.8 per cent. Following previous studies (Mackenzie et al., 1986), we operationalised
favourability as totalnetprocessing thatwas formed subtracting the numberof
negativeassociationsfrom thenumberof positiveassociationselicited by each
individual.The lowest favourability obtained was 2 2 (it means that there were more
negative than positive thoughts for this individual) and the highest was 13.
Finally, attitude towards the brand was measured with a semantic differential scale
(Mackenzie et al.,1986).Attitude was measured after each exposure (alpha ¼ 0.96 for
the first exposure, alpha ¼ 0.95 for the second exposure). Our dependent variable is the
attitudes reported after each exposure.
Manipulation check was assessed by evaluating perceived consistency among
brand messages,congruency among brands and the promotional activity,and finally
brand familiarity(seeAppendix).Perceived consistency wasassessed with a
seven-point Likert scale (Gwinner and Eaton, 1999; Speed and Thompson, 2000) with a
standardised coefficient alpha of 0.893.The measures of congruency between brand
and the promotionalactivity were adapted from the scale of perceived consistency
between messages (standardised coefficient alpha ¼ 0.924).Finally,brand familiarity
was measured on a seven-point semantic differential scale based on traditional scales
extractedfrom brand extensionliterature(Simoninand Ruth, 1998)with a
standardised coefficient alpha of 0.983.
Other variables were also measured in order to controlfor potentialconfounding
effects.Attitude towards the promotional activity was measured with two items on a
seven-pointLikert scale (Speedand Thompson,2000).For this measurethe
standardised coefficient alpha was 0.782. Attitude towards the specific destination was
measured with one item.
For all multi-item scales we used the mean value of the summed scale to test the
research hypotheses.Furtherdescriptivestatisticsfor all research variablesare
provided in Table IV.
Results
Manipulation checks
Before testing the hypotheses, we analysed the perceived consistency between messages
using a one-way ANOVA.Results showed that there was a significant difference in
perceived consistency (Mhigh¼ 5.62;Mmoderate¼ 4.44;F(1.191)¼ 50.858,p , 0.00).
Individuals also perceived a higher consistency for both brands when asked about
the Caribbean trip (high consistency condition) than when asked about the Tunisia trip
(moderate consistency).These differences were significant between conditions (Fanta:
Mhigh ¼ 5.22;Mmoderate¼ 4.65F (1.94)¼ 5.749p ¼ 0.018;Fungo: M high¼ 5.07;
Mmoderate¼ 4.20,F (1.96)¼ 11.454 p ¼ 0.001)though no differenceswereobtained
between brands ata high consistency level(t(93)¼ 0.715;p ¼ 0.476).This result
indicates that the fictitious brand successfully resembled the image of the familiar
brand.In addition,both brandsdiffered in termsof familiarity (MFanta¼ 6.90;
X Fungo¼ 1.52;F (1.189)¼ 2444.57;p , 0.00).
Finally,we also needed to controlfor possible differences in terms ofattitudes
towards trips,in general,and towards the specific destination in particular across
Revitalising
brands
41