Professional Documents
Culture Documents
uk/wrap
A Thesis Submitted for the Degree of PhD at the University of Warwick
http://go.warwick.ac.uk/wrap/2452
This thesis is made available online and is protected by original copyright.
Please scroll down to view the document itself.
Please refer to the repository record for this item for information to help you to
cite it. Our policy information is available from the repository home page.
16
Elif Karaosmanoglu
TABLE OF CONTENTS
TABLE OF FIGURES
A'
LIST OF TABLES
Vii
DEDICATION
ix
ACKNOWLEDGEMENT
DECLARATION
Xi
ABSTRACT
Xiii
I INTRODUCTION
1.1.Relevance.
of theResearch
1.2.Aim of theResearch
1.3.TheContextandtheRespondent
Baseof theStudy
1.4.Methodology
andtheMethodsUsed
10
1.5.Contributionof theStudy
13
1.6.Outlineof theThesis
15
1.7.Definitionsof Constructs
andConcepts
18
11LITERATURE
REVIEW
22
2.1.Introduction
22
in Corporate
2.2.Paradigms
ImageandCorporate
IdentityStudies
23
22.1. Visuallgraphicdesignparadigm
23
2.2.2.Integratedcommunication
paradigm
24
2.2.3.Organisational
studiesparadigm
26
2.2.4.Marketingparadigm
28
2.2.5.Interdisciplinary
paradigm
31
36
38
42
44
46
48
2.6.1.1.Symbolism
48
2.6.1.2.Communication
51
2.6.1.3.Behaviour
53
2.62. Unplannedcommunicationjactors
55
2.6.2.1.Interpersonalcommunication
57
2.6.2.2.Intermediarycommunication
58
2.6.2.3.Intrapersonalcommunication
59
67
2.8. Summary
71
III CONCEPTUAL
FRAMEWORK
73
3.1. Introduction
73
76
3.2.1.Symbolism
79
3.2.2.Communication
80
3.2.3.Behaviour
81
3.3. Unplannedcommunicationfactorsandcorporateimage
83
3.3.1.Interpersonalcommunication
83
3.3.2.Intermediarycommunication
84
3.3.3.Intrapersonalcommunication
85
ii
87
3.3.3.2. Consumer-companyidentification
88
89
91
93
94
95
97
3.6. Summary
98
IV METHODOLOGY
AND RESEARCH
DESIGN
101
4.1. Introduction
101
101
105
108
115
115
119
123
121
127
135
140
4.61. Method
140
iii
141
143
145
146
151
154
4.8. Summary
V ANALYSIS
AND FINDINGS
155
5.1. Introduction
155
5.2. Sampling
155
155
158
162
166
167
176
5.4.3. Assessmentofmodelfit
189
5.5. Summary
200
VI DISCUSSION
202
6.1. Introduction
202
203
211
216
222
iv
223
6.7. Summary
225
VII CONCLUSION
227
7.1. Introduction
227
227
229
233
237
238
244
248
7.6. Summary
252
Ap
Pp
endix I- ConsultancyInterviewsQuestionSheet
254
Appendix2- CompanyInterviewsQuestionSheet
256
Appendix3- FocusGroupDiscussionsQuestionSheet
259
261
262
263
264
270
276
References
277
TABLE OF FIGURES
17
75
100
106
191
vi
LIST OF TABLES
Table 2.1. The characteristicsof main perspectivesin corporate image and
key
identity
and
some
studies
authors.
corporate
34
99
Table 4.1. Details of in-depth interviews with consultants and senior managers.
110
Table 4.2. The details of focus groups and the core points discussed.
114
116
121
126
Table 4.6. Reliability and factor analysesresultsfor eachconstructon the basisof 129
the Pilot Study 1.
Table4.7. Reliability measuresfor the constructson the basisof Pilot Study2.134
Table 5.1. Samplesizeestimationon the basisof the McDonald's andthe Renault- 157
Mais datasets.
Table 5.2. Demographics of the McDonald's and the Renault-Mais cases 159
figures.
to
the
population
main
compared
Table 5.3. The variableswith possiblemissing data patternsin the McDonald's 164
dataon the basisof West.
Table 5.4. The variableswith possiblemissing datapatternsin the Renault-Mais 165
dataon the basisof Nest.
Table 5.5. Exploratory factor analysis results" for the corporate identity mix
(company-controlled communication) elements.
170
Table 5.6. Exploratory factor analysis and Cronbach alpha results for the 173
(uncontrolled)
communication elements.
unplanned
Table 5.7. Exploratory factor analysis results for the dependent, antecedent and 175
the control variables.
183
vii
190
Table 5.13. The results in terms of the rejected and supported hypotheses.
195
viii
DEDICATION
is
for
secondto none.
support
me
and
ix
ACKNOWLEDGEMENT
I would like to thank many peoplefor their personaland practical supportduring my
PhD quest at Warwick. First, I would like to expressmy deepestgratitude to my
principal supervisor,ProfessorTC Melewar(now at Brunel University),for believingin
my project and being a great exampleof an academicand a strong person in life.
Second,I would like to sincerelythank my secondsupervisor,MaureenMeadows(now
been
Open
has
University),
very constructiveandpatientwith my writing.
at
who
I would like to thank my professors Prof Selime Sezgin and Prof Zeynep Bilgin, who are
the reason of my career in academia,and to my head in ITU, Prof Nimet Uray, who had
to deal with all the bureaucracyto help me complete my PhD adventure in abroad. I also
would like to mention my colleagues in Middlesex University Business School, who
have given me the chanceto gain more experiencein teaching and research.
I am also very grateful to my PhD colleagues Cristiana.Lages, Leong Yow and Cagri
Yalkin here who constantly helped me out of difficult situations during my PhD
experience. I thank my officernate and secondyear housernate,Nina Seppala,for being
a very encouraging and lively friend. I also thank my colleague, Dr Noppom
Srivoravilai, for his regular support during the data analysis phase of my thesis. I am
Giannakis,
housernates,
Dr
Milialis
Dr
Olga
Mourouti,
thankful
to
and
my
also always
Ioannis Bardis, who made my London days a beautiful dream.
DECLARATION
This is to declarethat:
"
"
All verbatim extracts have been distinguished and the sources specifically
"I
acknowledged.
During the preparation of this thesis, some papers were prepared as listed below.
The remaining parts of the thesis have not yet been published.
and Melewar,
T.
C.
(2004) "Linking
Corporate
xi
Signo
Date:
xii
ABSTRACT
This thesis aims to extend the current knowledge about corporate image formation
process by developing a comprehensive model which incorporates corporate identity
mix
communication
factors
(i. e.
interpersonal,
intermediary
and
intrapersonal
discussions.
The main surveydatais derivedfrom a cross-sectionalsurveywhich
group
is conducted in Turkey. The data drawn from 439 questionnairesis analysedby
multivariate data analysis techniques including exploratory factor analysis (EFA),
Cronbach alpha, confirmatory factor analysis (CFA) and structural -equation modelling
xiii
The findings of this studyshowthat not every aspectof corporateidentity mix elements
(e.g. corporateaesthetics,staff apparel)is consideredto be salient communicatorsof
corporate identity by consumers.They also indicate that when consumersattribute
positive feelingsto a company'scorporatevisual identity systems,when they think that
marketingcommunicationactivities reflect corporatevalues,and when they perceivea
company as socially responsible, they tend to form favourable image of that
organisation.Moreover,the resultshighlight that interpersonalcommunication(wordintrapersonal
friends
information
and
and
relatives)
of-mouth
exchangeamong close
factors (i. e. corporate associations gap, consumer-company identification, emotional
image
determinants
of
corporate
appeal, consumer-company value congruence) are
formation.
However, it
image;
immediate
their
corporate
associations about
communication may not activate
boundary
be
dependent
image
conditions.
may
on other
relationships with corporate
Finally, it is shown that the perceived integration of corporate identity mix elements is
linked to corporate image formation through the corporate associationsgap. Overall, this
research shows that corporate image formation process is a complex phenomenon.
xiv
I INTRODUCTION
and
external
impact
factors
in
in
to
their
on consumers'
order
examine
one
model
communication
(Comelissen,
been
identities
has
tested
empirically
not
evaluationsof companies'
2000).
In the following sections, first why the above subject has significance in the
field
(Section
1.1),
and second,what this researchaims to examineare
marketing
1.2).
Third,
(Section
the context of the study and the targetedpopulation
explained
for the datacollection are described(Section 1.3).Fourth, how the investigationwas
data
and
which
collection and analysismethodswere usedare presented
conducted,
(Section 1.4). Fifth, the contribution of the research is discussed(Section 1.5).
Finally, the structure of the chaptersand the definitions of the key conceptsare
provided(Section 1.6).
informs
for
the media, signals a company'snew
advertising,
provides an objective
direction or name change, createsan understandingof and familiarity with the
companybehind a product or service(Bernstein,1984;Dowling, 1986),encourages
favourable behaviour towards the company and adds value to general promotion
(Kennedy,1977).
Keller (2000) claims that "the power of a brand lies in the minds of consumersor
be
identity
(p.
157).
Similarly,
the
can
assessed
strengthof a company's
customers"
by examining how the intended identity results in the minds of its stakeholders.
Organisationswant to ascertainthat their core valuesand principles are understood
in the sameway they want to position them in the marketplace.Gray and Balmer
(1998) statethat a strategicallyplannedand co-ordinatedcommunicationprogramme
rather than autonomouscommunicationactivities can build a strong imageand help
organisationsachieve a match betweentheir communicatedidentity qualities and
peoples' associationsaboutthem (Abratt, 1989;Gray and Smeltzer,1987;Melewar,
2003; Van Riel, 1995). For example, in 1983, TRW launched a television
advertisementwhich had the slogan "The Future Isn't What It Used to Be"
(Hartigan, 1987,p. 67). This messagewas in line with the company'score value of
being future focusedin that it supportedtheir promiseof being the producerof stateof-the-art products,while their printed advertisementsimultaneouslystressedthat it
was a companywhich generatesnew ideas (Hartigan, 1987). It is evident in this
example that companies are keen on conveying consistent messagesabout
themselvesand henceachievingan overlap betweenwhat they claim and how they
are perceived.
Brown and Dacin (1997) have found that there is a close link betweenconsumers'
associationsof a company's manufacturingabilities and social responsibility and
how peopleevaluatethat companyand its products.In his 1996study, Drurnwright
concluded that "organisational-levcl associationsenable the benefits of strongerorganisationalidentification to be widespreadand pervasive" (p. 84). Similarly,
Bhattacharyaand Sen (2003) claim that the more consumersidentify themselves
inclined
loyal
be
the
the
to
they
that
more
and
with a company, more
company
will
they will be to promote it. They will also be more resilient to negativeinformation
aboutthat company.
3. Does the perceived integration of corporate identity mix elements(companycontrolled communicationelements)have an impact on achieving the match
between the projected corporate identity traits and consumers' perceptions of
thosevalues(corporateassociationsgap)?
identity
borders
The
of corporate
mix
9
elements (company-controlled
The model to examine the relationships between the corporate identity mix
elements (company-controlled communication elements), the unplanned
(uncontrolled) communication elements, the perceived integration and the
corporateimageconstructswas testedon the basisof the empirical data.
I-
I- McDonald's and Renault-Mais were' among the first ten most reputable
in
by
2000,
based
The
Capital
the
survey
completed
companies
on
reputation
10
The scales were purifled on the basis of the qualitative and the quantitative
assessment
of the questionnaire.Academicjudges were recruitedfor the qualitative
Zeller,
in
(Carmines
to
the
the
and
satisfy
stage order
content validity of
measures
1979;De Vellis, 1991;De Vaus, 1996).Two pilot testsprecededthe main survey in
factor
Exploratory
indicators
the
to
the
quantitatively.
order simplify
constructs
of
analysis (EFA) and reliability check using the Cronbach alpha statistic were
II
employed on the basis of the data collected from the pilot studies (Aaker, 1997;
Babin et al., 2000; Hair et al., 1998;Tabachnickand Fidell, 2000).
Before merging the two data sets (McDonald's and Renault-Mais)for exploratory
factor analysis and measurementmodel and structural model testing, they were
separatelyexposedto initial dataanalysisin order to diagnoseoutliers, missing data
and variables with non-normal distributions (Hair et al., 1998; Tabachnick and
Fidell, 2000).This was followed by the applicationof exploratoryfactor analysisand
the Cronbach alpha test (Aaker, 1997; Babin et al., 2000; Hair et al., 1998;
Tabachnick and Fidell, 2000). Structural Equation Modelling was used for model
testing since it is considered to be a powerful technique for testing causal
relationshipsbetweenconstructs(Diamantopoulos,1994;Fornell and Larcker, 1981;
Steenkampand Baumgartner,2000). A two-step approachwas taken for the model
testing (Andersonand Gerbing, 1988).First the unidimensionalityof the constructs
was validated on the basis of the measurementmodels by the application of
confirmatory factor analysis(CFA). Following that step,the modelswere testedby
applying structural equation modelling which simultaneously estimates the
relationships among constructs (Anderson and Gerbing, 1988; Diamantopoulos,
1994;Steenkampand Baumgartner,2000).
12
1.5. Contribution
of the Study
Balmer, 1997) in the consumer context. Additionally it considers the role of non-
13
Ibis research also brings new insights on the current conceptualisation and
operationalisation of the constructs such as the corporate image, corporate
associations,corporateidentity mix elements,unplannedcommunicationfactorsand
integrated communication. Although the majority of the sub-elementsof these
been
have
discussed,
has
been
their
not
scales
measurement
constructs
conceptually
developedand verified in the consumers'context. For example,the salienceof the
corporate identity mix elementsfrom the perspectivesof consumersis examined.
Furthermore, some constructs from other fields, such as identification and
integration,are adaptedto the context of this study. The robustnessof all the scales
developed and adapted is confirmed by the application of confirmatory factor
analysis(CFA).
14
15
in
by
It
the
the
methodology
study. continues presenting
adoptedand
context which
it was studied.
Chapter III presentsthe theoretical framework which links the above concepts
together.It depictsthe indicators for the constructsand sets out the hypothesesfor
the empirical testing.
16
17
Chapter VI
VI. DISCUSSION
6.1. Introduction
6.2. Measurement Scale Purification
6.3. Determinants of Corporate Image: Effects of corporate identity mix elements
6.4. Determinants of Corporate Image: Effects of unplanned communication elements
6.5. Integration of Corporate Identity Mix Elements and Its Relation to Corporate Image
6.6. Corporate Identity Mix Elements versus Unplanned Communication Factors
6.6. Summary
Chapter VII
VII. CONCLUSION
7.1. Introduction
7.2. Theoretical Implications of the Study
7.3. Managerial Implications of the Study
7.4. Limitations of the Study
7.5. Future ResearchAvenues
7.6. Summary
18
Corporate
identity
elements
mix
(company-controlled
communication
appearance), communication
Marketing
communication
19
Manager behaviour: The way the managers of a company represent their company
EmpIoyee behaviour: The way the employeesof a company treat its customers
(Kennedy,1977;Kiriakidou andMillward, 2000).
Intermediary communication: The communication between a person and nonperson communicator(massmedia, NGOs, governmentalinstitutions) regarding a
brand,a product,an organisation,or a service.
20
21
11LITERATURE
REVIEW
2.1. Introduction
workforce (Melewar et al., 2005). Simoeset al. (2005) assertthat it is important for
decision-makersto understandwhich factors affect the image that individuals form
about their organisations,in order to gain competitive advantage.Although the
literature in corporateidentity and corporateimage areashas discussedthis issue,a
comprehensive view has not been provided which delineates the types of
image
formation
influence
have
corporate
may
an
on
communication which
(Comelissen,2000).
22
Since early writers in fields of the corporate image and corporate identity were
practitioners,the main emphasisof corporate identity studies conducteduntil the
1980's was on the visual/graphic design featuresof organisations(Balmer, 1995,
1998; Van Riel and Balmer, 1997; Simoeset al., 2005). The visual/graphicdesign
schoolauthors(e.g. Carter, 1975,1976;Henrion and Parkin, 1966;Margulies, 1977;
Olins, 1978, Pilditch, 1970; Selame and Selarne, 1975) initially argued that
symbolism used by an organisationsuch as nomenclatures,logos, symbols, house
styles, stationary,uniforms, vehicles etc. brings visibility to a companyand should
be kept fashionable(Balmer,2001a; Van Riel and Balmer, 1997).
23
After Kennedy's study (1977), this underlying assumptionhas been revised, since
she demonstratedthat employeesof an organisation play a role in building the
identity of an organisationand in its communicationto externalstakeholders.It was
acknowledgedthat corporate identity influences the shared values, beliefs and
behaviours of organisational members on which the corporate culture is built
(Balmer, 1995; Downey, 1986), and therefore corporatesymbolism should reflect
characteristicsof organisationalculture aswell as any changesto it dueto alterations
in corporatestrategyto better position the company in the market place (Balmer,
1995; Van Riel and Balmer, 1997).Accordingly, the focus of researchin this area
has shifted to the examinationof how visual expressionsof an organisationwere
designedto reflect its core values and principles (Balmer, 1995; Van Riel and
Balmer, 1997).The major conceptualdevelopmentof the visual/graphicschoolwas
introducedby Olins (1978),who proposedthat organisationsexpresstheir corporate
culture and corporatestrategymainly by using three visual identity styles, namely
monolithic, endorsedand brandedidentities.
24
The pioneeringmodel of this era was developedby Abratt (1989) who launchedthe
concepts of 'interface' between corporate identity and corporate image which
defined the 'consistency' betweenthe projected identity and the perceivedimage.
His model was the first attemptto show a linear link betweencorporatepersonality,
corporateidentity, and corporateimagewhich setsout the foundation for corporate
image management.In the model, the importanceof a communicationprocesswas
emphasised.He argued that it is imperative to understandwhether the corporate
identity of an organisationis communicatedeffectively in order to make sure that
it
intended
Gray
2003).
(Bick
and
as
aL,
stakeholdersof an organisationperceive
et
Smeltzer(1987) statedthat a strategicallydesignedand co-ordinatedcommunication
programmecan build a strong imageand reinforce its consonancewith identity. Van
Riel (1995) arguedthat management,organisationaland marketing communication
in
integrated
be
terms
of
constitute
communication
should
efforts - which
corporate
executionand organisationalmessagesthey convey in order to eliminate the risk of
delivering contradictingcuesaboutan organisation'sidentity.
25
Cristensen and Askegaard (2001) argue that the employees' reactions to company
interpretations
interactions
their
their
of
with external publics gives
policies and
decision-makers
development
in
terms of establishing and
to
the
of
strategy
shape
communicating
corporate
identity.
Therefore,
the
relationship
between
organisational members and their companies in terms of how they internalise their
interest
is
distinctive
to
of
and enduring characteristics
organisations' central,
corporate identity management (Albert and Whetten, 1985, Balmer, 2001a).
26
27
Jerseythat the ability of senior managementto read the gap between its internal
reality and external image defines how the cultural atmosphere inside an
organisationcan turn into an undesirableenvironment.Hatch and Schultz(1997) and
Simoes et al. (2005) state that the senior managementof an organisation is
responsible for creating an organisationalclimate which nurtures the consensus
betweenemployeesabout their organisation'score values and main purpose,which
in turn fosters a more favourable perceived organisational identity and greater
organisationalidentification. This has beenconsideredby marketingacademics(e.g.
Balmer, 2001a; De Chernatonyand Harris, 2000; Harris and De Chernatony,2001;
Keller, 1999;Kennedy, 1977; Schultzand De Cherriatony,2002) as one of the vital
image
in
about an organisationin the minds of its
conditions creating a positive
externalaudiences.
28
Hatch and Schultz (2003) asserted that complex markets and sophisticated
consumershaveforced companiesto position their product brandsby differentiating
their organisations.Acknowledging that, marketing academicsstarted considering
the corporationas a brand in its entirety (Balmer, 2001a; Balmer and Gray, 2003;
Bickerton, 2000; Knox and Bickerton, 2003; McDonald et al., 2001, Simoeset al.,
2005).They arguedthat the organisationhasbecomea strategicelementin branding,
sincecorporatebrandingprovidesan opportunity to use a company'score valuesas
a part of its strategicselling points (Hatch and Schultz, 2000,2003). Keller (2003)
stated that emphasisingthe company behind its brands brings brand leveraging
opportunities by extending consumers'knowledge of about a company's product
brands.
29
30
members.
2.2.5. Interdisciplinary
paradigm
In 1995, the International Corporate Identity Group (ICIG) stated that corporate
identity and its managementis a multifaceted phenomenon(Balmer, 1995,1998).
They claim that corporateidentity managementrelatesto a company's values and
its
personality, disseminationof organisationalvalues
principles which constitute
internally, employees'sensemaking abouttheir organisation'sidentity, externalising
corporateidentity by formal communicationactivities, and the influence of tertiary
communication factors on perception of corporate identity (Balmer, 2001a).
Accordingly they formulatedthe following statement:
Birkigt and Stadler(1986) and Van Riel (1995) proposedthat corporateidentity cues
are revealed internally and externally by strategically planned symbolic,
communicativeand behaviouralactivities of an organisation.Similarly, The ICIG
statementasserts that understandingof corporate identity and corporate image
concepts requires harmonising the perspectives of visual/graphic design, of
communication and integration, and of organisational studies and marketing.
Moreover, it states that corporate branding concerns multi-stakeholders and
managementof it requiresboth internalisingand extemalisingof corporatevaluesby
formal communicationefforts and taking necessaryactions towards the possible
intervening influence of informal communication factors on their planned
communicationactivities (Balmer, 1998;Balmer and Gray, 2003,Hatch and Schultz,
2003). By this statementit has beenconfirmed that corporateidentity management
has an interdisciplinary nature (Van Riel and Balmer, 1997) and therefore while
developing conceptual frameworks for studying corporate identity and corporate
image related issuesdifferent conceptsfrom other disciplines should be integrated.
Simoes et al. (2005) also note that researchersshould ground their analysis and
discussionin a broadrangeof disciplines.
Recent conceptualand empirical articles (e.g. Balmer, 2001a; Bick et al., 2003;
Comelissen, 2000; Christensenand Askegaard, 2001; Dacin and Brown, 2002;
Melewar et al., 2003; Melewar and Karaosmanoglu,2005; Simoes,2001; Shoes et
32
al., 2005; Van Riel and Balmer,. 1997) highlight the fact that corporate identity
management and corporate image positioning stbdies should follow
33
Itt
en
cm
9ji
15
f..
u
2
4
4
g1,0 1 ""Li
0.
>
22
in
>0
0 .M
i2 12 Q r.
>, .2 jo.
b.
I-
U
-cj
Ce r.
=
0
4.
9-
M2
Q..
r c;r
8m &
:Li
ce
> "0
2 -Z%
92.0
:2-9
=
vi
-; g
=
2
z g
0
rA
9= :2
. .2 .
2
CJ
>
5Q
"0
Q
ci
cl
l=
to
4,
e; g.2gur,
>u
.t
to
7D
cu
9
.-Z5=
.-
e.
Q
A
-9-,
cu
ADi
v2
- -B
e
0
4 ,A
0
:s
e
-31.1
'c2,cu
Q
02
to
Ici
4.
-0
V)
m. rl ;
xu
9)
-5 , -,A
0 '-r- w
Gn -
& g r)
tu
Ei
92.0
'
Co
Ln
12
r_
Z,
vi
1.
p .5bd
0
e
(Y,
Ici
16.9
ei
.
.t-
Ilw 2'.=
gl
in
1..
12
'
c3
u .'
-,> 2
Vi
0%
0
m
Zo
ce
_b4
(:
"Ci
Q
1
1-<
00
=
'00
2.
*,
-0
0
l=
cl
4) C%
8
cz cez,
g>s,
r-
00
ti
-,
tQ CYN10
0-
CD (7,
3 -0
;B -9
<
00 U
ZA
-. 2
:g:
;:
.-U
9 Z-.
Z (7,2
Z2
<
to U2
41
ch
-. '
iz
9
-W> 2
0
(>
0me
>
Cw 00 -0
c3
2.
I, E
C
.-I
s',
'
9>
Ei
t
g!
t;
,,
.t
t
9
u P.00
t),
4
E
(7,
-1
Ei
CY% 0 ,
c>
-t, - g2
l! 5
0
-0=*CI
-EIr-i,
r-12,
00 (7,
10
0999
9Z
4 .
r- V)
ce
* LD
to
40. -
j-,
CZ.
Co
.28-.
42,
.re
cl
2
-u.!m
2e0.
MU
Ei
,5
tu
to
%.5
ro ,
00u
u2
a0
a 1?
r_
IU UM r.
ci
0
b
U2
-
6
., 2
Ln
m
=0
J'- t20 0
0E
. in
0 Azj
k.
8 c,
&.5 -2
Ei.2 Z
-to
-0
,5
.-0e
tu Z
Ei -mUX.
e90
9e.-e 1-0-
'CJ
te
rd
>
1,
E'
r-
Co
9-
-!u
r-
m>2
ci
2
4)
-9
.9u0n
41
r. mu>2-g
r- = ri
.2
e Z)
.8,2
E
ri
K
1-i'
Irl
Ch
ilU
-A
,r-
LD
i...
r. r,
gp
E
M.,
-2
ce
00
5gv
-0
. , 2
zi
0- -O>
cl
21
0S
t
E.
u
2V8 rceb= 7;
ae ..
b
C
E
U0e
.-. - 0
0
',Zi
r_
to
t!
0%
Ici
g -;2uti
e92 Icic
<
CD
9-0
r-
r-
00 52e0
.2 fA
r0cU.
I'm
2h
-d g.
u
>.
. r:
Ei
(*
E,
n r:
0) e
0eA
-0-5
:25
-. --Z:e
m
'Or-M
0 ,
i= Ei u=&.
2 (.,' r-
00
2
-A
!2uI
= -,
j
,;
u"
g0,
.-2 12
,
fi
1.
t.
29
:2
-r, 0
cm
= cu J=
"0
0
0
A
-.
2; >
:g.
-'
22
Z
, n.
-20,
r= > -,=U =
9
r-
cm
"0 N.
E
0 >,
-:S
sM
.0r.
bo
4.)
CD
'
E
ce
e
g"R
4)
Co .-
93
. 0 4) 0
eL
E
ci
2 .--0
r-
92. tu
>,
C:-
-::
,
0-
Z$ 0
ce CO
"0
W)
en
g.
t"
"0
U0
2*
CU
>Z
0
Ew
0
ci
r-
"0
9.1
0 C>
Qm=
tn ,0
- 10
*0
U02
.2 >
00
u
00Ajn.
; t
55 0
i
.
8u
2
r.
,r q5
(> (DN 2
ON
im - -ci
E
22=
(7,
>,
-c
;Eu,
4.
-j
L)
>0
:1.
GM EE
Z'
-v
ci
m r- 0
.. 0,
34
-0
t)
v
e =00
S
c\
0 :2e
Gn
2: 3
0
im
m
%. (D
uE
CD 0. NX
5j 't-
`>
"0
0
c:
.0 -i
-
ce
0-
5
U
00
E, 0
ZEZ
't
N 121
Z
cl,
:1:.
-to
tE
:1- --
r:
e9u
li
u2
ul
r- '0
g9 uu
u
1-19
5
0Z
c>
ja
Co
cn -0
0
', 0
41
,<-,
Co
u
.0
E2
:<
10
-0
ce r- =
'(DN
(:
Zm
cu'CJ o'
-- tn
im
-j
u0
CY% >
c>
CY%Q.
P,
=
-2
-% c
r-
0,
c zr,
(>
8
z
, , 2 C). -Z.
-a -4)
21. Z,
=
c CD
CD
O, 0 -j
- ch
c
t.
'
(N 0 CD (D l';j
00
4)
(>
rq
0U
0
>, Z
t4
-Z
10 CD
ce
:5
r.
-r- 49V, iz
u
1-..
vi
:3b-, 2 e
E3
5-
12 1>
9%
2:.
-; u- 00
"0
ImNumuU
cl
-0
0%
cy,
-
12.pr-ju
.85
,Z
al
00
4)
-0 -u
b4 kn
-0u
4)
,
'm
C2N - u C, 3 Gn
...
L, tu ON
00Ez
1
,Z .ga
'98g
Ei
0-, 3 J, 42
c 5
2, ,u 1-
c>
ci 0
0 %0
00 im c
z-wl
t:u 00
c,
N
,Uu,.
cn
:ir.
V
r4
-0
-0
1.
tu
1-.-0 0A*v
4.
u'08
,
',
2A
-jc
u
rn
.5
'cj
A<
e
fi
a',
l Z;
m
le 'xE
E 'J
l-. g',-m
:s 0' -m'EE
E2
:1
121.
cm
ZO
ce
a.
lu
cd U
00
tu
0UU-52
r-
42
Co ce 0
:i
j2
.2
.-
u-J,
.
-d *Z
bo
",
m9
0Z
-5
- tz jz 4
Z
.-,
.=
ZJ
Ei tu M
0Q
;j52
th -= 2 &0
Ew
0
-5
c2rn
the organisationare formed. This study takes the secondview into considerationand
(i.
influence
factors
the
e.
which
may
consumers'
evaluations
of
companies
examines
image).
corporate
Brown
individuals
into
Dacin
come
contact.
and
associations)when companiesand
(2002) argue that regardless of what a company wants to establish as its identity, the
mental picture people hold about it defines what its actual identity is. Similarly,
36
interdisciplinary
individuals.
is
This
identity
the
to
with
view
consistent
company's
identity
2.2.5),
Section
(See
that
and corporate
which
suggests
corporate
paradigm
image concepts are central to an organisation's success and understanding the
disciplines
(Brown
from
drawing
insights
between
them
several
necessitates
relationship
into
2005).
Accordingly,
Simoes
this
2006;
takes
this
account
gap
et al.,
research
et al.,
(See
Figure
3.1),
from
develops
the
which
model
conceptual
receivers'
standpoint
a
and
integrates interpersonal, intermediary and intrapersonal factors into the existing models
in
it
the consumers' context.
tests
and
Even though discussions on the conceptualisation of corporate image has started in the
image,
identity,
identity,
1950's,
the
terins
organisational
corporate
organisational
early
by
different
interchangeably
image
have
been
authors while
used
corporate
often
and
their definitions have overlapped (Bick et al., 2003; Brown, 1998; Simoes et al., 2005).
38 -
Each paradigm presentedhere has defined the corporate image concept through its own
Since each discipline approachedthe image concept with its own parameters,it is
difficult to formulatea generaldefinition of the concept(Brown ct al., 2006).However,
comparing the definition of image in marketing and organisation studies contexts could
Marketing literature refers to corporate image through two different angles. One group
of academics refer to corporate image as the overall impression held by the several
39
segmentsof the public (Barich and Kotler, 1991; Berstein, 1984; Bevis, 1967; Dowling,
Organisational studies consider the image concept from the employees' perspective and
by
identity
They
seen
outsiders.
company's
name these three perceptions as
Hatch
(Dutton
Gioia
2000;
Dukerich,
1991;
identity
and
et
al.,
and
organisational
Schultz, 1997; Pratt and Foreman, 2000; Whetten and Mackey, 2002), construed
external image (Dutton et al., 1994; Gioia et al., 2000) and desired organisational image
Although the subjectsat the centre of the definitions differ in thesetwo approaches,
Brown (1998) and Stem et al. (2001) concludedthat image is basedon perceptions
is
individuals;
it
in
the
therefore
an individual-levelconcept.However,
stored
minds of
image
definition
has
limited
(2001)
Stem
to the total gestalt
the
et al.
of corporate
while
40
impression of individuals about an organisation on the basis of their beliefs, feelings and
(1998)
image
is
formed
Brown
that
argues
experiences,
corporate
also
as a result of
specific associationsthat individuals link to an organisation,and has proposed,an
overarching concept of corporate associations. He defined it as the totality of
be
of
associations;and accordinglycorporateassociationsshould studiedas antecedents
the corporate image and not as components (cf. Brown, 1998).
(See Dutton et al., 1994; Hatch and Schultz, 1997). Moreover, Peteraf and Shanley
(1997) argued that managers tend to position their companies as a part of a strategic
41
in
industry.
Therefore, it could be assumed that this
the
same
group of organisations
comparative positioning of organisations with reference to each other would have
implications for an individual's perception of a particular organisation in an industry.
Fornbrun and Shanley (1990) claimed that while individuals are mentally positioning a
company, they engage in more complex inference making by comparing companies to
their counterparts.
public's opinion about that company, and his/her comparative evaluation of that
company versus its counterpartsin the sameindustry (Williams and Moffit, 1997).
As mentioned before (See Section 2.2.4), some marketing authors (e.g. Dowling, 1986;
42
and corporate branding fields (e.g. Balmer, 2001a; Balmer and Gray, 2003; Davies and
43
in
individual's
describing
that
an
emerge
mind
while
an
characteristics'associations
it
(Brown,
identity
1998).
However,
should be noted that
attributes
organisation's
decision-makersof an organisationdeterminewhal associationsthey would like their
stakeholdersto hold about their companies(Balmer and Soenen,1999; Brown, 1998;
Brown and Dacin, 1997;Brown et al., 2006; Van Rekom, 1997;Van Riel, 1995).The
be
to
that
the
management of an organisation chooses
set of organisational attributions
Balmer, 1997). After the shift towards recognising the importance of organisational
image
formation
in
(SeeKennedy,1977),the meaningof corporate
members corporate
identity expanded beyond the visual expressions of an organisation incorporating
corporate personality (Balmer, 1998).
describe corporate identity. They argued that corporate identity cannot be a direct
internal
expression of
reality or the personality of an organisation, since "... companies
balance internal preoccupations of organisational identity with external imperatives... "
2 It should be noted that what Cornelissenand Harris (2001) refer to as human identity metaphoris
differentfrom the onereferredto in the definition of corporateassociations
(SeeSection2.3). The former
refers to the values commonly sharedby organisationalmemberswhich constitutean organisation's
personality;the latteris the meansto describecompanies'identitiesby a company'sstakeholders.
45
interaction
between
social
a company and its businessenvironment including other
organisations and its stakeholders.
This view is supportedby Cheney and Vibbert (1987), who assertedthat corporate
identity is a dialectical phenomenon. Along with strategic,planning by decision-makers
This dcfinition of corporate idcntity (See Section 2.5) implies that the managementof
intentionally
organisations
choose certain company characteristics to position their
46
Van Riel, 1995). Balmer (1997) states that all planned expressions of an organisation
Riel
(1995)
Van
convey its identity.
assertsthat company-orientedmessagesare
disseminatedby company-drivencommunication efforts manifested as symbolism,
communicationandbehaviour.
Accordingto Cornelissen(2000)company-controlledcommunicationtakesplaceamidst
informal
information
during
forms
The
sources.
one-to-one
other
of message
exchanged
I
from
(interpersonal
the
communication),
received
communication
news
mass media,
NGOs, governmental institutions etc. (intermediary communication) and psychological
consequencesof the previous experiencesand images stored in the mind (intrapersonal
define
level
the
also
attachment
communication)
of consumers to companies and the
47
The managementof a company's identity lies at the heart of the planned communication
activities of an organisation (Gray and Smeltzer, 1985; 1987). Gray and Balmer (1998)
disseminate
to
tools
that
several
communication
channels
use
and
state
companies
corporate messagesin order to position their organisations in their stakeholders' minds.
Van Riel (1995) arguesthat the values and principles which define a company's identity
its
in
be
management, organisational and marketing communication
should
embedded
be
forms
he
Thus,
that
can
all
of
company
communication
efforts
suggests
activities.
headings
identity
following
the
corporate
mix
which constitute
classified under
elements: symbolism, communication, and behaviour.
2.6.1.1. Symbolism
48
Olins (1989) stated that an organisation's symbolism conveys cues about its goals.
hence
customer loyalty. Melewar and Saunders(1999)
products and services,and
emphasisedthat standardisedvisual identification also plays a role in a company's
successin the global context.Rosson(2003) reportedthat especiallyduring mergerand
acquisitions visual identity change
helps to
leading
it
is
for
to.
organisationstands andprovidesguidanceaboutwhere
increases
likelihood
the
of achievinggreatermarket
on an organisation'spublications
visibility (Melewarand Saunders,2000; Schmittet al., 1995;Topalian,1984).
49
determines
how
people respond to a company's products and the
a spokesperson
persuasionlevel of organisational messages.
50
1999,2000; Topalian, 1984), 2) staff apparel (Dowling, 1994; Gray and Balmer, 1998;
Kennedy, 1977; Van Riel, 1995), and 3) corporate aestheticsincluding printed material
(e.g. stationery, promotional literature etc.) and exterior and interior desing of company
buildings (e.g. headquarters, plants, retail stores, offices etc.) (Schmitt et al., 1995;
Topalian, 1984).
2.6.1.2. Communication
Hatch and Schultz (2003) and Knox (2004) stated that in recent years, increased
competition,highly demandingconsumersand faster innovationshaveforced decisionmakersto changetheir marketingstrategiesfrom being productbrand orientedtowaids
corporatelevel branding. Keller (2003) argued that treating organisationsas brands
enablecompaniesto distinguishthemselvesfrom their equivalents,which in turn may
supporttheir productsandservices.
Rossiter and Percy (1996) assertedthat positioning a brand requires to define what the
major premises of a brand are and what it offers to its targeted audiences.Harris and De
Chematony (2001) noted that "a set of functionally distinct capabilities that differentiate
a brand should be derived from the brand's core values" (p.444). Aaker (1996) claimed
that a brand's major qualities represent its identity, and defined brand identity as "a
unique set of brand associations that the brand aspires to create and maintain. These
from
imply
brand
for
that
to
represent
associations
what
stands
and
a promise customers
the organisation members." (p. 68). Simoes and Dibb (2001) and Simoes et al. (2005)
51
definition
(1996)
Aaker's
that
argue
overlaps with the definition of corporate identity
(See Section 2.5), and implies that organisationalvalues should underline individual
brands' characteristics.Keller and Aaker (1992b)statethat embeddingcompanyvalues
in product brands may reduce the risk of negative response to brand extensions.
Moreover, Gflrhan-Canli and Batra (2004) note that products endorsedwith core
company characteristicsmay be evaluated more positively.
is
by
view supported Duncan and Moriarty (1998) who statedthat "brand messages
originate at the corporate, marketing, and marketing communication level." (p. 6).
influence
individuals'
efforts
not
only
responsesto product
marketingcommunication
brands,but also their perceptionsand evaluationsof corporatebrands.Drawing on this
argument,Van Riel (1995) claimed that marketingcommunicationmix elements(e.g.
advertising, sponsorship, public relations activities, corporate advertising, sales
promotionsetc.) should be usedto convey the distinctive qualities of an organisation
(i. e. corporate identity).
52
2.6.1.3. Behaviour
In line with Birkigt and Stadler (1986), Van Riel (1995) states that an organisation
should take its behavioural aspect into account while communicating its identity to its
issues.
andenviromnental
behaviour
integrate
to
started
employee
as a part of organisational-lcvelvalue
dissemination. Balmer and Wilson (1998) assert that "corporate identity refers to an
organisation's unique characteristics which are rooted in the behaviour of employees"
brand
both
be
the
to
at the productand corporate
of
messages,
ambassadors
considered
level (Balmer and Wilkinson, 1991; Harris and De Chernatony, 2001). Tberefore, it can
53
be concluded that employee behaviour creates a basis for corporate image formation
While several authors (e.g. De Chematony and Harris, 2000; Harris and De Chernatony,
2001; Hatch and Schultz, 1997; Keller, 1999; Kennedy, 1977; Schultz and De
Chernatony, 2002) mostly highlighted the role of managers in developing a shared
vision of their company within their workforce, Van Riel (1995) mentioned that
activities, and appearin the media and in public occasionsas spokespersonof their
organisations. Scott and Lane (2000) and Simoes (200 1) state.that top managersare the
first group of organisational members who are formallY charged to represent their
1998;
be
identified
in
(Argenti,
to
the
companies
want
valueswhich
with
marketplace
Gray and Balmer, 1998; Hunt and Grunig, 1994).
54
Economic Forum, 1999) argue that since in recent years consumershave become more
life,
issues
they
to
the
the
quality
natural
of
envirorunent
and
related
consciousabout
scrutinise company's actions more to assesswhether they
-contribute
to the society in
(e.
business.
Previous
doing
they
researchon corporate social responsibility g.
which
are
Bhattacharya(2001)demonstrated
that consumers'associationsaboutcompanies'social
individuals'
favourability
level
determine
the
of
responsibility actions
company
evaluations.
in
how
the media and
their
values
organisations'
managers
represent
well
consumers,
issues.
how
and
social
company
supports
environmental
mucha
public occasions,and
The definition of corporate identity (See Section 2.5) suggeststhat managing corporate
there is a one-way linear link between corporate identity and corporate image
(Comelissen, 2000). Some authors (e.g. Cheney and Vibbert, 1987; Dacin and Brown,
2002) point out that companies'. identities are redefined according to stakeholders'
responseswhich are basedon their interactionswith each other and other sourcesof
information as well as company-drivencommunicationactivities. Proctor and Kitchen
(2002) support this approach by stating that in the post-modem era, consumers are not
just the passive receivers of company communication, rather, they shape what
organisations should be.
56
previous experiences and images stored in the mind). The following sections explain
thesefactorsin detail.
Researcherssuch as Bristor (1990), Buttle (1998) and Duhan et al. (1997) comment that
information
flow among consumersnot only affects individuals'
word-of-mouth
approachestowardsproductsand services,but it may also influencetheir views about
companies. Williams and Moffitt
I
57
and services. Accordingly, Comelissen (2000) and Dacin and Brown (2002) posited that
mouth information not only occurs betweenindividuals, but also is spreadby some
communicationmediums.CristiansenandTax (2001)highlight that recenttechnological
advancessuch as electronic bulletin boards, e-mail and the Internet createda new
information
by
about
environment which consumerscan easily accessa vast amountof
businesses
andtheir actions.
In line with Kotler (1988), Van Riel (1995) comment that communication fhanagers'
58
(Comelissen, 2000).
59
following
intrapersonal.
communic ion.
a. Corporate associations gap. Brown (1998) and Brown et al. (2006) stated that
60
expected. On the basis of this discussion, the corporate associations gap concept is
Many authors in the fields of corporate identity and corporate image mention that
favourable
to
makers should aim
engender
attitude towards their companiesby
I
61
Ashfort and Mael (1989) stated that these definitions above encompass one's
with authors such as Brewer (1991), Kramer (1991), and Tajfel and Turner (1985),
Ashfort and Mael (1989) argued that individuals associatethemselveswith certain
groups to define who they are and what they belong to, and their categorisation of
failure of the group.Accordingly, Mael and Ashfort (1992) define identificationas "the
perceived onenesswith an organisation and the experience of an organisation's
successesand failures as one's owrf ' (Mael and Ashfort, 1992, p. 103).
organisational
identification (e.g. Gwinner and Swanson,2003; Pratt, 1998; Scott and Lane, 2000)
haveshownthat identificationcan alsoplay a role in relationshipbuilding betweennonformal members and organisations.Aaker (1996) assertedthat consumersalso engagein
deeper cognitive attachment with companies. Similarly, Bhattacharya and Sen (2003)
claimed that consumersmay categorisethemselvesin social groups which they associate .
62
with the companyand its productsand services,but also to what they construeabout
a
that company'sreputationamongthe public (Bhattacharya,
and Sen,2003).
63
Organisational identification scholars such as Ashfort and Mael (1989), Bergami and
Bagozzi (2000) and Dutton et al. (1994) suggestthat individuals are inclined to be
attractedto companieswhich are evaluatedas beingtrustworthyandrespectfulby others
that they think of highly. Otheracademics(e.g. Hogg andTerry, 2000;Pratt,2000; Scott
and Lane, 2000) imply that when those attributions have a salient position in an
individual's working memory;it is likely that this individual will focusandelaborateon
those companieswhich have these qualities. Therefore,it could be inferred that an
organisation'semotionalappealcan be defined as the emotionalpredispositionof an
individual towards a company on the basis of its reputation of trustworthiness and
respectfulnessin the public mind.
Fombrun et al. (2000) demonstrated that the emotional appeal of organisations with
is
to
these
two
one of the essential components of reputation
regard
attributes
is
Since
management.
corporatereputation consideredto be an aggregateform of
corporateimagesheld about a companyover time (Gotsi and Wilson, 2001), it can be
formation
is
determinant
image
that
assumed
of corporate
emotional appeal a
process.
to'usethemassocialidentifiers.
64
65
Previous research on the self-image concept has shown that self-congruence theory has
behavioural and attitudinal consequences.For example, Sirgy and Samli (1985) reported
that there is a relationship between people's perception of their self-image and their
evaluations of product image as well as of store image. Hong and Zinkhan (1995)
showed that self-concept determines brand preference, brand attitudes and purchase
intentions. Sen and Bhattacharya (2001) argued that self-image concept can be studied
at the company-level, and demonstrated that when an organisation's values match
individuals' own self-image, they are more likely to evaluate that company positively.
All in all, in line with authors such as Bhattacharya and Sen (2003), Comelissen (2000)
views about companies.Proctor and Kitchen (2002) commentthat in the future the
successof companieswill be mostly dependenton positioning organisationsin their
to understandwhich
stakeholders'minds.Therefore,it is importantfor decision-makers
communicationtools and channelsare more influential in its constituents'decisionmaking processes (Abratt, 1989). Brown and Dacin (1997) state that even though
organisations are putting substantial effort into managing their identities, they still do
not know whether it is tlfe planned communication or external responseto their efforts
66
As discussedin the integrated communications paradigm section (See Section 2.2.2), the
realisation of the importance of consistency between the presentation of corporate
identity and its perception led researchersto pay attention to the concept of integrated
communications (Bernstein, 1984; Gray and Smeltzer, 1985; Van Riel and Balmer,
inefficiency in efforts and materials used (Beard, 1997) by strategic planning and
execution (Duncan and Everett, 1993).
67
Since the launch of the term 'integration' to marketing theory, various definitions have
in integratedmarketingcommunicationsfield. Duncan
beenusedmostly by researchers
and Moriarty (1993) defined integrated marketing communications as "the strategic coordination of all messagesand media used by an organisation to influence its perceived
"IMC
that
relations)
All thesedefinitions emphasisethree main points which are also supportedby Low's
(2000) study: 1) a common communicationmessage,2) co-ordinatedplanning and
68
eachcommunicationactivities.
Van Riel (1995) asserts that the concept of integration not only relates to marketing
but
also concerns organisational and management
communication activities,
communication activities which are embodied by symbolism, communication and
behaviour of a company. He claims that integration is also sought at the corporate level
69
proper selectionof channelsand tools to carry them and the better co-ordinationof
company-driven communication efforts are the foundation for motivating consumersto
form similar corporate associations to what is projected (Balmer, 1998; Gray and
Duncan and Moriarty (1998) claim that consumersare inclined to integrate brand
by
highlighting
Sandra
(1997)
Prof
Moriarty's
Harris
this
messages.
supports
view
words: "The brain does not distinguish an advertising message from a PR message.
70
As a conclusion,integrationin the contextof corporateidentity managemententailscoordinatedexecutionof corporateidentity mix elementsand alignmentof theseelements
to serve a common corporate identity messageby conveying similar messagesto each
other(SeeLow, 2000).
2.8. Summary
In light of the interdisciplinary paradigm which suggests that concepts from different
71
transmits company related information to individuals while they are forming an image
depicts
framework,
the
the
theoretical
The
which
of a company.
next chapter outlays
identity
between
mix elements, unplanned communication
corporate
relationships
factors, integrated communication and the corporate image concept.
72
3.1. Introduction
Brown, 2002). Although it is well studied and argued that a company's planned
communicationefforts influencethe imagepeoplehold aboutthat organisation(Balmer,
1998; Balmer and Soenen, 1999), there is a paucity of researchin the corporate identity
and corporate image literatures on understanding the impact of company-driven
communication efforts and external communication factors on corporate image
Based on the attribution theory, the conceptual model (Figure 3.1) proposes that
consumers9positive attributions about a company's symbolic expressions (corporate
visual identity systems, corporate aesthetics, staff apparel), marketing communication
mix elements (products/services, brand advertising, corporate advertising, public
relations activities, sponsorships,direct selling etc.) and behavioural aspects (manager,
increase
behaviour)
the likelihood of achieving a favourable
employee, and company
image about that company in the minds of its stakeholders.
75
Gray and Balmer (1998) state that a company's identity is translated into an image in the
76
antecedent of the corporate image. In other words, when consumers have positive
attitudes towards an organisation's formal communication efforts (i. e. symbolism,
communication, and behaviour), they will be more likely to interpret the messagessent
by those communicators positively and hence will be more likely to form a favourable
image about that company.
This assumption is grounded in the attribution theory (Graham, 1991; Kelley, 1967;
behaviour upon which they build their reactions towards that particular behaviour
(Kelley and Michela, 1980). It suggeststhat attributions are contingent on three factors:
1985,1986), the attribution theory has been used extensively in marketing studies
(Folkes, 1984; Valle and Wallendorf, 1977). There has been evidenceto show the
behaviour
between
attributions
about
employee
negative
and customer
relationships
complaints (e.g. Curren and Folkes, 1987; Richins, 1983), customer satisfaction (e.g.
77
Oliver and DeSarbo, 1988) and repurchase intentions (Folkes et al., 1987). However,
In the literature, it has been argued that consumers' evaluations of corporate image are
3 It should be noted that Weiner (1986) also arguesthat people's attributionswill be contingenton
situationswhich force other peoplebehavein a certainway. In the contextof this study, theseexternal
forcescanbe the businessenvironment,competition,industryidentity andso on (SeeBalmerandSoenen,
1999)which can force companiesto changetheir communicationstrategies.Sincethe studyonly, focuses
on corporateidentity mix elements,thosefactorsare not discussedhere.Furthermore,it is assumedthat
consumerswill not be able to easily judge whether a company'splanned communicationactivities
changeddue to such external forces. They would rather see the immediateresult of such forced
modificationsin communicationactivitiesof an organisation.
78
3.2.1. Symbolism
Although there has been a shift from a simplistic view of 'corporate identity as visual
expressionsof an organisation' towards a multi-faceted approach of 'corporate identity
as all expressions of a company' (Cornelissen and Harris, 2001), the importance of
symbolism in communicating corporate identity is still widely acknowledged (Baker
and Balmer, 1997; Van Riel and Balmer, 1997).
is
identity
identity
by
corporate
of a company projected
corporate visual
systems
(logo/symbol, name, slogan, colour and typography) (Dowling, 1994; Mel-ewar and
Saunders,1999,2000; Topalian, 1984) and their applications on staff apparel (Dowling,
79
reactions (Henderson and Cote, 1998). Van Riel (1995) notes that well designed and
favourably.
The same affect transfer can be assumedfor the other symbolic
more
aspects of an organisation including the other elements of corporate visual. identity
systems, corporate aesthetics and staff apparel. Thus it is assumed that consumers'
Hla: The more positive the attitude that consumershave towardsa company'svisual
identity systems,the more favourable the image they have about that company.
3.2.2. Communication
80
direct
promotion,
sponsorship
and
selling are also considered
services, advertising, sales
as part of a corporate identity mix (Barich and Kotler 1991; Van Riel 1995). Corporate
1994),
(Hunt
Grunig,
1998)
(Argenti,
and
public
relations
and
activities
advertising
individual
familiarity
directed
than
recognition
and
rather
which are
at company
products/services,
contributeto communicatinga company'sidentity aswell. Barich and
Kotler (1991) and Keller (2001) statethat marketingcommunicationactivities not only
aim to positiona company'sproductsand servicesin the market,but alsoto promotethe
it
is
itself.
Therefore,
concludedthat the favourability of consumers'views
company
by
how
be
their
well the marketing
attributions
about a companywill
enhanced
about
intended
in
identity
the
to
that
wants
create
a company
communicationactivities reflect
the mindsof consumers.Henceit is proposedthat;
3.2.3. Behaviour
The elements of this dimersion are employee behaviour, manager behaviour and
company behaviour (Van Riel, 1995). Ind (1997) claims "perceptions of an organisation
(p.
Since
indirectly,
by
83).
directly
determined,
employees
managers
and
staff'
or
are
1986;
face
1977;
Dowling,
(Kennedy,
the
of
an
organisation
and senior managers are
Hernsley, 1998), managers' interactions with external audiences at social events
81
(Argenti, 1998; Hunt and Grunig, 1994) and employees' contact with consumers
have
image
favourable
they
the
the
aboutthat company.
well,
more
H3c: The more consumers perceive a company as socially responsible, the more
favourable the image they have about that company.
82
closeenvironment),intermediary(word-of-mouthdisseminatedby massmedia,NGOs,
govermnental institutions) and intrapersonal (psychological consequencesof previous
in
images
experiences and
stored the mind). Accordingly, he argues that reception of
following unplannedcommunicationelementsareexamined.
consideredas an input into consumerdecision making (Halstead,2002; HarrisonWalker, 2001; Lau and Ng, 2001; Richins, 1983).In consumerresearch,it is widely
information
informal,
(i.
that
accepted
person-to-person
exchange e. WOM) about an
determinants
is
the
major
of consumerattitude and
organisationor a product one of
83
behaviour change (Brown and Reingen, 1987; Cristiansen and Tax, 2000). Several
researchers(e.g. Bristor, 1990; Duhan et al., 1997) consider that positive word-of-mouth
information received from friends and relatives is a significant communication factor in
is
it
favourable
In
this
about
'
engendering
view
a company or a product.
respect.
concluded that positive informal information received from interpersonal sources can
result in a positive attitude towards a company, creating a more favourable image of that
company (Comelissen, 2000; Dacin and Brown, 2002). Thus, it is hypothesisedthat:
H4: The more positive word-of-mouth consumersreceive from their close friends and
have
favourable
image
the
they
the
more
about that
relatives about a company,
company.
about social and environmentalissues has led to the search for more company
behaviourrelatedissues(Brown and Dacin, 1997;Senand Bhattacharya,2001).Cases
such as Enron (corporate malfeasance) and McDonald's (obesity issues) show that
H5: The more positive word-of-mouth consumers receive from intermediary sources,
85
appeal,andconsumer-company
valuecongruence.
image (consumer-company
value congruence).Furthermore,this researcharguesthat
emotional appeal is an antecedent of consumer-company identification and that
consumer-company value congruence is a determinant of
consumer-company
identification and the corporateassociationsgap. Finally, it claims that along with the
direct impacts of emotional appeal and consumer-companyvalue congruenceon
corporateimage, there is a mediated relationship between the former concept and
identificationas well as betweenthe latter
corporateimagethroughconsumer-company
image
concept and corporate
via both consumer-company identification and corporate
associationsgap. In the following, the propositions for each relationship are presented.
86
87
For example, Weiss et al. (1999) showed that the difference between an organisation's
own perceptionof its reputationand its salesforce's judgement has implications for
salesforce structure.Similarly, it is assumedthat the differencebetweenconsumers'
associationsabouta company'sintendedidentity and that organisation'sown definition
of it (i.e. the corporateassociationsgap) affects consumers'overall assessment
of a
company's image (Dacin and Brown, 2002). Therefore, it is proposedthat;
H6a: The smaller the corporate associationsgap, the more favourable the image
consumershaveabouta company.
Scott and Lane, 2000) showed that people tend to identify themselves with
organisationswith which they are not formally involved. Bhattacharyaand Sen(2003)
claim that when consumers see that a company is well respected by others whose
be
develop
be
likely
in
to
to
they
they
more
opinions
will
a similar view order
value,
88
able to be a part of that group of people. Drawing on this argument, it is proposed that
is
identification
consumer-company
one of a psychologicalstatewhich could affect
how consumersperceive a company as a whole. Therefore, it is posited that;
3.3.3.3.Emotional appeal
in
impression
their
to
about
organisatioris the market place as
order createa general
being good corporatecitizens (Fornbrunand Shanley, 1990). Bhattacharyaand Sen
(2003) claim that- consumersare emotionally attracted to companieswhich have
feel
identities
the
they
trusted
public,
worthy when they
among
since
respected and
89
construe that the organisations they consider as reputable are also well regarded by
individuals they hold in high esteem.Drawing on this conclusion, it is proposed that the
emotional predisposition of consumers towards a company which is triggered by the
collectively shared public view about that organisation will influence the image held
about that company. Hence,
H6c: The more consumers find a company's identity emotionally appealing, the more
The argument above by Bhattacharya and Sen (2003) also implies that the relationship
between emotional appeal and corporate image is likely to be contingent upon an
individual's feeling of self-worth. They claim this issue relates to the consumer-
90
Researchon person-organisation fit, and self-concept and value congruence has shown
"I" and the linkage betweenthis particular attribute and the self-imageof that person.
For example,when a product is consideredto be a luxury item by an individual, this
individual
himself/herself
"I
define
that
the
then
self-schema
and
as
may activate
may
am a luxury product user".
that when consumersperceive a close match between their own self-image and a
's
have
image.
O'Reilly
tend
they
to
al.
et
product's
positive associationsabouta store's
in
(1991)
a
similar
study
showed
result an organisationalcontext, that is, the match
between personal and organisational values stimulates positive emotions in employees
91
towards their organisations which may lead to higher staff motivation and to greater
Sen
organisationalcommitment.
and Bhattacharya(2001) demonstratedthat when
consumers perceive that an organisation's social responsibility (CSR) and company
ability (CA) values match with their own self-image, they are more likely to evaluate
that company positively. Accordingly, it is proposed that:
Me: The greater the consumer-company value congruence,.the more favourable the
imageconsumershaveabouta comPany.
92
Dutton et al. (1994) claim that "[T]he greater the attractivenessof the perceived identity
of an organisation, the stronger [is] a person's identification with it" (p. 244). This
proposition implies that a company should have an emotionally appealing point for
consumers before they engage in deeper relationship building with companies such as
using them social identifiers. As mentioned in Section 3.3.3.3, when consumers think
that peoplethey value also respectand trust companiesthat they find reputable,they
will be more likely to considertheseorganisationsin helping to define who they are in
their social environments(Bhattacharyaand Sen,2003). Therefore,it is assumedthat
the emotional appeal of an organisation is a determinant of consumer-company
identification,andhenceit is proposedthat;
93
inclined
be
dcfining
to define
they
may
more
similar
attributeswith organisations,
themselves in relation to these companies (i. e. consumer-company identification) in
order to fulfil the needof associatingthemselveswith a certain social group (Sen and
Bhattacharya, 2001).
that;
H6h: The greater the consumer-company value congruence, the greater the consumercompany identification.
94
Furthermore, Bhattacharya and Sen (2003) assert that consistent projection of intended
integration
in
identity
the
addresses
of corporate
mix elements
role of perceived
corporate image formation.
3.4. Perceived integration of corporate identity mix elements and corporate image
in
image
in
identities
favourable
to
the mindsof
order
achieve
companies'
consistently
a
their stakeholders(Abratt, 1989;Berstein,1984).Balmer and Soenen(1999) arguethat
different people may have different perceptionsof an organisation'sidentity, since
individuals' associationsabout companies are subject to their interpretationsof
basis
feelings
the
their
attributes
on
of
own
organisational
values,
and experiences
(Brown, 1998;Dacin andBrown, 2002).
Psychologicalresearchhasdemonstrated
that coherentimagesprovide greaterinfluence
in
(Moriarty,
1996),
"where
consistency
communication can be seen as a
on publics
necessary condition... " (Cornelissen and Lock, 2001, p. 428). This premise and the
95
build
(i.
behaviour)
a strongimageand
elements e. symbolism,communicationand
can
reinforceits consonancewith the communicatedidentity (e.g. Gray and Smeltzer,1987;
Van Rekom, 1997; Van Riel, 1995; Van Riel and Balmer, 1997). -i
96
identity values as consonant with what it intends to convey (i. e. smaller corporate
associations
gap),andasa resultthey hold a morefavourablecompanyimage.
H6j: The greater t4e perceived integration of corporate identity mix elements,the
smaller the corporate associationsgap.
97
'word-of-mouth
from
the
corporation',
or third party communications',
communications
individuals'
in understanding how the constituents of
feelings
'emotions
of
and
and
organisations build impressions of them and make a comparison between organisational
factors'
impact
image
formation.
relative
corporate
and external communication
on
Williams and Moffit (1997) reported that although the business and personal factors
play a role in corporate image formation, the organisational factors (i. e. company
buildings, employees, advertising, philanthropy and being a larger insurer) are more
is
following
hypothesis
factors.
Drawing
this
the
than
on
conclusion,
powerful
external
developed which assumesa similar comparison f6r the two major groups of identity
identity
(corporate
mix elements versus unplanned communication
communicators
factors) defined in this study:
3.6. Summary
This chapter reviewed the literature in corporate identity and corporate image areasand
incorporated insights from different fields in order to build the conceptual model
illustrated in Figure 3.1. On the basis of the literature, the relationships between the
identity
image
mix elements as well as unplanned
construct and corporate
corporate
in
hypotheses,
factors
which are summarised
are explained and relevant
communication
98
Table 3.1 and illustrated in Figure j. 2 are stated. In the next chapter, the researchdesign
for
develop
to
constructsas well as to test the model proposedis
adopted
scales
outlined.
Table3.1.The hypotheses.
Hypotheses
HIa: The more positive the attitude that consumershave towards a company's corporate visual
identity systems(CVIS), the more favourable the image they have about that company (IMAG).
Hlb: The more positive the attitude that consumers have towards a company's corporate aesthetics
(AEST), the more favourable the image they have about that company (IMAG).
Hlc: The more positive the attitude that consumers have towards a company's staff apparels (EAPP),
the more favourable the image they have about that company (IMAG).
H2: The more consumersperceive the marketing communication activities of a company reflect its
identity well (COMM), the more favourable the image they have about that company (IMAG)
H3a: The more consumersperceive the managersof a company as representing the company's values
well (MBEH), the more favourable the image they have about that company (IMAG).
H3b: The more consumersperceive the employees of a company as treating customerswell (EBEH),
the more favourable the image they have about that company (IMAG)
H3c: The more consumersperceive a company as socially responsible (CBEH), the more favourable
the image they have about that company (IMAG)
H4: The more positive word-of-mouth consumersreceive from their close friends and relatives about
a company (CWOM), the more favourable the image they have about that company (IMAG).
H5: The more positive word-of-mouth consumers receive from intermediary sources such as media,
NGOs, opinion leader etc. about a company (EWOM), the more favourable the image they have
about that company (IMAG .
H6a: The smaller the corporate associations gap (IGAP), the more favourable the image consumers
have about the company (IMAG).
H6b: The greater the consumer-company identification (IDNT), the more favourable the image they
have about that company (IMAG).
H6c: The more consumers perceive a company's identity emotionally appealing (EMOT), the more
favourable the image they have about that company (IMAG).
H6d: There relationship between emotional appeal (EMOT) and corporate image (IMAG) is mediated
by consumer-companyidefitification (IDNT).
Me: The greater the consumer-companyvalue congruence (CNGR), the more favourable the image
consumershave about a company (IMAG).
H6f. The relationship between consumer-company value congruence (CNGR) and corporate image
(IMAG) is mediated by corporate associationsgap (IGAP).
H6g: The more consumersperceive a company's identity emotionally appealing (EMOT), the greater
the consumer-companyidentification QDNT).
H6h: The greater the consumer-company value congruence (CNGR), the greater the consumercompany identification (IDNT).
H6i: The greater the consumer-company value congruence (CNGR), the smaller the corporate
associationsgap (IGAP).
H6j: The greater the perceived integration of corporate identity mix elements (INTG), the smaller the
corporate associationsgap (IGAP).
H6k: The relationship between the perceived integration of the corporate identity mix (INTG) and
corporate image (IMAG) is mediated by corporate associationsgap (IGAP).
H7: The effect of the corporate identity mix elements (company-controlled communication elements)
is stronger than the unplanned (uncontrolled) communication factors on corporate image.
99
. e
c
0
4)
-4
-um
41).-9.1+1
00
.2
gt
.2
tv
C
C
.....
......
45
0
0
IL)
pl
"Ci
9
Cd
ed
cn
0
0.
!19
tw
.0
SD
.8
zog
Ei
IV METHODOLOGY
4.1. Introduction
Although paradigms can be classified in different ways, they are conventionally grouped
b)
(idealism)
thoughts:
two
a)
positivism
and,
phenomenology
of
schools
major
under
101
Creswell,1994).
directedto theory verification rather than theory generation,it follows a hypotheticodeductive approach (Buffell and Morgan, 1979; Deshpande,1983). Due to the
descriptiveandtheorytestingnatureof this research,a methodologywhich relies on the
is
deemed
in
in
identify
to
the
appropriate
positivist paradigm
order
possibleregularities
corporate image communication (Buffell and Morgan, 1979).
AcademicssuchasMingers(2001)andRobey(1996)suggestthat paradigmsshouldnot
be treatedas mutually exclusive.They believe that the use of different paradigmsand
their pertinentinvestigationmethodscould be appropriatefor achievingcertainresearch
objectives.Although the positivist paradigmwas principally adoptedfor this study,the
following reasonsrequired the consideration of inductive investigation methods:
I-
The literature review revealedthat the scopeof corporateidentity mix has gone
beyond the visual expressionsof an organisation (symbolism) including promotional
the
content of
corporate identity
mix
elements (symbolism,
103
2- Turkey was chosen as the context for this research. Since Turkey is an emerging
little
in
in
has
been
Turkey
that
the
conducted
of previousresearchrevealed
research
field of corporate image and its related areas4.This situation and Turkey being an
emerging market required the researcherto carry out preliminary fieldwork in order
to gain further insights into the practices of corporate image, and the corporate
communication managementof global companies operating in Turkey.
4 Seethe Higher Education Institute of Turkey's website hl! R://www. yok. P-ov.tr/YokTezSrv for the
researchconducted in Turkey.
104
Along with authors advocating pluralist methodologies (e.g. Deshpande, 1983; Mingers,
2001; Robey, 1996) the researcherbelieves that ignoring the possible contribution of the
methods deemed as relevant to non-positivist approaches (e.g. in-depth interviews)
possibly limits the understanding of researcherswho use the positivist approach. The
results of investigations which adopt methodologies using both qualitative and
quantitative methods can provide more insightful knowledge of issues of interest
(Churchill, 1979; Dickson and Albaum, 1977; Dodd and Whipple, 1976; Mingers, 2001;
Sieber, 1973). For example, exploring the field of interest by carrying out some
preliminary in-depth interviews with key informants in the field could lead to some new
insights which would not have been captured by a literature review alone, Accordingly,
this research adopts Churchill's paradigm (1979) which integrates a qualitative
paradigm while being predominantly quantitative in nature (Figure 4.1).
1. Specifydomainof construct I
2. Generatesample of items
Literaturesearch
Contentvalidity (academics,intcrviewees)
Lexical anddesigncheck(academicsand
businessdoctoralresearchers)
1
3. Collect data
4. Purify measure
Cronbach'scoefficientalpha
Phase 3: Main Survev
Actual survey
S.Collectdata
6. Assessreliability
7. Assessvalidity
Cronbach'scoefficientalpha
Content validity
Construct validity
106
A survey method was adopted in the main data collection stage, since positivism entails
testing the hypothesesdeveloped on large samples (Carson et al. 2001). Van Riel et al.
(1998) suggest that surveys provide an opportunity to contact a large audience with
moderatecost (time and funding).
The research methods in corporate image studies differ depending on the scope of the
study. When the study goes beyond investigating the salient attributes of an
image,
the general tendency of choosing qualitative methods shifts to
organisation's
quantitative techniques (Van Riel et al., 1998). The studies by Andreassen and
Lindestad (1998), GUrhan-Canli (1996), Kennedy (1977), LeBlanc and Nguyen (1998),
Simoes (2001), Stuart (1995) and Williams and Moffit (1997) demonstratedthat when a
in
image
identity
to
the
and corporate
researcherwants examine
concepts of corporate
relation to other marketing concepts, quantitative methods are more appropriate than
large
by
These
their
all
conducted
qualitative ones.
researchers
studies
recruiting a
107
This study carried out an exploratory research for the following reasons: (1) to gain an
been
issues
have
(Eisenhardt,
1989),
to
emerge
not
which might
order allow possible
diagnosedduring the literaturereview stage.The information gatheredon the basisof
the qualitative data helped to ensure that all possible items were investigated and
in
its
(Bonoma,
1985).
theory
operation
with preliminarynotionsof
in
based
Istanbul, which have global clients. They provided some
consultancy agencies
I
In line with Churchill (1979), Melewar (2001) and Van Riel (1995) suggestthat any
study in the corporate identity management field should first embark on a situation
analysis via exploratory interviews with company managers.Birkigt and Stadler (1986)
This associationis also a member of the International Public Relations Association based in London.
109
;e
e
EI
(L)
f!
00
>
ZE
-,i
Ei
Ci. g
2
Co
4)
.-r.
A.-=
00
2
,
iR
t
.
-2
=
Iti
4)
re, '
E 1so,
:1
- 'Ei
cu
,-,
-lc
=
u0
4)
-0
42
u
Z
0 .-
ci
0.
E-
Z A
2 ci
4,0
0
0
Ei
>0
rw)
:22
r.
1
(L)
fi
ro-
C2W
0)
t)
r.
c)
U. 0
:i
-84
cu
.5
CD
c
v
r-
r-
%0
ir
c>
r4
r.
r.
4)
0
&g1
0
12
r.
(1
00
CJ
Q
A
tn
0.
1 lri
l
r-
r:
0A
cA&
2
0
0
u
00
4-. t
-0
9)
1881
4)
ju
Ici r4
g
r
Z!
==
Ei
-8
u ,
4)
0
CJ
ci
ci
0,
*C
11
\O
r.
o. 2 - g
2,. In n
cu-
M
c)
2E0
cl
92
92.
4)
4-
%0
(A
>%
r- Ef A
> m
9
5
'
1
0
21
0
,
11
<Z
C0
9: 0
(J 1-
F,
,5
.0 -
55
,4 g> e) c2 0
u.
rA
0.,
Ei ci
0
C) (424
U 1,11,
JD
Co 9)
9-9
0
E- 1
E
0 Ei u E =
12a0
>% 0
C)'0- ,> E k2 0 - -, .121
E =>m
fi
Ici
c:
tu 0
R
4
0
000
z
(L)
-
im 0
b-.
E
u
Ici
C)
t2
0
..
=
E t)
000
>,
0
g-e
Z
4
2
4
>
0y
u3
( " ,
12 ,
;
fj 4".
k)
92.
r.
U
0
t...
Cd
=u29
:22E
&
32
-E-4
ce
cli>
Em0 - 5
0S
2
9-2
E
0
10
E
Ew
0 n 0 o
r_
00
Q0
E l> & Ag
0
rA
0 -Z
t2
0 -8 g
.2E
12
.9
9.1
r= b :20
0 .2
-9
ce
li
rA C2W
r=
0
=
'-9
tCo
CI)
u
ul
-0
>
9)
r.
9
0 ,
cl.
2
.
m
Qgo
2*
ro)
Gn
E
0
cu
.2
ig
00
cn 4) 1-.
.5ee
u
g0S
`s
1-.
Kl> A
.2t
M
Ei
0
ce
>
2
.!
iz
v2
rc-4
0
MO
v)
.
1.1
49
4,e
r.
A
;t
.-
g
E>
rj
4.)
-2
p
Z2
.2 =52
- Zi (D *
4)
.E
cq
Z
rA
CD
.5
.-
e-., 2
6-
X 0 Gn
3
C,
92.
A2
tu
oz
tn
.-
r.
cL
..4
:i
1
> g
CD ;CD
: -7
m
+,
ein
*g
g ,
:1
g
N2 - - 2:
'Z
e .2 9) .t
i ] w <
d
.0
924
Vi
M
9)
(n
cq
0
s2
9
le
10
The companies were chosen from the list of the Reputation Survey 2000 by 'The
7
Since exploratory investigations are less concerned with reaching a group of subjects
who are representative of the main population (Miles and Huberman, 1994), the
112
The next section explains how the information was integrated into questionnaire
development.
113
15
0
bd g
e
Gn
.u
G A
2,2,
18)
4)
2
0
e 12
0U
.(
0
Q
r.
%L U
JD
CU *0
s,
L.
9
0
(.
c.
-2
-; g
0 0
.
. ;50
.2
'
7.
,
1
5
Ei
.
m2 E
E=
A
c
. N.
GSA
Ei 0E
m. m5
.
"
e
:'
.E
:.
.;;;
g
2 JO,
"m
t) c13 c2.
0
u
tu
-0
-;5 A-0 r. 0
8
-OS
.
5
0.
g2e8
zi
m9>,
2
.E
A
4)
12
- -0
wi
cm
4m
ta
0
92. E
4)
4)
>
0
z72
m ,
(d. *-
0
.e
ci
c41,1
Fm
- 4.
0
= u)
rA
020
LZ
".
02t
ce .
-. S
u
CJ
42
1
>,
2cl 92.
9 *a
>,
0
1
2 Ei g3
- 0
in
e
0
-0
2
lg
-a
;
u
0 ci
00
2
4
g
.
2
;
tu r.
in u
0
0
"0
v2
2
10
12
Ici
Ow
(D
,0
.2<
tu
4)
[2
.2
-,1 -
ego
(L)
16.
-0
A' u
m
wl -4)
= ',r4- 4)
&. %g
-u
e:
'
>
CK1
%_,
=2 9
Ei
Ei
rn
ZZ
4)
Gn
00
2
0 10
r4
ho
L:
=O
23
<
41,
GO2
<
311 >
LA J--
<
u2 Co.
A
ri2
c2.
vi
10
rm
The first step in questionnaire development is specifying the domain of the construct. In
The focus of this study is the role of company driven and non-company driven
communication in corporate image building. Therefore, the literature review comprises
identity,
image,
into
corporate
of studies
corporate
corporate reputation, corporate
basis
integrated
On
the
tion.
marketing
communica
of theoretical
communicationand
information obtained, the conceptual framework (See Figure 3.1 in Chapter III) was
ON (D
-0 8
c93
(D
lu
IZ
cui >0
-13 ,
-0 e0
>
>,
ZZ M
<
1-1
CD
C>
011
94
.
ICJ U
ce
-- >, 00
%0 2
'Zi
Gn
; -Z
r-
r(>
(>
00
0
rON
(>
4)
(1)
10
9
"Ci
m
r-
Ei em
(L)
'ZJ fl
-. ===
>
9) 2
ce
:2=0
ce
m2IM
0.
tb 9 z
iz -,
u
,= 'E
iz
'-,
cu
4)
:99
10
r-
ti
Z
"i
.
w
Z
0
Ei
0
mc
<
<<
"Ci
m
:i
10
CM
.-
1-"ri
r. 0 2u , ID
X
x.'
23
"5j ' .
25
(L)
e-. 0
CD
tu
E
CA
f4 ,,2,
C)
c> CD
all
0
%-0
(A
cn
V)
-iz
,
>
Iti
9
Ici
=l- 0
gh
D;
ei jz
0
r.
<
(A
0
&)
Gor
gl
0
u
5
,
,
-3
10 rA 5 I
Z
ce
2
0
qi
.
i.
f- .5
-e
0
ce
tQ
E-9
1111
9x 0
o c rj
C)
*2
0
tu
Gn
(A
<
cm
>
!nu
>,
gA
.2
4) (J 2 Cl. C'.
l
Ei CD.
t9 ,
12
0
om
b-
Z
w
x
L
0
<
0.
>%
u
r. r.
e
44-4
E!
CL. u g
.4 4)
<
5 1-
(A 0
ffl
4)
10
.,
2 ,j
0
9)
ww
u
0
N-g
2
10
w
Awi
= M
j2 E
>
<
+
15
0
0
u
,CU0 <
coi
CO
qi
C.)
CJ
:t
00
0
zi
t)
-e
-,
-e
>:
Z.
x
4.
52
tu
"0
4)
tz8e uj
(D
,
9
9"
r.9-- C)
2-4; 2a
92.
.-
,
'"'
'n
C)
k
Ei
,
'
-9
-I
CD
c Z
(>
2
r,
,:.
lt.;
-10
S ci
9=
e:
10
9m A
2= ,
00
J.,
(:
10
CD
JW2
= 00
aN
:2
(A
2-
CD
c>
iz
eg -0
0
0
0
C.
L4
JA
t 1-. -
CD
c>
-5
CA
(L)
00
rn
9
-0
'UJ
Gn
iz
>
2 E V) r. e'-e:
;t
Ei
cn
CA
rA
;4
l=
(x%
; lz,
all
*Z
;t -CI -
- r4
'b4
r-
cm
45
cn
:i
5
4,
=
0
A
tu
Ln M t
E0
-0
j
In 0
(Y, =mm
>
%Zp.
90
tu
p
all
>
t
:3
%-w kn
42
9
Im
-0
CYN
aN
du
2 i2
<
ij
'0 Zg
u1..
'0,.
000
<4 UM
00
ri
14
u
Ici
-.
<
ri
w
fi
W-9
QZ
Ln
tu xZ)
rA 0
4n
C)
4
<
0
rn
c6)
--,
=
E2
0
u0
4)
0.0 .-
tu
tu
.2
e
11.
4)
"e
u
a 4.
0
e.
0
0
fi
ei
fi
Z zu
s
4)
ci
fi
>
.
g
iv
-3
P
e
t. =
Gn
>
k-
92.
Co
*U1.-
10
92 S
81
.9
Ln
1 0
9'. q>
9)
,Z]
cu
<m
r.
&) tu
(D
ce
j2
10
Ei >
A .'4
rA
<U
(L)
C) ,
'
eg
< 1-, ch
00
m
-0 -u
(U :
r.
15..
8 -a
"Ci
w
zs 10
00
0000
e-
-0
w
10
! E -Z
08
,
2
0>
-0
0
0
e
'
,;
-'
1
, - 0. ,
0
C)
CL)
5
< eCb-4
0K
-0
w m
;e
ri
0
rA
Gn
Z
Z
u
2
0
;e
c3.
> >
K
-(V (U -
I ci
:2
4
t
c2.
I ti
0-0
0,
4
v
2 5,9
0
, >. r,
C>
.0=
>% z4
"CJ ci
00
V-4
Ei 0
0
1-4
4) -CJ
,0
.0=
<
10
0
u
e
r. im
Ici
j, -2
Ln 0
25
.
Wd
2 i
in
(D -
>,
r.
Qj -0
EI
22
2
IN
r. ig
10
119
Brown and Dacin (1997), Davies and Chun (2002), Gardberg:and Fombrun (2002),
Senand Bhattacharya(2001), and Westberg(1994).The traits mentionedby the two
companies (McDonald's and Renault-Mais) chosen as a reference for the
questionnaireswere also incorporated.Corporate image construct's measurement
by
Williams and Moffit (1997). For the other constructs,
based
the
scale
was
on
items were taken from non-empirical articles. The information from the interviews
with the consultantsand the company managersas well as from, focus group
discussionswere also incorporated(Table 4.3). The questionnairestatementsfor
those items were formed by the researcher. Finally, the first draft of the
questionnairewas designed.Table 4.4 shows the initial number of items for each
construct.
items, consumer-companyidentiflcation -6
items, consumer-company
for
interpersonal
items
items),
6
items
for
44
6
communication,
value congruenceintermediary communication, 44 items for corporate associationsgap (corporate
identity traits), 5 items for perceived integration of controlled communication,3
items for company knowledge, and 3 items for corporateimage. In total the first
draft of the questionnairecomprisedof 169items(Table 4.4).
120
Corporate Image
44 items
Corporate Associations Gap (Corporate Identity Traits)
Corporate Identity Mix (Company-controlled Communication)
Symbolism
4 items
Name
Logo
4 items
4 items
Corporate Visual Identity Systems (CV1S)
Slogan
4 items
Colour
4 itcms
I Typography
4 itcms
Corporate Aesthetics
6 itcms
StaffApparel
Communication
9 itcMS
Marketing Communication Mix
Behaviour
2 itcms
Management Behaviour
2 items
Employee Behaviour
6 itcms
Company Behaviour
Unplanned (Uncontrolled) Communication
6 itcms
Interpersonal Communication
6 items
Intermediary Communication
Intrapersonal Communication
3 items
Emotional appeal
6 items
Consumcr-company Idwfification
44 items
Consumer-company Value Congruence
5 items
Perceived Integration
3 items
Company knowledge (Contro Variable)
* The same corporate identity traits were used in order to measure the corporate associations gap ancl
the consumer-company value congruence. Please see Section 4.4.4 for further explanation of the
calculation of these constructs during the analysis.
This third step of Churchill's (1979) paradigm involves reliability and validity
testing for the scale items. It was conducted by an expert judge assessment
(academics),a practitioner check (communicationconsultantsand communication
first
draft
interviewed)
the
the
of the questionnaireto a
application
of
and
managers
200
1).
individuals
1979;
Melewar,
(Churchill,
sampleof
Before using a survey instrument,it must be testedto ensurethat the data collected
how
(1995)
is
Litwin
"what
As
be
matters
states,
not
meaningfuland accurate.
will
121
quantitative the data are but how well the survey instrument performs" (p. 4).
Thereforevalidity and reliability assessments
were consideredat this stage.
Reliability entailsthat "... measuresare free from error and thereforeyield consistent
results" (Peter,1979,p. 6). It thus showswhetherthe scalesare reproducible(Litwin,
1995).Although there are different methodsfor measuringreliability (SeeCarmines
and Zeller, 1979;Litwin, 1995)Cronbach'salpha as an internal consistencymethod
is widely used by researchers.It is a measurethat reflects how well the different
itemspurport to measuredifferent aspectsof a construct(Carminesand Zeller, 1979;
Litwin, 1995).Nunnally (1978) and De Vaus (2002) statethat an alpha equal to or
above0.70 showsthat the itemsmakea reliable set.
122
4.5.3.1.Qualitative assessment
123
both its English and Turkish versions (Craig and Douglas, 2000). They raised
heavy
the
contentand complex wording of the perceivedintegration
concernsabout
and consumer-companyidentification scales. Subsequently,the researcherrevised
these constructs' wording in Turkish. Finally, the Turkish version of the
questionnairewas back translated into English by the researcherand another
bilingual businessPhD student (Craig and Douglas, 2000; Malhotra and Birks,
2003). The following corporate identity traits, which were translatedinto English
differently by the latter two researchers,were taken out: 'conservative','cooperative,
'risk averse', 'enlightened', 'fair,
'stylish', 'tough', 'spirited', 'warm', 'glamorous', 'strong', and 'high quality'. Minor
corrections were made for the rest of the sentencesin the questionnaire.The
translatorstook an iterative approachwhile revising the questions.They double
checkedboth Turkish and English versionsuntil they agreedon the wording of each
sentence(Craig and Douglas,2000). Basedon this filtering, a refined list of a total of
21 items for corporateidentity traits was concluded.The numberof the items for the
other constructsremainedthe same.
An
124
125
items;
items;
items;
items).
logo
list
The
slogan
colour/typography
of
-4
-4
-4
-4
the constructsand the reducednumberof items are illustratedin Table 4.5.
Table 4.5. The constructsand the numberof initial and final items.
Constructs
No. of
Initial
Items
3 items
Corporate Image
Corporate AssociationsGap
items
44
(CorporateIdentity Traits) (1)
Corporate Identity Mix (Company-controlled Communication)
Symbolism
Name
4 items
Corporate
Logo
4 items
VisualIdentity
Slogan
4 items
Systems(CV1S) Colour (2)
4 items
4 items
I Typography(2)
CorporateAesthetics
4 items
Staff apparel
6 items
Communicati2n
---T-9
Marketing CommunicationMix
items
Behaviour
ManagementBehaviour
2 items
EmployeeBehaviour
2 items
CompanyBehaviour
6 items
Unplanned (Uncontrolled) Communication
6 items
InterpersonalCommunication
6 items
IntermediaryCommunication
IntrapersonalCommunication
Emotionalappeal
3 items
Consumer-company
Identification
6 items
Consumer-company
Value Congruence(3)
44 items
PerceivedIntegration
5 items
Company knowledge (Control Variable)
3 items
After
After
Qualitative
Quantitative
Assessment Assessment
3 items
3 items
21 items
21 items
4 items
4 items
4 items
13 items
4i tems*
4 items
3 items
4 items
3 items
6 items
5 Tt-cms
Excluded
2 items
6 items
I items
6 items
5 items
5 items
items
4 items
3 items
6 items
21 items
5 items
3 it-ems
3 items
5 items
21 items
5 items
3 items
126
a. Pilot Study 1
127
The responsesfrom each version were input separatelyinto the statistical package
for social sciences(SPSS).The correlations,factor structuresand coefficient alphas
were computedto havean initial impressionof the reliability of the scalescreatedby
the researcher,as well as the ones adopted from the other studies.Although each
sampledid not provide a minimum numberof five casesper item (Nunnally, 1978;
Peter, 1979)and the measureof samplingadequacywas below the acceptablelevel
for a few constructs(Kaiser-MayerOlkin's measureof sampling adequacystatistic
less than 0.50) (Hair et al, 1998),the results of reliability testing as well as factor
analysiswere encouraging(SeeTable 4.6).
" McPilotI and RMPilotI versionsincludedthe following questionsof the final questionnaire:1-7
and 16-26.
McPilot2 and RMPilot 2 versionsincludedthe following questionsof the final questionnaire:1-3,1112,17-26.
McPilot 3 and RMPilot3 versionsncludedthe following questionsof the final questionnaire:1-12,
18-26.
PleaseseeAppendix 7 andAppendix 8 for the final versionof the questionnaire.
128
w Go
r
.
99
.2.;
>,
0,
Zo
4
Ill
.
No,
0
ON
1.11,
(D
Ci "R
I*: "R
t-
00 110
tn
<
W
'CIO,
061
en tt!
0
22,::
09
co
co
cc W)
(71
m
o
1-0
",
1.01
N.
1.10
'It
N.
.4
00
ell 00
C5
06
r
00
C-i
ON en
00 1.0
(71,
mg
.2
lei lp
[
w 40
c0
C,
'"!
C4
Cc cl
.:
,
> 1
40
01
,<
rA
CD <
en
en
C>
lag
:b ,
2&
QZ
>
r-Zi
ZL
u
,'c,- s lu
m'.o
cc
Tw
,a
a
a
cc
en M
0.0.
en en
en en
r)
89-e4)
''
O
11
00
r-
C71,
00
-cd
r
-.> &0
A 00
C4
-!(:
U
r a,
C'% 00
0,
00
.g0,
to
C71A
cq
en
-t
kn
00
00 0000 00
Z,
2uCA
z to
>% GM -5
19, m
E25
12
.
4,
tA 10
c:
n r,-
C; C5
.0
.0
w m '04
W)
94
N
Q
.0A
No
tn
10
en
en
CS
00
-A
CD
tn
00 00
u
rA
V)
r0
S
0w
z
E E
2
.-
0
;t:
A A
en
C'4
'E'
w 2
vE'
2
on
k.
vi
2t1
10 re 0
>
110.
2 -c,
.02
0>
-0
cl
9
P.
li ,
Ei
.
-0
v211
la*E
2
.2
,
:jQ
=
ci
15
0
0
421
z
C6
a
C4
co
'D
C)
W)
Iu
-.=
.2
A
z
CL
0
.2
44
0 u0 u
'24.
W
l
13l
1-.
0
00
co
92
4) .-.
2 0 cia2 8 :2 ja 0
21
c1
0
in.
vr 0u
u
W
bf
rN
"
.2
CJ4 Z.
,
' 0,2
t3
U2
13 Ul
I
0
65
r,- 13-c, Z;
Gn
:i
-c-&0
17
ll
19
-0 1
M
tj
CC
ZZZZ
0.93
od
en
cc
> 04
C)
C4
00
till
<
10
tn
00
40
93
>
00
t-
N.",
WI)
.Ol
C4
m
00
10
ell
1.01
r-
la
wl
00
tn
1
2
CD
.Ol
t-
1.01
I 'D
r-
Be
O
,
o-t
rn
re
52
.
C.)
12.
4,
vC) .0
U,
11
00
00
41
ON
Z
A
00
CL.
t-
%0
00
oo
00
2 02
CD
C4
AA
CD
10
W)
cc
<
09 09 _.
a -Z
0
AD
'a
0.
0
g
i
.- n
C5 C;
fn
(7
.9
00
all
ON
co
00
(D
rn
-4
C4
o
ttn
.5 i
i
A
0
.
r.
0 0
:s
0
8.
40 8
-tj
.0.,
CO)
cl
. '0
0 1. - (4
-0.*E
-a
>
.-6.
-0
.20
cl
'A
t J22 "i
'0,
'0 j .-.
& :3
cd
cr
2=0r.
2,
-O
Q=2 -a .
"= :5t;-E
'.. 8
o
cd
la
";
PC
*z:
% *9
;3
a Q
R:
Q
Ic
.-
>
c
c
2 '*'
'"')
,
u lU
V0 ci
pa
t
4
--r
L.
-g
at
9 o< g, 4
Lon
4zzz
en4
Factor analysis was run for each of the constructs to test whether they were
defined.
Nunnally
they
theoretically
multi-dimensional
as
are
and/or
unidimensional
(1978) mentions that this assessmentis necessaryto ensure that the empirical
latent
The
root criterion (Eigenvalue >1.00) was used to
measuresare valid.
determine the number of factors (Hair et al., 1998). The items which had
12
less
0.60
than
as well as the oneswith lessthan 0.50 factor loadings
communalities
were excludedin eachrun (Hair et al., 1998).The items which were loadedto more
than two factors, as well as to the theoretically unexpectedfactors,were taken out.
Finally, factors with single items were dropped. The Bartlett Test of Sphericity
(BTS) was significant for all factor analysesrun, which shows that correlations
among variables were present(Hair et al., 1998). Except Staff apparel, employee
behaviour, intermediary communication and perceived integration constructs,the
Kaiser-Mayer-Olkin measureof sampling adequacy(MSA) was-above0.60 for the
rest of the constructswhich indicatesan acceptablelevel (Hair et al., 1998)(Table
4.6).
All the constructsshowed high reliability levels except the perceived integration
construct.This was due to the fact that the factor analysisfor this constructresulted
in two factors (rather than one factor) which was not expectedtheoretically (See
Table 4.6). As Carminesand Zeller (1979) state,this result was due to a non-random
measurementerror. The two items which were loadedto one factor were negatively
factor
loaded
items
to
the
the
the
which were
second
rest of
constructed,whereas
were stated positively. It was concluded that the structure of the responsesets
triggered a general tendency to respondto the items in a similar manner, which
12For the following constructsthe items which had communaliticslessthan 0.50 were excludedfrom
the analysis since the communalities for the majority of the items were between 0.50 and 0.65:
corporateimage,companybehaviour,communicationand interpersonalcommunication.
131
b. Pilot Study 2
After the refinement of the scales on the basis of the Pilot Study 1, all items were
132
Table 4.7. Reliability measuresfor the constructson the basisof Pilot Study2.
Constructs
No of
items
3
Corporate Image
21
Corporate Associations Gap (Corporate Identity Traits)
Corporate Identity Mix (Company-controlled Communication)
Cronbach
alpha
Me Data
0.824
N/A
Cronbach
alpha
RM Data
0671
N/A
Symbolism
Corporate Visual Identity Systems
0.913
13
0.897
Corporate aesthetics
0.858
4
0.889
1 0.761
1
Staff apparel
3
0.915
Communication
0.919
0.919
Marketing Communication Mix
5
Behaviour
I
N/A
Emplq)ve Behaviour
N/A
6
0.893
0.791
Company Behaviour
Unplanned (Uncontrolled) Communication
Interpersonal communication
4
0.877
0.934
Intermediary communication
4
0.828
0.910
Intrapersonal communication
3
0.957
0.721
Emotional appeal
0.906
Consumer-company identification
5
0.905
21
N/A
N/A
Consumer-company congruence
5
0.890
0.872
Perceived integration
3
0.824
Company knowledge (Control Variable)
0.915
For the corporate identity traits, reliability test was not run, since those items were only used in
order to calculate the difference scores for the corporate associations gap as well as the consumercompany value congruence constructs rather than proving it as a scale. Please see Section 4.5.4 for
the explanation of these constructs' calculation.
Note: Due to the small number of cases,the missing values were replaced with the mean.
134
135
Petty and Cacioppo (1986) in their elaboration likelihood model assert that the
involvement of the subject to the object to be evaluated defines the level of
in
elaborationoccurring the communicationprocess.For example,an individual may
be motivatedand able to assessa company's productsand servicesclosely but may
be less interestedin its buildings. While for the former situation a high elaboration
forming,
in the latter situation a personrefers to
leading
to
opinion
rational
occurs
peripheralcues(low level of elaboration)such as the shapeof the building without
attachingany meaningto it. On the basis of this discussion,it could be concluded
that the respondent's psychological and physical distance from the aspects of
corporate identity mix (corporate communication) elements and unplanned
(uncontrolled) communicationelementswill vary on the high elaborationand low
elaborationcontinuum.Therefore, Stuart (1995) proposesthat basedon the totality
is
level
individual's
the
concrete
and
abstract
conviction,
of
a middle
of elaboration
more likely to occur during the assessment
of the communicationchannelsand tools
that the respondentscome across(SeePoiesz, 1988;Van Riel, 1995).According to
Van Riel (1995), in that situation, attitude measurementis the proper method for
corporateimageresearchsince it allows the respondentsto compareand appraisethe
image
factors.
communication
of
and
attributes
136
137
in
VII
Appendices 7 and 8). The calculation of the
(See
question
main survey
differencescoreswas as follows:
138
The consumer-company
value congruenceis calculatedas follows:
Value Congruence
C=Ei Aij - Sij i where,C- Consumer-company
Aij - Evaluation of corporateassociationby individual along
corporateidentity trait i byjth individual;
Sij - Evaluation of an individual's own personality along
16
dimension
byjth
individual
i
personality
16It should be noted that same adjectiveswere used to measurethe corporate identity traits and
difference
be
in
to
traits
able
calculate
scoreson the samecharacteristics.
order
personality
139
Churchill (1991) states that survey research is presumed'to have high external
validity; that is, the results can be generalisedto a population. The structuredundisguisedsurvey, where a formal questionnaireis prepared,is the most popular
data collection method becauseof the simplicity and flexibility of the research
technique(Churchill, 1999;Van Riel et. al., 1998).
the researcherto focus on Turkey as the context of the study . Garten(1997) states
that Turkey is one of the emerging markets which may influence the world trade
substantially. The studies of corporate image as well as corporate identity and
corporatereputationhave coveredthe US and a numberof Europeancountriessuch
as the United Kingdom, Netherlands, Germany, Belgium and Greece (e.g.
Andreassenand Lindestad, 1998; Balmer, 1995; Boyle, 1996; Brown and Dacin,
1997; Fill and Diminopolu, 1999; Fornbrun et al., 2000; GOrhan-Canli, 1996;
Kennedy, 1977; Simoes,2001; Thevissen,2002; Van Riel, 2002; Westberg,1994;
Wiedmann,2002).
Burgess and Steenkamp (2006) argue that economic instability and social diversity
may disrupt implementation of marketing approaches developed for highly
industrialised countries in emerging markets such as Turkey. Therefore, a continuous
issues
is
important
in
information
these
on
and
analysis
order to
gathering
marketing
be able to respond to any likely changes in the market. In terms of corporate identity
in
in
its
the
change
social
cultural
especially
and
norms
and
perception,
and values
17Pleaseseethe HigherEducationInstituteof Turkey'swebsitehttp://www.yok.jzov.tr/YokTezSrv
in Turkey.
for thethesesconducted
141
be
for
may
of
concern
markets
organisations. Social and cultural
emerging
environment may not only have an impact on how consumers' evaluations of
companiesas a whole and their perceptionsof specific company identity traits in
comparisonto their own values, but also on which communicationchannelsattract
attention of individuals while forming corporate image. Therefore, developing a
comprehensivemodel which integratescompany-drivencommunicationfactors and
externalfactorsand testing it in an emergingmarket environmentmay be important
for organisationsin order to understandthe dynamicsof corporateimage formation
in termsof internationalmarketingpractices.
The survey was conductedin Istanbul. According to the year 2000 censusfigures,
this city is the most populatedcity in Turkey with approximately10 million people,
which is inth of the whole Turkish population(DIE, 2003). It is locatedin the most
142
populated (Marmara) region of Turkey with a 25% share of the total population
(Euromonitor, April 20oo). its population has increased12.4 times between 1923
and 2000, whereasthe overall population in Turkey has increasedonly five times for
the sameperiod (DIE, 2003). These figures indicate that Istanbul can be considered
as embracingthe Turkish peoplefrom different backgrounds,therebyshowinga high
potentialof representingthe overall population.
Data analysiswas completedin three stages.In the first stage,the content and the
relevanceof the multi-item scaleswere refined on the basisof the quantitativeand
qualitative data gatheredfrom the exploratory stageof the research.In the second
143
basis
the
on
validated
of the quantitativedata obtained from the
stage,scaleswere
main survey.Lastly, the final model was tested.
144
According to Hair et al. (1998) and J6reskogand S6rbom (1993), this analysis is
in
be
for
further
to
testing
scales
need
summated
constructed
very useful when
dimensionality
helps
It
to
the
assess
and the
proposed
structural models.
interest.
for
In this study,
the
selected
variables
of
of
each
construct
appropriateness
the principal componentapproachwas used since the aim of the researcherwas to
identify a minimum set of variableswhich accountedfor the maximum variancein
the data (Hair et al., 1998).The numberof factors was defined on the basis of the
latent root criterion (Eigen value >1.00). The use of this methodwas to ensurethat
"any individual factor should accountfor the varianceof at least a single variable"
(Hair et al., 1998,p. 103). In order to achievethe best possibleinterpretationof the
factors, the varimax rotation method was used. This is an. orthogonal rotation
techniquewhich is suitablefor reducingthe numberof variablesto smaller subsets.
Additionally, the significanceof the factor loadingswhich determinesthe correlation
145
betweenthe variable and the underlying factor was assessed.The factor loadings
above +1-0.50 were considered practically significant (Hair et , al., 1998).
Furthermore,the communalitieswhich indicatethe amountof varianceeachvariable
shareswith the rest of the variablesin the analysiswere examined(Hair et al., 1998).
The variableswith communalitiesless than 0.60 was deemedas not contributing to
the variance explained and were therefore dropped from the analysis (De Vaus,
2002).
For the factors derived from the exploratory factor analysis, Cronbachalpha was
computedin order to test whether each subsetof items were internally consistent
(Carmines and Zeller, 1979; Litwin, 1995; Parasuramanet al., 1998). That is a
method which is widely used in social sciences(Churchill et al., 1974; Churchill,
1979;De Vaus, 2002; Litwin, 1995;Peter, 1979).The valuesequalto or above0.70
were consideredto be of an acceptablelevel of reliability (De Vaus, 1996;Nunnally,
1978).
4.7.2. Confirmatory
1991). In order to claim that a construct is valid the following criteria should be
assessed:(1) unidimensionality of a construct (Gerbing and Anderson, 1988;
Steenkampand Trijp, 1991)(2) reliability, (3) convergentvalidity, (4) discriminant
validity, and (5) nomological validity (Gerbing and Anderson, 1988; Peter, 1981;
SteenkampandTrijp, 1991).
set of logical (theoretical)indicators (Hair et al., 1998; Hattie, 1985; Gerbing and
Anderson, 1988; Steenkampand Trijp, 1991). It permits the computation of the
criteria for assessingnomological validity, convergent validity and discriminant
validity.
.2
Fornell and Larcker (1981) mention that "neither [pj2 nor p, alone] measures'the
in
by
is
that
the
captured
construct relation to the amount of
amount of variance
variancedue to measurementerror" (p. 45). They suggestthat the averagevariance
extracted(p,) should be calculatedin order to assessthe total varianceaccountedfor
by the latent construct and its indicators. Diamantopoulosand Siguaw (2000)
mention that " ... p, values [the averagevarianceextracted]lessthan 0.50 indicate
that measurementerror accountsfor a greateramount of variance in the indicators
than doesthe underlying construct" (p. 91). This raisesquestionsabout the validity
of the indicatorsand the measureused (Diamantopoulosand Siguaw,2000; Fornell
149
error varianceof the indicator and E is summationover the indicators of the latent
variable (Diamantopoulosand Siguaw,2000; Fornell and Larcker, 1981;Hair et al.,
1998).
150
151
The chi-squarestatistic()?) is " a test of perfect fit in which the null hypothesisis
...
that the model fits the population data perfectly" (Diamantopoulosand Siguaw,
2000, p. 83). When it is statistically significant, it indicatesthat the null hypothesisis
rejected.Unlike the conventionalhypothesestesting, this is a condition for claiming
good model fit in structuralmodel'estimation.Chi-squarevalue is computedas "(N1) F. i,, where N is the sample size, F,,i,, is the value of the fitting function
at
...
convergence[and] ... the relevant degreesof freedom is calculatedas Y2k(k+l) - 1,
where k is number of observed variables and t is number of parametersto be
estimated"(Diamantopoulosand Siguaw,2000, p.83).
152
The Normated-fit index (NFI) compares the base model with the suggested model
nuli
(Xnujjdegrees
freedom.
is
It
of
)?proposed)/
computed as
without considering the
where ;? is chi-square value (Hairet al., 1998, p. 657). It can have values between 0
is
is
0.90
Even
1.00.
threshold
though
there
above
value,
and
not an absolute
and
1998).
indication
fit
(Hair
al.,
of
good
et
as
an
mostly recommended
The non-normated fit index (NNFI) (Tucker-Lewis index) compares the null and the
into
both
degrees
freedom
by
taking
the
of
models
of
proposed models
2
/
dfnull)
/
/
df,,
dfproposed)]
is
[()?
/
(X
It
O?
calculated
as
consideration.
null
proposed
null
ull) where ;? is chi-square value and df is degrees of freedom (Hair et al., 1998). Unlike
1.0
indices,
in
incremental
fit
NNFI
indices
than
take
the
values
greater
all
(Diamantopolous and Siguaw, 2000). However, the recommend level of good fit
value is the same (0.90 and above is as accepted good fit) (Doll et al., 1994; Hair et
al., 1998).
153
4.8. Summary
Throughout this chapter,the researchdesign of the study and the details of each
information
from
design
incorporated
have
been
The
three
research
presented.
stage
stages of data collection. First, the exploratory researchwas conducted which
did
focus
discussions.
This
interviews
in-depth
not
phase
and
group
composedof
it
design
insights
into
interest,
the
the
of
also
supported
subject
of
provide
more
only
the research instrument. Following that step, two pilot studies were conducted
last
The
stageconsistedof a survey,the resultsof which are explained
successively.
in the next chapter.
154
5.1. Introduction
The findings of the main survey of this thesis are presentedin this chapter. It is
composedof four main sections.In the following sections,first the details of the
sampling in ten-nsof samplesize and respondentprofiles are presented.Second,the
initial dataexaminationis discussed.Third, the scalevalidation and the measurement
models are explained. Finally, the structural model and the related findings are
illustrated.
5.2. Sampling
5.2.1. Sample size
1. nl= 22* S2 /H
2* X2
155
2*
2. n2 =s z2/H 92
where, s-
parameter(cr)
z- standardnon-nalvariable
H' - Level of precision(In this caseH was multiplied
with X in orderto get an absolutevalue)
(Malhotra and Birks, 2003)
3. n3 = 0.83332* Z2/H 92where,0.8333- standarddeviationof the population"
z- standardnormal variable
H' - Level of precision (In this caseH was multiplied
with X in order to get an absolutevalue)
(Malhotraand Birks, 2003)
156
0
C14
%D
to)
92
N
ICD 00
r-
W
4
91
CD
10
Vi I-q 09
cl
lz
iz
00
en
rn
%C
Go
CN
0 0
(14
%0
14,
00
Nt
eq
C)
m
00
Ir
kn
00
tn
(D
4r)
0
6
qn
CD
tn
CD
d
en
C)
o
W %0
1.0
tel
00
t-
m
00
eq
en
t-
tf) 0
00
en
en
en
Itt r
en
I-T
en
(4
00
rto
in m
10o
t-
oq
I-
r-
It
tn
q
tn
Q
6
4.
4.
0
cn
&n
Cd
9i
r.
&.
rA
10
i
I.
.
"0
r_
-5
Cd 0c
bo -EL
CL
0
Q
9
eq
en
cu
r.
Gn
0
Ln ':
Eo
a S
0s
-s
U
g
EU
.
5
r
V.
E
,
0
a
E
8.
0
0E
B
4
E
.Eg
0
E
>
U E
c
40
4-2
- -s
eel 00 t- r- rq Cq CS
>-.-:,
en
eq
93
Ch (71
8
1. C13
t.
t
a)
rA :s
&.
0Z =
15
. 0
0
0
9
en
Cq
en
; 51 ,p,
'd
'n Cd
Eo
U0
> "B v -S g
EU
'a W5
- r.
(a w E wE 8 0
:
T
-s
E
0
.
4W.
>
0
2.8
=
.
>1
10%00
1010 oo
en
It
en
-a
4K ,o
00
ej
1.0
N
as
in
;t 'o
t-
G* t-
as
eq
E
o
wl
om
ON
00 ON
Gq
C aq
llq
C oq
t-
iz
Q
en
en en
en
en
en
en
oq C
en en
en
In
fn
en
4.
0
7EL
;Ja,z
E 0
-Ei
9
0 ;
m.
E
CL2
a --
E 8 8
.R , -5
0 co 0
in
iz
C:
R-g
o
"g
>, 4) La .5 -E =
.2
;;
r-
-.
UFO
>
00
.2q
.ca
j
C;
:3
cd
al
:s
Gm
0
*
C -e
C13 0
0
> 1 r_ 9
>,
9
qu 16
.-
>
Ej
'-=
A8
r-
9
.,;
T
8 (.)
I
I
As mentioned in Section 4.6.2, the data for the main survey was collected in
Istanbul,Turkey. Two samplesof individuals were recruitedfrom eight boroughsof
Istanbul. Since it is important to ensurethat a sampleis representativeof the main
figures
demographics
1999),
(Churchill,
the
the
of
sample
were computed
population
and comparedwith the main population figures both for the McDonald's and the
Renault-Maiscases.Table 5.2 summarisesthe profile of eachgroupof respondents.
The characteristics of the respondents such as age, gender, marital status, education
level, income level and employment status were asked in the questionnaire. It was
found that 48.4% of the respondents were female in the McDonald's case, while
44% of the respondents were female in the Renault-Mais case. These figures
distributions
for
both
the
that
gender
of the respondent groups were
reflected
= 49.1%; 7r.,,j,= 50.4%). Almost 60% of
representative of the main population Orfemal.
55.7%
the respondents in both data sets were between the ages 18 and 35 (PMC18-35'*=
figure
(nJ8.35=
When
58.3%).
to
the
compared
population
main
and pRmig-35=
55.9%), this finding showed that the targeted respondentswere recruited properly.
158
McDonald's Data
(%)
Renault-Mais
Data
Main Population
Gender
Female
48.4
44.0
49.
Male
51.6
56.0
50.4
Age
18-35
55.7
58.3
55.9
36-55
33.5
31.7
34.7
56+
10.9
10.1
9.4
Marital status'
Single
35.7
34.9
35.5
Married
64.3
65.1
69.5
Education level
Secondaryschooland below
57.0
57.8
64.9
28.5
28.4
23.4
High school
14.5
13.8
University and above
11.7
Income leve12
Lessthan 1,000YTL
41.2
36.2
n/a
1,001TL - 2,000 YTL
46.6
49.1
n/a
2,001- 3,000YTL
7.7
7.8
n/a
1.8
3,001 YTL and above
6.0
n/a
Employment status
43.9
Employed
49.1
47.0
Unemployed
56.1
50.5
53.0
Occupation
Top executiveor manager,
Ownerof a largeor mediumsize
7.1
14.7
5.4
company,Lawyer, dentist,
arc ec etc.
Office/Clerical staff, Civil
11.9
12.4
13.0
servant
24.9
22.0
Worker, Craftsman
28.6777]
1) Widow and divorced individuals were addedto the single category,and individuals living with
their partnerswere includedin the marriedcategory.
2) The amountof net income per month is presentedin YTL rather than billion TL (See questions
XIII in Appendix 7 and 8) in this table, sincethe last six digits were removedfrom the currencyin
Turkey by January2005. For example,I billion TL correspondto 1000YTL.
Note 1: McDonald's casesamplesize N= 221 except income level (N - 115); Renault-Maiscase
samplesizeN= 218 exceptemploymentstatus(N = 217) and incomelevel (N - 215)
Note 2: The figures of the main populationare gatheredfrom DIE (StateInstitute of StatisticsPrime
Ministry Republic of Turkey) (2003) 2000 Censusof PopulationTurkey: Socialand Economic
Characteristicsof Population,DIE Printing Division, Ankara.
Note 3: n/a: not availablein DIE statistics.
159
both
level
the
and
employment
characteristic
samples were very
of
status, education
11.7%, 7remployed=
close to the main population figures (7r,i,,&= 30.5%, 7runiversity+=
47%). Almost all of the percentages related to the occupation profile of the
respondents in both samples were close to the main population figures except for the
first occupation category in Table 5.3 for the Renault-Mais data. There was a larger
discrepancy between the sample and the main population values for this occupation
finding
14.7%
5.4%).
However,
(PRM-occupation=
this
vs.
7toccupation=
was not
group
considered to be a serious problem. Since cars are much more expensive products
than fast food, it could be expected that a higher proportion of the respondents who
in
interested
familiar
answering the Renault-Mais questionnaire could
were
with and
be-from higher salaried positions such as top executives, middle managers, doctors
etc.
in the Renault-Mais
160
knowledge(tmc=-1.346,pmc=0.180 > 0.05). 39% of the respondentsin the RenaultMais samplementionedthat they or a member of their family used to or currently
own a Renault car. There was no statistically significant difference betweencar
ownersand non-ownerson the basisof the companyknowledgethat they haveabout
Renault-Mais(tRm=1.026,ppm=0.307 > 0.05). All of theseresults showedthat the
respondentsin each samplehad a sufficient level of contact with and information
about the casecompaniesin order to have the ability to answerthe questionsin the
questionnaire.
161
The aim of the initial data examinationwas to ensurethat the data was ready for
further applicationof the multivariate dataanalysistechnique.As well as helpingthe
basic
to
a
understandingof the data, this Mage also gave the
researcher attain
opportunity to detectany violation of the underlying assumptionsof the multivariate
techniquesapplied (Hair et al., 1998).As recommendedby Hair et al. (1998) and
Tabachnickand Fidell (2000), the following analyseswere performed:(1) graphical
(4)
data,
(2)
(3)
descriptive
of
analysis
missing
and
analysis, analysisof outliers,
and
test of normality.
First, the frequencytableswere tabulatedboth for the McDonald's and the RenaultMais data setsin order to find any valuesoutside the value range for eachvariable
due to wrong insertionof the codesinto the SPSSdata sheet.Next, the shapeof the
distributions for eachvariable was examinedby looking at the histogramsand stem
and leaf diagrams.Finally, box and whisker plots were plotted and outliers noted
(Hair et al., 1998;Tabachnickand Fidell, 2000).
Hair et al. (1998) mention that it is betterto deletethe outliers which are considered
to be non-representative
of any observationsin the population;otherwisethey should
be retainedin order to increasethe generalisability of the multivariate analysis.The
for
25%
items
the
than
of
more
were removed
answers
with
missing
questionnaires
from both the McDonald'sand the Renault-Maissamples.Coincidently,this groupof
casesalso included all of the problematic outliers. As a result of the procedure
described,31 caseswere excludedfrom the McDonald's sampleand 37 caseswere
162
The missing value analysis for the McDonald's data showed that there was a
potential pattern in the data towards not respondingto the variables presentedin
Table 5.3. Many of the t valueswere significant (p < 0.05) for eachvariable on the
have
In
(1998)
if
Hair
these
that
the
variables.
other
situations
et al.
suggest
variables
a theoretically core role in the research,they could still be kept for further analysis.
On the basis this argument,the items in Table 5.3 were still included in the data
analysisof the measurementmodels(SeeSection5.4.2).
163
Table 5.3. The variableswith possiblemissing data patternsin the McDonald's data
on the basisof West.
Construct
CorporateVisual Identity
Systems(CVIS)
CompanyBehaviour
Marketing Communication
Mix
IntermediaryCommunication
Perceivedintegration
Item
0 The sloganof the companycommunicateswhat it standsfor
0 It is easyto recall the sloganof the company
9 The sloganof the companymakesme havepositive feelings
towardsthe company
0 This companysupportscorporategiving
* This companytreatsits employeesvery well
0 This companycaresaboutenvironmentalissues
0 This companytreatsthe local public very well
0 The sponsorshipactivities of the companyreflect the image
definedin the paragraphabove
0 The media,opinion leaders,governmentinstitutionsandNGOs
etc. talk aboutthis companyfrequently
01 hearpositive things aboutthis companyfrom the media,
opinion leaders,governmentinstitutionsand NGOs etc.
0 The media,opinion leaders,governmentinstitutionsandNGOs
etc. areproud of this company
0 The media,opinion leaders,governmentinstitutionsandNGOs
etc. recommendthis company
9 It seemsthat all communicationactivities of the companyare
plannedand executedby the samepersonor people
Note: During the t-test, the variables that were missing values for less than 5% ofcases were omitted
from the analysis. It was considered that the likelihood of a missing value pattern occurring
was small when less than 5% of the respondents did not answer that question (variable).
Little's MCAR test showeda significant difference (p < 0.05) betweenthe observed
iormer
test (Hair
and expectedmissingdatapatternswhich confirms the result of the
in
deal
bias
In
1998).
to
the
order
with
potential
results due to non-random
et al.,
missing values, the main data analysis for the McDonald's casewas basedon the
data in which the missing values were replacedwith estimatedmean (EM) values.
SinceEM, being a maximum likelihood estimationmethod,makesthe most accurate
and reasonableestimatespossiblecomparedto meansubstitution,casesubstitution,
regressionimputation and multiple imputation (Hair et al., 1998),it was considered
to be the most appropriatetechniqueto solve the missing data pattern issue in the
McDonald's data.
164
CorporateAesthetics
Staff Apparel
EmployeeBehaviour
CompanyBehaviour
Item
0 The architecture and the interior design of the company's
buildings, sales and after sales service centres etc.
communicatewhat it standsfor
01 like the architectureand the interior designof the company's
buildings, salesand after salesservicecentresetc.
0 The architecture and the interior design of the company's
buildings, sales and after salesservice centresetc. are easily
recognised
0 The architecture and the interior design of the company's
buildings, sales and after sales service centresetc. make me
havepositive feelingstowardsthe company
0 The apparel of the employeessuch as salesmen,after sales
servicestaff etc. communicatewhat the companystandsfor
01 like the apparel of the employeessuch as salesmen,after
salesservicestaff etc.
0 The apparel of the employeessuch as salesmen,after sales
service staff etc. make me have positive feelings towards the
company
0 The employeesof the companysuch as salesmen,after sales
servicestaff etc. treat customersvery well
0 This companysupportscorporategiving
0 This companytreatsits employeesvery well
0 This companycaresaboutenvironmentalissues
0 This companytreatsthe local public very well
0 This companyrespectsconsumerright
10This companytreatspeoplewith high standards
Note: During the West, the variables that were missing values for less tilan 5% or cases were omittc(i
from the analysis. It was considered that the likelihood of a missing value pattern occurring
did
less
5%
the
than
not answer that question (variable).
respondents
of
was small when
Although the West analysis for the Renault-Maisdata showed a potential missing
for
few
in
5.4,
for
items
Table
t-values
the
eachvariable on the other
value pattern
between
difference
observationswith missing values
variablesshoweda significant
indicated
This
that a missing data pattern was
result
and without missing values.
unlikely to be presentin the Renault-Maisdata.The Little's MCAR test corroborated
this finding. The significance level of the Little's MCAR test was 0.77, which
showedthat the missing valuesmay be regardedas completely random (Hair et al.,
1998). In these cases,Hair et al. (1998) suggestthat any imputation method for
data
be
Since
the
merged
setswere usedin further stages
used.
missingvaluescould
165
in managerialand social
is
it
unlikely that the statistical assumptionswill ever be met in a
scienceresearch
in
(p.
Bentler
Chou
(1987)
81).
that
and
state
structural equation
strict sense"
likelihood
(ML) estimators are almost
"...
theory
maximum
normal
modelling
always acceptableeven when data are non-normally distributed" (p.89). Although
instead
free
distribution
ML
of
asymptotic
estimation methodsmay
application of
yield an untrustworthyX2(Chi-square)statistic and standarderrors, if the model fit
indices show reliable results it may be concluded that this problem has been
data
1987).
Therefore,
Chou,
(Bentler
the
transformation
of
and
none
overcome
approachesto achievenormality were appliedat this stage.
166
Exploratory factor analysis (EFA) is a useful technique in the early stages of scale
(Aaker,
1997;
Babin et al., 2000) since it allows the
and
validation
refinement
researcherto have a preliminary understandingof the relationships between the
I
indicators and their relevantconstructs.It becomesespeciallyuseful when there is
167
in
investigation
known
little
(Gerbing and
theory
the
about
constructs
under
very
Anderson,1988).Sincemost of the itemswere generatedfrom anecdotalarticlesand
somewere adaptedfrom empirical studies(SeeTable 4.3), it was necessaryto apply
EFA.
Table 5.5 shows that four major factors were suggestedas the dimensionsof the
corporate identity mix elements (company-controlledcommunication elements).
That is to say that the eigen valuesof four underlying factors were bigger than one
(Hair et al., 1998). The sample was'adequatefor the factor analysis in that the
Kaiser-Mayer-Olkin measureof sampling adequacy(MSA) was 0.924, which is
consideredas marvellous (Kaiser, 1974),and the Bartlett Test of Sphericity (BTS)
168
bivariate
items
the
that
the
correlations
among
were significantly
suggested
scales'
different from zero (BTS= 6142.91, p=0.000). These four factors captured an
acceptablelevel of 66.91% of the variance(Hair et al., 1998).
In the literaturereview (SeeChapter111),corporateidentity mix elements(companycontrolled communicationelements)were defined as consisting of the following
corporatevisual identity systems(CVIS), corporateaesthetics
multi-item measureS22:
(AEST), staff apparel (EAPP), company behaviour (CBEH) and marketing
by
(COMM).
This
factor
was
elements
structure
not
confirmed
mix
communication
the outcomeof the EFA (Table 5.5).
Even though all of the 13 items for the CVIS constructwere not reservedby the
items
in
first
factor were able to cover the content of
7
the
the
remaining
analysis,
the CVIS. In contrastto the theory, the items for the corporateaestheticsand the
factor
in
dimensions
than
two
one
separate
rather
constructs
resulted
staff apparel
(Factor 2). The items for the company behaviour (Factor 3) and the marketing
communication mix elements (Factor 4) were retained as theoretically expected
(Table 5.5).
measuredwith only one item it was not includedin exploratoryfactor analysiswhich was aimedto
refine the multi-item measures.
169
Table 5.5. Exploratory factor analysis results for the corporate identity mix
(company-controlledcommunication)elements.
Factors and Related Items
Corporate Visual Identity Systems (CVIS)
I like the name of the company
The name of the company makes me have positive feelings towards the company
11ike the logo of the company
The logo of the company makes me have positive feelings towards the company
The slogan of the company makes me have positive feelings towards the
company
I like the colour and typography used on all visual materials of the company
The colour and typography used on all visual materials of the company make me
positive feelings towards the company
_have
Corporate Design (CD)
I like the architecture and the interior design of the company's buildings,
Factor
Loadings
0.633
0.750
0.664
0.765
0.756
Cronbach
Alpha
0.895
0.653
0.750
0.558
restaurantsetc.
The architectureand the interior designof the company'sbuildings,restaurants
0.555
0.805
etc. makeme havepositive feelingstowardsthe company
1 like the appearance
0.785
and apparelof the employeessuchas cashiers,
etc.
waiterstwaitresses
The appearance
and apparelof the employeessuchas cashiers,waiters/waitresses 0.775
etc. makeme havenositive feelinastowardsthe comnanv
Company Behaviour (CBEH)
0.671
This companysupportscorporategiving
0.710
This companytreatsits employeesvery well
0.905
0.751
This companycaresaboutenvironmentalissues
This companytreatsthe local public very well
0.799
This companyrespectsconsumerright
0.798
This companytreatspeoplewith high standards
0.803
Marketing Communication Mix (COMM)
The company'stv, radio or printed advertisingsaboutits productsand services
0.832
reflect the imagedefinedin the paragraphabove
The public relationsactivities of the companyto promoteitself aswell as its
0.868
productsreflect the imagedefinedin the paragraphabove
0.901
The productsand servicesof the companyreflect the imagedefinedin the
0.735
paragraphabove
The sponsorshipactivities of the companyreflect the imagedefinedin the
0.789
paragraphabove
The company'stv, radio or printed advertisingsto promoteitself reflect the
0.800
imagedefinedin the paragraphabove
Kaiser-Mayer-Olkinmeasureof samplingadequacy- 0.924
Total varianceexplained= 66.91%
Bartlett Test of Sphericity--6142.91, p=0.000
Note 1: The items which had communalitiesless than 0.50, the ones with less than 0.50 factor
loadingsas well as the oneswhich were loadedto more than one factor were excluded(Hair et
al., 1998).
Note 2: Principal componentanalysisand orthogonalvarimax rotation were used.
170
by the Cronbachalphameasure.
The internalconsistencyof eachfactor was assessed
In the light of the conclusionabove,the following rearrangementsof the items and
their underlyingconstructswith their re-specifiedlabelswere consideredasthe most
relevant dimensions of the corporate identity mix elements (company-controlled
communicationelements):
171
Van Riel, 1995)as well as itself (Argenti, 1998;Hunt and Grunig, 1994).Consistent
initial
items
the
theory,
the
all
were retainedin this factor.
with
The three factor structure below (Table 5.6) was concluded for the unplanned
(uncontrolled)communicationelements.The factorswhich had eigen valuesgreater
than 1.00were preserved.The MSA measureof samplingadequacy(MSA = 0.910)
(Kaiser, 1974)andBartlett Test of Sphericity(BTS=7624.34,p=0.000)demonstrated
that the EFA was appliedcorrectly. The amountof varianceexplainedby thesethree
factorswas 78.95%(Hair et al., 1998).
0.70,
Cronbachalpha statistics for each factor (ccf,,,
0.947
0.877
>
>
=
ccf.,
2=
t,,,,
t(,,
0.70, (XfactoO 0.964 > 0.70) confirmed that the items in each factor were intemally
consistent(Nunnally, 1978) (Table 5.6). This result indicatedthat thesefactorscan
be consideredasthe basisfor the conflrmatory factor analysis(CFA) application.
In contrastto the theory, it was found that both the intermediary(EWOM) and the
interpersonalcommunication(CWOM) items were loadedon the samefactor rather
than two distinct dimensions.The researcherconcluded that while people were
forming an image of a company,the reception of informal information about that
company may be more important than from whom or from which media that
information was disseminated.Thereby the loading of all items onto one single
factor was consideredtheoretically acceptable.The second and the third factors
items
for
is,
the emotional appeal (EMOT) and the
that
all
as
expected,
resulted
consumer-companyidentification (IDNT) constructs loaded to their underlying
dimensions.
172
Table 5.6. Exploratory factor analysisand Cronbachalpha results for the unplanned
(uncontrolled)communicationelements.
Factors and Related Items
Factor
Loadings
Word-of-mouth (WM)
My close friends and relatives talk about this company frequently
1 hear positive things about this company from my close friends and relatives
My close friends and relatives recommend this company to me
My close friends and relatives encourage me to make purchases from this
company
The media, opinion leaders, government institutions and NGOs etc. talk about
this company frequently
I hear positive things about this company from the media, opinion leaders,
government institutions and NGOs etc.
The media, opinion leaders, government institutions and NGOs etc. are proud
of this company
The media, opinion leaders, government institutions and NGOs etc.
recommend this company
Cronbach
Alpha
0.815
0.815
0.827
0.849
0.860
0 947
.
0.832
0.853
0.844
0 877
.
0 964
.
78.95%
Note 1: All items were retained.There were no items with communalitiesless than 0.50 and with
factor loadings less than 0.50. None of the items were loadedto more than one factor either
(Hair et al., 1998).
Note 2: Principal componentanalysisand orthogonalvarimax rotation were used.
On the basis of the EFA results, the factors for unplanned (uncontrolled)
communicationelementswere describedand namedasthe following:
Factor 2-
174
Factor
Loadings
0.886
0.932
0.904
0.840
0.859
0.808
Cronbach
Alpha
0.922
0.822
0.872
0.871
0.811
0.917
company
0.831
It seemsthat all communicationactivities of the companyare plannedand
executedby the samepersonor people
0.870
communicationactivities of the companyaim to convey a commonmessage
_All
Kaiser-Mayer-Olkinmeasureof samplingadequacy= 0.835
Total varianceexplained= 78.28%
Test of Sphericity--3259.93,p= 0.000
_Bartlett
Note 1: All items were retained.There were no items with communalitiesless than 0.50 and with
factor loadingsless than 0.50. None of the items were loadedto more than one factor either
(Hair et al., 1998).
Note 2: Principal componentanalysisand orthogonalvarimax rotation were used.
175
Factor 3-
5.4.2. Confirmatory
modelling which "allows testsof the significanceof all pattem coefficients ... [and]
provides a particularly useful framework for formal comparisons of the
176
interest
of
model
with next most likely theoreticalalternatives"(p.422).
substantive
In this method, first the validity of the constructsis tested by confirmatory factor
analysis (CFA) i.e. measurementmodel assessment.Second, the relationships
betweenthose constructs(structural equation model) are examined.In this section
the measurementmodels for the corporate identity mix elements (companycontrolled communicationelements),the unplanned(uncontrolled communication)
elementsandthe dependentantecedentand control variablesare discussed.
177
The maximum likelihood (ML) estimation method was used in all measurement
model estimationsby CFA. Even though ideally the asymptotic distribution free
estimation methods should be used when the normality assumption of the
is
analysis
violated (SeeSection5.3 for the test of normality), ML could
multivariate
still be applicableas an estimationmethodwhen the samplesize doesnot meet Hair
et al.'s (1998) criterion of having at least five observationsfor each variable
(Anderson and Gerbing, 1988; Bentler and Chou, 1987).The model fit indicators
were utilised in model validation in order to solve the likely problem of an unreliable
due
(Chi-square)
to ML application (Bentler and
standard
statistic
and
errors
X2
Chou, 1987).Moreover,the covariancematrix was usedin all the CFA testssince it
yields more reliable values for the X2statistic and measurementerrors comparedto
the correlation matrix Q6reskog and S6rbom, 1993). All the parametersof the
indicatorsandthe latentvariableswere estimatedfreely.
In the first run of the CFA for the corporate identity mix elements (companycontrolled communicationelements),the model fit measuresdid not indicatea valid
four factor solution. The RMSEA (root mean squareerror of approximation)was
above the acceptablelevel of 0.08 (Garver and Mentzer, 1999) and the X2 statistic
was statistically significant (p < 0.05) (Andersonand Gerbing, 1982).Andersonand
Gerbing (1988) suggestthat in order to improve the model fit, the problematic
indicators can be treated in the following ways: 1) removing items from further
analysis,2) relating itemsto a different factor, 3) relating the itemsto more than one
dimension, and 4) using correlatedmeasurementerrors. The first two options are
recommendedover the two latter ones,since it is more likely that unidimensionality
178
Accordingly, the items of the corporate visual identity systems (CVIS) and the
corporatedesign (CD) constructswere merged becausethese two dimensionsare
theoreticallycloserto eachother than the other two factors (companybehaviourand
marketing communicationmix). Moreover, the items which required modification
belongedto the corporatedesignfactor. The three factor measurementmodel did not
show a good fit, either. Consequently,the four problematic items which were
initially included in CD were iteratively excludedfrom further analysisand the 7.2
differencesbetweenthe previousand the current modelswere computedin eachrun
between
found
difference
the measurement
was
until no statistically significant
1991).
1988;
Steenkamp
Trijp,
Exclusion
Gerbing,
(Anderson
of all
and
and
models
the items initially defined as relating to the staff apparel(EAPP) and the corporate
In
be
the
(AEST)
theoretically
to
acceptable.
considered
constructs
was
aesthetics
design
did
focus
the
the
of
not mention
groups
preliminary research,membersof
staff apparel before the researcherasked them to elaborateon what they thought
front
line
the
the
employeesof the organisationsthat they chose
of
appearance
about
to talk about. Similarly, the discussionabout the exterior and the interior designof
the companies' buildings was initiated by the researchernot by the focus group
attendants.This was consideredas an indication that while consumersare evaluating
the
image
to
tend
those
they
as
not
consider
elements
salient
as
a company's
items
identity
Therefore,
those
and the relatedconstructs
systems.
corporatevisual
were excludedfrom further analysis.
179
Additionally, three items were excluded from the corporatevisual identity systems
items
from
behaviour
(CBEH)
factor
(CVIS),
two
the
company
and one
construct
item from the marketing communicationmix elements(COMM) dimensiondue to
high modification values.The deleteditems under corporatevisual identity systems
are as follows: 1) 1 like the nameof the company,2) 1 like the logo of the company
and, 3) 1 like the colour and typographyusedon all visual materialsof the company.
Discardingthose items was consideredtheoretically acceptablesince the remaining
four items still coveredthe theoretically defined items of name, logo, slogan and
colour andtypography.
Lastly, the following item from the marketing communication mix construct
(COMM) was deleted:"The public relations activities of the companyto promote
itself as well as its products reflect the image defined in the paragraphabove".
During the pilot testingof the questionnaire,it was noticedthat the respondentswere
between
difference
the public relationsactivities and sponsorship
the
not clear about
180
activities. Sinceit was not diagnosedon the basisof the exploratory factor analysis
data
to
test
the
pilot
as well as the main survey data,the item was
applied
which was
factor
higher
However,
the
confirmatory
assessment
analysis
showed
a
still reserved.
covariancebetweenthat and the remaining items of the construct. Therefore,this
item was excluded.
The figures in Table 5.8 are evidencethat convergentvalidity was achieved.All the t
values of the items were significantly greaterthan the critical value of 1.96 at the
0.95 confidence level (Bagozzi et al., 1991; Chau, 1997). All indicators showed
higher individual reliabilities (SMC) than 0.50. The three constructs had high
for
(p
0.70)
the
>
each
and
average
variance
extracted
composite reliabilities
23Since the X2 statistic is sensitive to sample size, this was consideredan acceptablecondition
(Bagozzi et al., 1991;Hair et al., 1998).Thus the measurementmodel was assessed
on the basisof
the other fit indices.
181
182
intermediary
(CWOM),
communication (EWOM) and consumercommunication
company identification (IDNT) constructs were revised, whereas the emotional
appeal (EMOT) dimensionremainedthe same.The following stepswere taken to
model structurein Table 5.9:
reachthe measurement
CFI
0.99
SNIC
AGFI
0.92
t- value
0.73
0.83
0.73
22.52
22.76
21.90
0.77
23.59
0.77
20.75
0.76
22.58
0.82
0.85
0.50
16.22
19.01
13.13
23.94
0.71
26.95
0.73
32.41
0.87
36.90
0.91
26.30
0.84
Average Variance
Composite Reliability (p) Cronbach Alpha (ct)
Internal Consistency
Extracted (AVE)
0.763
0.906
InterpersonalCommunication
0.906
0.767
0.908
IntermediaryCommunication
0.906
0.722
Emotional appeal
0.885
0.877
0.960
0.828
Consumer-company
Identification
0.956
0 X'-Chi square;df - degreesof freedom;RMSEA - Root meansquareerror of approximation;GFI - Goodness-offit index; NFI - Normated-fit index; CH - Comparative-fitindex; AGFI - Adjusted goodness-of-fit index; SMC Squaredmultiple coffelation(Variable's own individual reliability)
Nomological validity: Satisfied.RMSEA < 0.80; NFI, NNFI andCH -> 0.95; GFI andAGFI ->0.90
Convergent validity: Satisfied.All t-values-> 1.96(significant at 0.95 confidencelevel); All SMC andAVE ->
0.50, All p>0.70
First, even though the exploratory factor analysis for the unplanned(uncontrolled)
communication elements suggested merging the items for the interpersonal
183
Second,the item "When I talk aboutthis companyI say"we" insteadof "they ..was
deletedfrom the consumer-companyidentification construct(IDNT), since its error
term was correlated highly with those of the other items. This exclusion did not
jeopardise the theoretical essenceof the construct (Anderson and Gerbing, 1988),
sincefive of the manifestvariableswere still preserved.
184
Finally, the corporate image (IMAG), the perceived integration (INTG) and the
company knowledge (KNOW) constructs' validities were assessedby CFA. The
goodness-of-fitindices evidencedthat the measurementmodel was valid, thereby
confirming nomological validity (Lages, 2000; Steenkampand Trijp, 1991) (Table
5.10).The RMSEA measurewas 0.054,which is within the acceptablerangeof 0.05
and 0.08 (Garver and Mentzer, 1999), and NFI, NNFI, CFI were above the 0.95
threshold value and GFI and AGFI were above the 0.90 critical value (Bagozzi et al.,
1991; Doll et al., 1994; Forriell and Larcker, 1981; Garver and Mentzer, 1999; Hair
et al., 1998; Mueller, 1996)25 (Table 5.10). All of the items for the company
knowledge (KNOW) and the corporate image (IMAG) constructs were retained.
However, the indicator - "All communication activities of the company have similar
its
deleted
integration
from
(INTG)
the
objectives" - was
perceived
construct
since
error term correlated highly with the other manifest variables' error terms and the
modification indices suggestedre-specification of the factor (Anderson and Gerbing,
2SSimilarto theformermeasurement
thex wasnot considered
model'sfit assessment,
asa modelfit
indicatorsinceit statisticallysuggested
a worsefit. It was concludedthat this result may have
occurreddueto the sensitivityof the X2statisticto samplesize(Bagozziet al., 1991;Hair et al.,
1998).
185
The result of the CFA application showedthat the convergentvalidity was satisfied.
All t valuesof the manifestvariableswere higher than the critical value of 1.96at the
0.95 confidence level (Bagozzi et al., 1991; Chau, 1997). Almost all indicators'
individual reliabilities (SMQ were abovethe thresholdvalue of 0.50 exceptthe last
item underthe corporateimageconstruct(Fomell and Larcker, 1981).Sincethis last
item's SMC value was close to 0.50 (the value= 0.44), the retentionof the item was
consideredacceptable.All of the constructsshowedhigh compositereliabilities (All
p>0.70) and the averagevarianceextractedfor eachfactor was above0.50 (Fornell
186
and Larcker, 1981). Additionally, the Cronbach alphas for the three underlying
dimensions were 0.922,0.822, and 0.896,. respectively, thereby exceeding the
thresholdvalueof 0.70 (Nunnally, 1978)(Table 5.10).
00
.6
W)
'
Tw
.e
; -
H a 9
0 ;c:) :0
Vi I
%
tt
A
0,
%D 0,
,,6
zo
(4
"
(4
VII
tn
V)
"S
.r
9
M,d, d,
d
S
-S
'o 0 -06
8
3d Md
%0 "i
VII CIS
Wi
el
ON
M
`0
Oi
a,.
40
:o
o
1
1
HH
0
0
01
l
0
u5
o
u6
G
w
1
U
U0
'0
U
G5
:
a:
0
8
o
0
--p:
-)
U
o. , o 0. " . , . '
-. " ) -1
u
-. o
: o : o : o : o : o : o
,u
0:
40
: o ; C)
0
U-Md
00 ON00 ONZ, ,, M=
-4
Tu
zo
r-
(7, t 'o, M :!
00
'-1
-. ? -0
.,
v2
GM
0
10
C4 eq
o o o o
o
en
0: 0
: o :
U: w
1=41
' 1
06 F-
:E
Gn
rA
co
0 : 0
: o -:: o
90
CD
9:9
I'li
10
0 > 'A l o 01
M U5 u 'MUYOYU: V:
Q
Y
Rt
:
R U, ::0 :Q 0--O: Ot
:p
-: 40
uc,::) U,:Sd 3 U,
m
t-
F.
gI
0
!2
00
11 '
--
JA
zS
00 Ch 00 CS
0% 1- 00 %n
-en
n
2
00 (7,
00 (7,
"4 --
en
c!
-,
N6
00
,
P.,
r.-
o : o
w w5 1
W
44
S S b
u
Ici
r_
ce
"Ci
t-
00
10
eq
oo
00
t10
N
No
oo ON
o
eq
d:
%6
C4
.
0
CIO,
A
t-
C5 u :Su
o
L) z- =1
14
00
M
00
-
en
t-
5
0
00
,4
P
'
R00'- --
r'r
I
zR
,
Go
.2-i- _1 -,
%0
0
0
: 0
o
o
o
0
a
:
:
0 40
0
0 b
>
> >
uo
>
U
5
P
C U5u:>u
)
u
L
u
U,
u
,4
u
u
m
u
U
"E
u
U,
u.
uW
"
., u >u
-.
.u
,6
g>u
gu
g .9.
-ut
u
1
.u
- i -E
m
3
%,
iz
.-
35
:Sd u
B B
2
r,
5
d
U2 .,
'?
'o,
.
.9g
4)
tj .
'o, .
I
3dI
:Su1
:SU,:5 (5 :5
During the analysis, no constraints were imposed to almost all of the parameters
relating to manifest and latent variables except employee behaviour (EBEH),
consumer-company value congruence (CNGR) and corporate associations gap
(IGAP). Since these the first concept was measured with only one indicator and the
two latter concepts were calculated as difference scores of aggregated values (See
Section 4.5.4), the link between the manifest and the latent variable is set to unity and
the measurementerror of the indicator is set to zero (Bagozzi, 1980b; Diamantopoulos
and Siguaw, 2000; J6reskog and S6rbom, 1993). The covariance matrix was preferred
to the correlation matrix due to the fact that the latter causesproblems in terms of the
X2 (chi-square) statistic and measurementerror computation (J6reskog and S6rbom,
1993).
Before analysingthe structurallinks, the overall fit of the modelto the observeddata
was examinedin order to assesswhetherthe model was valid. Table 5.12 represents
the figures for the goodness-of-fltindices.Although the X2 value was.statistically
.
significant (X2= 1187.44,dF--582) at a 0.000 significancelevel, thereby indicating
189
Modelfit: RMSEA
< 0.05;NFI,NNFIandCH -> 0.95,GFIandAGHwithintheacceptable
of 0.80and0.90.
range
The research hypotheseswere tested on the basis of the structural model above (Figure
5.1). An examination of the path estimates and t values in Figure 5.1 illustrates that
twelve of the paths had statistically significant coefficients. Nine of the t values were
above the 1.96 critical value at the 0.05 significance level. The t values for the
relationships between company behaviour (CBEH) and corporate image (IMAG),
26As mentionedin Section5.3.2.,the X2statisticwas de-cmphasizedin the assessment
of model fit.
Sinceit is sensitiveto samplesize(Bagozziet al., 1991;Hair et al., 1998),the other fit indiccgwere
usedto assessthe validity of the overall model.
190
0.03 (0.73/H3brejected
R2- 0.64
zz:
0.08 (1.49)*/112suppbrted
0.22 (2.92)/H4supported
A
nf%
V. V7
0.06 (-0.85)/H5rejected
11% Al%
116bsupported
R- 0.09
EWOM
0.24(5.33)
H6g supported
EMOT
(4.51)
0.21
.
116asupported
IDNT
R2- 0.52
(-4.75)/li6h
supported
-0.10
)7z
INTG
(-2.43)
0.15
-!
116csupported
(4.75)
-0.26
H6j supported
0.62 (13.19)/1-16i
supportcd
CNGR
191
Regarding the direct relationships, it was found that only three of the corporate
identity mix (company-controlledcommunication) elements, i.e. corporate visual
identity systems(CVIS), companybehaviour(CBEH) and marketing communication
I
mix elements (COMM), had a statistically significant impact on corporate image
(IMAG). CVIS was significant at the 0.95 confidence level (tcvls= 2.96 > 1.96) (Hla
accepted), whereas CBEH and COMM were significant at the 0.90 confidence level
(tCBEI-17
1.66 > 1.283 and tcomm= 1.49 > 1.283) (142and H3c accepted). The influence
of employee behaviour (EBEH) on corporate image (IMAG)
(tEBEH"'
0.73 < 1.96) (H3b rejected). Among the corporate identity mix elements, the
CVIS construct had the highest impact on the corporate image construct (yCVIS-IMM
0.22) (Figure 5.1). That is, ceteris paribus a one unit increase in CVIS resulted in a
0.22 increase in IMAG. The two coefficients for the company behaviour (CBEM and
the marketing communication mix (COMM)
0.08,
0.10
and
constructs were
respectively.
192
(IDNT),
(CNGR), corporate associations gap (IGAP), emotional appeal (EMOT) and corporate
image (IMAG) were confirmed. The t values of the paths between the first four
constructs and IMAG were 2.92,2.43,
statistically higher than the critical t value of 1.96 at the 0.05 significance level (H4,
H6b, Me, H6a accepted). The relationship between EMOT and IMAG was significant
1.87 > 1.283) (H6c accepted). However, no
at the 0.90 confidence l.evel OEMOT-IMAG-_
direct causal link was found between intermediary communication (EWOM) and
(H5
1.96)
<
corporate image (IMAG) OEWOWIMAG
rejected).
-0.85
193
(YEMOT-MNT'--
consumcr-companyvaluc congruencc(CNGR)
(ycNOR-IDNT-"2
0.24,
'0-109
tEMOT-IDNT
tCNGR-IDNf'--
5.33),
-4.75)
identification
(IDNT) demonstratedthat the two former
and consumer-company
dimensionspredictedthe latter factor. As a result the hypothesesH6g and H6h were
is,
Ilat
when consumershavea favourableemotionaldispositiontowardsa
supported.
companyand when they think that a company's valuesare congruentwith their own
values,they are more likely to identify themselveswith that organisation.
194
Result
Supported
Supported
Supported
Not
supported
Supported
Not
supported
Supported
Supported
Supported
Supported
Supported
Supported
Supported
Supported
Supported
Supported
Supported
Not
supported
(YEMOT-H)NTVIDNT-IMAG-
0.02,
tEMOT-IMAG=
Similarly, it was found that there was a statistically significant indirect relationship
195
196
the direct
links
between emotional
congruence, perceived integration and corporate image were excluded by setting the
gamma coefficients to zero for those relationships (Le model 1). Second, the gamma
direct
for
relationships were estimated freely (i. e. model 2) (See Figure
coefficients
5.1 for the illustration of direct links). The improvement in model fit was assessedby
the chi-square difference test (9).
the two models was computed and the result was compared with the chi-square value
2
freedom
(X
degree
The model with smaller chi-square was
of
43.841).
with one
critical-,
considered a better model (Diamantopoulos and Siguaw, 2000). The D2EMOT-IDNT-m4AG
(X2model
I-
X2mode12
=100.41 ,. X2critica1=1841)for testing the mediating role of
consumer-company identification
(IDNT)
partially
and IMAG.
(X2
Similarly, the D2CNGR-IGAP-U,
=13.55 >X2 ritic.1=1841) and the
4AG modell-X2 mode]2
(X2
1.37
demonstrated
D2CNGR-IGAP-IMAG
that
the
>X2
X2
=1
modell_ mode12
critica1=1841)
between
image
as
consumcr-company
relationships
value congruence and corporate
well as between perceived integration and corporate image are partially mediated by
corporate associations gap. These results confirmed that the incremental impact of
emotional appeal, consumer-company value congruence and perceived integration on
27
image
corporate
were valid
27All t valuesfor the pathsin all of the constrainedand unconstrainedmodelswere higherthan critical t
value of 1.96.All absoluteand incrementalgoodnessof fit indicators(i. e. CH, GFI, NFI, NNFI and
AGFI) showeda good fit (the valueswere above0.95) for all the modelstested,exceptthe RMSEA
valueswhich were much higherthanthe thresholdvalue of 0.08 for the unconstrainedmodels.
197
198
5.5. Summary
In this chapter, the findings of the study were presented.Firstly, the initial data
examination in order to prepare the data for further analysis was explained. The
characteristicsof the two respondentgroups were then illustrated. This section was
followed by the explanationof the exploratoryfactor analysesresultswhich helpedto
refine the scales for the confirmatory stage. The measurementmodels and the
structural model were assessedon the basis of 439 cases.The findings showed
statistically significant relationshipsbetweencorporateimage (IMAG) and almost all
forms of companycontrolled communicationelements(i. e. corporatevisual identity
systems - CVIS; company behaviour - CBEH; marketing communication mix
elements- COMM) exceptemployeebehaviour(EBEH). Except for the link between
intermediary communication (EWOM) and corporate image (IMAG), all the
200
201
VI DISCUSSION
6.1. Introduction
This chapter discussesthe results of the data analysis presentedin ChapterV with
supportfrom the theory presentedin the literaturereview and the information obtained
from the exploratoryinterviewsand focus group discussions(SeeTable 4.1 and Table
202
Since the scales for the sub-elementsof corporate identity mix elements were
developedprimarily on the basis of conceptualarticles, their operationalisationand
validation are presentedfirst. As explainedin ChapterIV, the item pool for scaleswas
subjectedto qualitative and quantitative refinement. Content validity of scaleswas
assessedby academics,and consequently,some items were excludedon the basisof
the information obtainedfrom the interviewsand focusgroup discussions.In addition,
the developedscaleswere testedby statisticaldatareductiontechniquesin Pilot Study
I (EFA) and in the main data analysis(EFA and CFA). As a result theoreticallyand
operationally valid and reliable scaleswere developedand hypothesistesting was
performed with the scales listed in Tables 5.8,5.9 and 5.10. In the following
paragraphs,some inferences are made on the basis of scale development and
refinementissues.
203
perceptionsof companyvalues, have shown that different traits may have different
meaningsfor different cultures and even sometimesexact translation of somewords
cannotbe achieved.This may raiseconcernsin using the samepersonalitytraits when
describingcompanies' identities and individuals' personalvalues. For example,the
researcherswho translatedthe questionnaireinto Turkish pointedout that the meaning
of 'cooperative' in Turkish suggestsa negativepersonalquality, in that a cooperative
individual is consideredto be disloyal to his close friends by sharing information
life
their
about
personal with third parties. Therefore,in line with Ekinci and Riley's
(2003) discussionaboutthe applicability of one set of adjectivesto both companyand
in different settings,it is arguedthat the developmentof
personalvalue assessments
anonymousscalesfor measuring*corporateassociationsgap and consumer-company
value congruenceis hard to achieve.However, observationof respondentsduring the
pilot studieshas shown that the identity traits pool was adequatefor this study, since
none of the individuals commentedon the irrelevanceof the adjectivesin terms of
their ability to describe both companiesand their personalities.Besides,concerns
aboutthe adequacyof the adjectivesin termsof their meaningsfor a Turkish audience,
as well as in terms of their relevance to both company and individual level
were addressedduring the qualitativeassessment.
assessments,
204
,6... we have a slogan 'Life is Ours, Life is Work'. Everybody [in the
company]shouldunderstandwhy we say 'for a better life, so that they
can give this messagewhen they are interactingwith others outsidethe
organisation.The managementof our organisationworks for achieving
that throughour department[PR department]...."
[CorporateCommunicationsManager,Pfizer]
The managersinterviewed also assertedthat only in special cases are their top
managers publicly seen. Otherwise, they are mostly involved in managing
relationships with governments and investors. This is evident in the following
statementby a manager:
205
believe that
his
This situation as shown in the extract above was also supportedby the information
obtainedfrom the focus group discussions.None of the discussantswere able to recall
seeing a global company's managerin public. As a result of this evidence,it was
concluded that manager behaviour is not a part of the behaviour aspect of the
corporateidentity mix elementsin the consumercontext. Moreover, since managers
are not seenpublicly, it is not possible for consumersto attribute anything to their
clothing and its relation to their overall judgement of a company.Accordingly, items
related to managers' physical appearancewere excluded from the staff apparel
concept.
206
in
the
is
related to
207
Another point is also inferred from the discussionswith focus group members.The
following statementshowsthat consumersdo not make immediateinferencesabout a
its
its
buildings,
image
the
or
while evaluating
appearanceof
company's overall
employees'physical appearance.Rather, they relate their views on buildings and on
staff apparels to their purchase experiences.The following quotations give some
supportinginsightsto this interpretation:
208
"... it is nice to seenice looking staff around [in Zara stores] it gives
...
you a pleasantfeeling aboutthe atmosphere..."
[A femaleattendant,FocusGroup #2]
Two main points can be drawn from these extracts. First, different stakeholdersin
different countries can base their views about a company's image on different
indicators.This inferenceis similar to the findings of reputationmanagementstudies
(Fombrun et al., 2000; Groenland, 2002; Thevissen, 2002). For example, while
Thevissen (2002) demonstratedthat Belgian consumers put more emphasis on
behaviour,communicationand personalfactorswhen they are evaluatinga company's
reputation, Brown and Perry (1994) argued that American businessanalysts and
Secondly,
financial
indicators.
to
attention
executives pay more
performance
be
factors
cannot
consumers' views about some company-driven communication
directly related to the corporate image concept, but to other concepts like store
atmosphereor store imageas it appearsto be in the quotationspresentedabove.The
209
210
internally
in their theoretically defined factorS29
homogenousand
that
they
are
and
externallydistinct from eachother (SeeSection5.4.1 and Section5.4.2). For example,
the corporateimageconstructwas confirmed as being a three-itemmeasureas defined
by Williams and Moffit (1997). Similarly, the consumer-companyidentification scale
resultedin one factor. Although the number of items in the scale was reduced,the
defined
items
by Ashfort and
the
the
still
captured
essence
concept
as
remaining
of
Mael (1989).
The objective of this study was to draw upon consumers'perspectivesto explore the
types of communicationthat determinecorporateimage formation. One of the two
29The items related to corporateaestheticsand staff apparel constructswere merged together and
redefinedas corporatedesign(SeeTable 5.5). However,this constructwas not found statisticallyvalid
by the CFA application. Although the items for interpersonal and intermediary communication
by CFA demonstratedthat
constructsappearedto composeone single factor, more stringentassessment
thesetwo were separateconcepts. -
211
the
basis
literature
is
the
the
of
communicators
groups
elicited
of
major
on
review
corporateidentity mix elements(i.e. symbolism,communicationand behaviour)(Van
Riel, 1995).As suggestedby the first researchquestion(See Section 1.2), the direct
impact of each of the sub-factorsunder symbolism, communicationand behaviour
identity
mix elementson corporateimageare examined.Basedon
aspectsof corporate
Attribution Theory (Kelley and Michela, 1980; Malle, 1999,2003; Weiner, 1974,
1986), it was hypothesisedthat people's attributions of a company's planned
communicationactivities affect their evaluationsof the company as a whole (i.e.
image).
corporate
212
213
214
215
6.4. Determinants
of Corporate
elements
216
information exchangeamong people that are strongly tied to each other (e.g. close
friends andrelatives)evokesattitudinal changetowardscompanies.
217
218
The results also supported the theoretical expectation that individuals' emotional
predisposition towards companies and the fit between organisational values and
consumers'descriptionsof their personalitycharacteristics(consumer-company
value
219
220
221
6.5. Integration
Corporate
of
Identity
Mix
to
Corporate Image
The third researchquestion of this study (See Section 1.2) addressesthe gap in
corporateimage researchrelatedto providing empirical support for the measurement
and relevanceof integratedcommunication in corporate image formation from the
perspectiveof receivers.Many of the conceptualarticles in the field (e.g. Abratt, 1989;
Bernstein, 1984; Einwiller and Will, 2002; Gray and Smeltzer, 1985,1987; Olins,
1978; Van Rekom, 1997; Van Riel, 1995) discussedthe integration issue from the
companies' standpoint and suggestedthat companies should integrate corporate
identity mix elementsto achieve favourable corporate image and similar corporate
associationsas they intendedto createin the minds of consumers.However,they have
failed to provide evidencefor how integration can be measuredin the consumers'
context and how consumers' attributions about the integration level of companydriven communicationefforts (i. e. corporateidentity mix elements)relateto corporate
imageformation.
222
Another issue was highlighted by Brown and Dacin (1997) that managers of
organisationsstill did not know whetherit was their plannedcommunicationefforts or
As
images
hold
defined
factors
their
about
consumers
companies.
what
external
also
Dacin and Brown (2002) suggested,in this respectit was fruitful to investigatethe
identity
impact
these
two
of
groups
communicators(i.e. corporate
major
relative
of
223
on corporateimage.
224
6.7. Summary
In this chapter, the research findings of this study are discussedin relation to
theoretical expectations.The chapter begins with comments on the results of the
measurementscale refinement and hypothesestesting. The outcomesindicated that
some sub-elementsof corporateidentity mix elements(i.e. corporateaesthetics,staff
apparel and managerbehaviour) may be removed from the proposedmodel. The
structuralequationmodelling applicationdemonstratedthat almostall of the proposed
relationshipsbetweenthe antecedentsdefinedand corporateimageexist. However,the
relationshipsbetweenemployeebehaviourand intermediarycommunicationwere not
statisticallysignificant.
225
found
for
However,
the relative
gap.
an
unclear
result
was
corporate associations
impact of eachcorporateidentity mix elementsand unplannedcommunicationfactors
on corporate image. Finally, the argumentthat 'the link betweenthe perceptionof
integrated company-driven communication and corporate image exists via the
corporateassociationsgap' was also confirmed.
226
VII CONCLUSION
7.1. Introduction
227
The major contribution of this research is that it is one of the first attempts to
incorporate corporate identity mix elementsand unplannedcommunicationfactors
into one model, and exarfiine each communicator's impact on corporate image
formation from the consumer'sperspective.Prior studieshave investigatedthe role of
some company-drivencommunicationactivities individually in conveying corporate
identity messagesto different audiences.However, a comprehensiveapproach,which
takes the communicatorsboth internal and externalto companiesinto consideration,
has not beentested.By examiningan integrativemodel in the consumercontext,this
research challenges the claims of 'anything a company does communicates its
identity' (Balmer, 1997,2001; Van Riel and Balmer, 1997)and 'corporate imageis a
compositeproductof multiple communicatorsincluding formal communicationefforts
factors'
(Bhattacharya and Sen, 2003; Cornelissen,
external
communication
and
2000). Additionally, it providesempirical evidenceabout the influence of integration
of company-controlledcommunicationactivities (corporateidentity mix elements)on
corporateimage. Therefore, drawing on the interdisciplinary paradigm (See Section
2.2.5), this study is one of the first empirical works that synthesisesconceptsfrom
corporateidentity and corporateimage managementword-of-mouth communication,
corporateassociations,organisationalidentification, value congruenceand integrated
marketingcommunicationliteraturesto explain the corporateimageformation process
in a more holistic manner.
228
229
implementationplans (e.g. Abratt, 1989; Dowling, 1986; Gray and Smeltzer, 1987).
They acknowledgethat all formal communicationactivities of an organisationshould
its
identity;
however,they stressthat theseefforts should
the
of
possess characteristics
be designedto meet the different expectationsof company'sconstituentgroups,and
specific setsof activities should be emphasisedmore in communicatinga company's
identity to eachtargetgroup of stakeholders.
230
231
This study showed that company behaviour towards social responsibility issues
determine what consumersthink of organisations.This finding is in line with the
researchin corporateassociationsand corporatesocial responsibility (e.g. Brown and
Dacin, 1997; Sen and Bhattacharya,2001) and it highlights that company aspects
which are not product related such as social responsibility efforts may act as a
232
233
resultsof the CFA demonstratedthat the definition of corporateimageas the net result
impression
incorporates
individuals'
their views
about
a
company
of an
overall
which
on the company'scomparativeposition in its sectorand what they construeaboutwhat
othersthink about that organisationis conceptuallyvalid. The hypothesistesting also
confirmed that the corporateassociationsgap was a determinantof corporateimage
formation. Therefore,the theoretical expectationsabout the definitions of corporate
image and the corporateassociationsgap were met. This inferencecomplementsthe
researchin corporateassociations(e.g. Balmer, 1998; Brown and Dacin, 1997) and
corporatebranding(e.g. Daviesand Chun,2002; Schultzand De Chematony,2002).
234
235
236
237
On the basis of the theoretical implications discussedin the previous section, this
study offers some practical guidelines for managersto achievea favourable image
about their organisationsin consumers' minds. This study's findings suggestthat
managersshouldrecognisethat corpprateimageformation is a complexprocessthat is
influenced by
multiple
238
The findings of this study about the relationship betweencompany behaviour and
corporate image suggest that managers should pay more attention to regular
being
is
the
perceived as socially
company
or
not
assessmentsshowing whether
found
(2001)
Sen
Bhattacharya
that consumers' evaluations of
and
responsible.
companiesarc especiallysensitiveto negativeattributions about theseorganisations'
reactionstoward issuesof socialresponsibility.In the light of this study's findings and
239
240
favourable
As
deliver
to
their
close
environments.
recommendations
companiesand
Bowman and Narayandas(2001) assert, each contact initiated by word-of-mouth
information disseminationby company related infortnation seekersis an important
factor for building betterrelationshipswith consumers.
241
The finding that the corporate associationsgap has a positive relationship with
corporate image suggests that communicating corporate identity is linked to
understandinghow specific corporateassociationsare held by different stakeholders,
including consumers,who are the subjectsof this study. Dacin and Brown (2002)
assert that companiesshould not only be aware of how consumersperceive the
specific qualities of their brands;they should also monitor how consumersevaluate
the unique and distinctive characteristicsof their organisations. Brown and Dacin
(1997) demonstratedthat specific corporateassociationshave a positive impact on
product evaluationsdirectly and through overall company evaluation (i.e. corporate
image). It can be arguedthat any mismatch betweenintendedcorporateassociations
and consumers'perceptionof theseassociations(i.e. corporateassociationsgap) may
also lead to the inconsistencybetweenthe specific product qualities and consumers'
in
be
As
them.
a
result,
products
of
organisation
may
misplaced
attributionsabout
an
the minds of consumers,and they may switch to competingbrands.In order to avoid
any marketsharelossdue to suchmechanisms,managersshouldensurethat companydriven communicationactivities reflect what they want their consumersto associate
with their organisations.
242
translatedinto using similar qualities in the creation of new products and services.
Similarly, managerscanusethe samevaluesto position their companiesin the market
place, and then revise the definition of intendedcorporateassociationsfor their own
organisations.If the consumer-companyvalue congruenceassessmentis carried out
periodically by companies,any changesin consumertastescould be detectedand
intended
definition
in
the
of
corporateassociations.By doing so, companies
reflected
can benefit from differentiatingthemselvesfrom their counterpartsat the right time in
the right manner.
243
Sincethis study is one of the first attemptsto addressthe role of corporateidentity mix
elementsand unplannedcommunicationfactors in corporateimage formation, there
are some limitations pertinentto sampling,survey approachand measurementissues
which arepresentedin the following paragraphs.
244
sample of individuals from the clusters selected(See Section 4.6.2). Even though
conveniencesamplescan be consideredappropriatefor theory testing, a probability
samplingtechniqueshould be used in order to eliminate the potential bias in terms of
validity and generalisabilityof the scalesusedin this study(SeeChurchill, 1999).
245
246
The sixth point is related to the collection of the information about the two case
identity
intended
corporate
companies'
associations.As mentioned in the research
design,a questionnaire(SeeAppendix 9) which included the compiled identity traits
was applied to the decision-makerswho are involved in the corporate identity or
corporate communication managementof their organisations.However, since the
distribution of this questionnairewas carried out by the key managerinterviewed in
each casecompany,there may be questionsabout whether the targetedrespondents
evaluatedtheir company's identity or not. Nonetheless,the introductory part of the
questionnaireclearly statedthat the questionnaireshould be distributed to managers
who are involved in the decision-makingprocessof corporateidentity management
and communication.The importanceof this issuewas also emphasisedin the followup emailssentto the key interviewees.
247
Finally, it should be mentionedthat the resultsmay show a potential bias, due to the
usageof difference scoresto measurethe corporateassociationsgap and consumercompanyvalue congruenceconcepts(See Peter et al., 1993).Despite the criticisms
about the reliability of difference score measuresand too high correlationsbetween
gap scores and theoretically related variables, the mathematicalcalculation of gap
scoreswas consideredas usablein this study, as the debateon the superiorityof direct
measurementof gap scoresis still unresolved(Ekinci and Riley, 2003).
Although there are some limitations pertinentto this study's findings, the researcher
believes that the potential bias in results can be overcome by conducting further
studies.In the next section,some future researchavenuesare suggestedto point out
the limitations discussedhere along with some further areasof investigationrelevant
to corporateimageformation and its consequences.
248
This study has grounded the argument about the impact of company-driven
communication on corporate image in Attribution Theory, which argues that
observers' (consumers)interpretationsof others' intentional behaviours(companies'
plannedcommunicationactivities) affect their reactionsto other people(Malle, 1999;
Malle, 2000). This research addressesconsumers' attributions of a company's
communicationof its identity by corporateidentity mix elements;however,it doesnot
take into accounthow the consumers'interpretationprocessof corporateidentity cues
occurs;this could be examinedin a further study.
249
250
251
7.6. Summary
252
corporatesocial responsibility,word-of-mouthcommunication,corporateassociations,
organisationalidentification,self-image,value congruenceand servicesmarketing.
253
Will you please give your opinion for the following definition of corporate
identity, which is done by Cees Van Riel who is an academic and expert in the
area? Please indicate your level of agreement and why or why not?
is the strategically planned expressions of selfpresentation of an organisation via the cues given by its symbols, behaviour and
communication"
5. According
to your experience what do global companies understand from
corporate identity, corporate identity management and corporate communication?
_
6. What could be the possible forms or formats of corporate communication to
communicate the identity of a global company?
What do you think constitutes symbolism, behaviour and communication? How
identity
activities
of
your
company's
communication
would you classify your
under these headings?
Definition:
"Corporate
identity
254
estions
by
symbolism,
conveyed
8. How much are the consumersaware of corporatecues
behaviour,communicationand corporatebrandstructureof global companies?
9. How do consumerselaborateof the information abouta global company?
10.Which channelsand tools of communicationhave more priority for consumersin
buildina?
imaae
r-nmorate
I would like to thank you againfor your kind cooperationandvaluabletime.
255
256
Questions
Source
Questions
Source
19.What do you think abouthow much the other factorsi.e. wordof-mouth, the information retained in the memory and
intermediaries have impact on the effectiveness of your The researcher
company's identity communication?How do they interplay
with your company'scorporatecommunication?
Abratt
and
20. Is thereany imagetracking mechanismin place?
Mofokeng, 2001
21. Has the companybeen involved iq any image-repairexercise Stuart, 1995
in the last five years?If yes,how was it handled?
Van Riel, 1995
22. What do your surveyssearchfor? What kind of information
The
researcher
for
by
looking
thesesurveys?
are you
23. (After a summaryof the interview) Is this summarysufficient
important
the
to
out
point
and salient issues?Do you The researcher
enough
want to add something,which you believewe didn't consider? I
I would like to thank you againfor your kind cooperationand valuabletime.
258
259
260
Appendix 4- Pilot Study 1 respondents' demographics for the three drafts of the
McP2
(McPl,
and McP3) for the McDonalds referencecompany.
questionnaire
mcpl
Female
Gender - Age/Year/Social Status
Male
Total
18-20
22
16
38
Age
21-23
35
24
59
46
51
Total
97
21
Second year
23
44
Year
8
Third year
15
7
Fourth year
20
14
34
Master
2
4
2
Total
51
46
97
25
23
A Social Eoup
48
Social status
13
28
B Social group
15
6
7
13
ocia group
6
2
8
C2 Social group
Total
51
97
46
mcpilot 2
21-26
46
37
83
Age
27-31
6
19
25
65
Total*
43
108
66
110
Master (t tal)
44
Year
A Social group
31
41
72
Social status
20
B Social group
7
13
9
CI Social group
4
5
C2 Social group
2
5
7
64
Total*
44
108
McPilot 3
43
15
28
18-20
Age
27
42
15
21-24
42
85
Total**
43
11
8
First year
3
Year
36
Second year
25
1
11
20
Third year
9
6
12
18
Fourth year
85
Total**
43
42
34
19
15
A Social group
Social status
20
10
10
B Social group
14
CI Social group
7
C2 Social group
is
6
9
Total***
41
83
42
Two caseswere missing.
One case was missing.
*** Three caseswere missing.
Note: McP I: McDonald's questionnaire draft 1; McP2: McDonald's questionnaire draft 2; McP3:
McDonald's questionnaire draft 3.
261
Appendix 5- Pilot Study 1 respondents' demographics for the three drafts of the
questionnaire (RM1, RM2 and RM3) for the Renault-Mais reference company.
RMP1
Gender - Age/Year/Social Status
18-20
Age
21-23
Total*
Year
Second year
Third year
Fourth year
Total
A Social group
Social status
B Social group
Cl Social group
C2 Social group
Total
RNIP2
21-26
Age
27-31
Total**
Year
Master (t tal)
A Social group
Social status
B Social group
Cl Social group
C2 Social group
Total***
RMP3
18-20
Age
21-24
Total
First year
Year
Second year
Third year
Fourth year
Total
A Social group
Social status
B Social group
CI Social group
C2 Social group
Total
Female
Male
Total
17
11
28
16
3
9
28
13
8
6
4
28
13
29
42
19
4
20
43
18
10
7
7
43
30
40
70
35
7
29
71
31
18
13
11
71
37
8
45
47
31
14
0
1
46
36
15
51
51
36
10
4
- 1
51
73
23
96
98
67
24
4
2
97
29
9
38
1
23
9
5
38
17
13
4
4
38
24
27
51
10
20
17
4
51
16
17
9
9
-51
53
36
89
11
43
26
9
89
33
30
13
Id
3
89
9
262
McPilot Study 2
Gender - Age/Class/SStatus
18-24
Age
25-31
32-38
Total*
Married+partnership
Marital status
Single+widow+divorced
Total*
A Social group
Social status
B+CI Social group
Total**
RMPilot Study
18-31
Age
32-52
Total
Married+partnership
Marital status
Single+widow+divorced
Total
A Social group
Social status
B+CI+C2 Social group
Total
Female
Male
Total
9
18
4
31
11
20
31
17
13 17
30
4
6
4
14
7
7
14
7
14
13
24
8
45
18
27
45
24
20
44
8
5
13
4
9
13
10
3
13
7
10
17
9
8
17
10
7
17
15
15
30
13
17
30
20
10
30
263
264
Yes
11.[low would you describe McDonald's as a company? Please write a couple ofsentences and/or adjectivcs about
what this company reminds you.
111.Pleasestate your general impression about McDonald's by ticking the most appropriate option below.
Very unfavourable
12345
I.Infavourable
Neutral
Faivourable
Very favourable
IV. What do you think about what impression other people have about McDonald's? Please tick the most
appropriate option below.
Very unfavourable
12345
V.
Unfavourable
Neutral
Favourable
Very favourable
Please state your impression about McDonald's compared to other companies in the same sector in(] tick
the most appropriate option below.
Verv unfavourable
12345
Unfavourable
Newral
Favouralple
Very favokil-able
VII. In the section below, there are some adjectives to understand the image you have about McDonald's in )our
describe
McDonald's
how
these
Please
adjectives
and tick the most appropriate option
much
state
mind.
below for each adjective.
Not at all
A little hit
Some
Neither/Nor
Pretty much
Ver.1 much
Completely
I lonest
12
Considerate
12
EffectiN c
Innovative
12
12
6
6
7
7
Sensitive
12
Expert
12
6
6
Trustworthy
12
Caring
12
Exciting
12
Leader
12
Responsible
12
Fun
12
6
6
Competiti%c
12
Young
12
Dynamic
12
Modern
12
Welcoming
12
Successful
12
Progressive
12
Capable
12
Creative
12
265
VIII. The section below is prepared to understand your impression about McDonald's various visual expressions.
Please evaluate McDonald's according to the l'ollowing items by ticking the most appropriate option below for
each statement.
Strougly
Slrougly
disagree
Disagree
Neutral
Agree
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
1
1
1
2
2
2
3
3
3
4
4
5
5
5
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
agree
Don't
know
IX. The section below is prepared to understand your impression abOUtthe 1ppC,
1r,1nCC
OfMCI)()j111d'Sbuil dingS.
Pleaseevaluate McDonald's according to the following items by ticking the most appropriate option hclow for
each statement.
Strongly
disagree
StrouglY
Don't
agree
knoA
Disagree
Neutral
Agree
5
.
X. The section below is prepared to understand your impression about the appearanceo fMcDorwId's stall. ['lease
evaluate McDonald's according to the f'ollowing items by ticking the most appropriate option below l6r cach
statement.
Strongly
disagree
Disagree
Neutral
Agree
Stronglj
agree
5
5
5
266
Don't
know
X1. This section below is prepared to understand your impression about McDonald's attitude and behaviour towards
various social issues. Please evaluate McDonald's according to the lollowing items by ticking tile most appropriate
option below for each statement.
Strongly
disagree
The employees
of the company's
restaurants
such as
Disagree
Neutral
Agree
Sfrongl
agree
Don't
know
12345
12345
12345
12345
12345
12345
12345
XII. In this section below, there is a paragraph which explains %hat image McDonald's wants to convey to tile public.
Please read the paragraph carefully and state your opinion about how much the following communication activities
reflect the defined image in the paragraph. Please tick the most appropriate option bclo\v for cach statement.
McDonald's is the biggest company in the fast-food sector in the world, which offlers healthy fast-l'Oodwith high
standards; provides a clean and hygienic environment and fast and high quality service, has fun, welcoming and
friendly restaurants and cares for the public, especially for the children.
Not at all
reflect
Somewhat
45
.15
45
uot reflect
Neilher/Nor
reflect
Complelely
reflect
Somewhat
reflect
45
.1
-1
Don't
kll()%r
X111.The section below is prepared to understand Oat you receive 1rom external COIIIIII kill iCdt0r,1,11)01.
lt MCDOMIld'S.
Pleasestate your opinion about the f'ollo\\ ing items by tickin g the most appropriate Option bclo\% 1,01cach stilicnicill.
slrongl
disagree
Disagree
Neutral
Agree
St I ou gly
agI cc
I2
I2
I2
I2
I2
2
I2
267
Don't
knu
XIV. In the section below, there are statements to understand how much knowledgeable you are about McDonald's.
Please state your degree of agreement with the flollowing items by ticking the most appropriate option below for
each statement.
I know
the products
and services
ofthis
company
very
well
Strongly
disagree
12345
Disagree
Neutral
Agree
Strongly
Don't
know
agree
12345
12345
XV. In the section below, there are some statements to understand how you feel about McDonald's. Please state your
degree ofagreement with the following items by ticking the most appropriate option below for each statement.
I fccl good
things
about
Strongly
disagree
12345
this company
Disagree
Neutral
Agree
Don't
know
Strongly
agree
12345
12345
12345
12345
12345
12345
12345
12345
XVI. The section below is prepared to understand your impression about what the interaction and relationship level of'
the communication activities of McDonald's with each other arc. Please state your degree of' agrccinent with the
fiollowing items by ticking the most appropriate option below for each statement.
Strongly
disagree
All
activities
communication
of the company
have similar
objectives
All communication activities of the company are aligned to
convey a common message
I receive similar messagesfrom each ofthe communication
activities of'thc company,
It seemsthat all communication activities of the company
are planned and executed by the same person or people
All
a common
XVII.
acti\
communication
company
aim to com c)
Nentral
below,
I lonest
Not at all
A little bit
1234567
12345
1235
12345
123
.15
Considerate
1234567
Eflcctive
1234567
Innovative
1234567
Sensitive
1234567
Expert
1234567
Trustworthy
1234567
Caring
1234567
Exciting
1234567
Leader
1234567
Responsible
1234567
Fun
1234567
Compel, itivC
1234567
Young
1234567
to understand
the personality
traits ol'a
by ticking the most appropriale
option
Some
Neither/Nor
Agree
12345
message
In the section
much
itics ofthe
Disagree
Strongly
agree
1'retty much
Very much
Coniplefel)
268
Don't
know
No I at all
A fillIc bit
Some
Neither/Nor
Prctl) much
N cf.) 1111101
41111pleld)
Dynamic
Modem
Welcoming
12
12
12
3
3
3
4
4
4
5
5
5
6
6
6
7
7
7
Successful
12
llrogrcssi%e
Capable
12
12
3
3
6
6
Creative
12
XV I 11.Your age
.......................
XIX Your gender
11 Female
171Male
11 Single
171Widow
11 Divorced
billion'Fl,
1,5
billion'I'l,
2,5
-
13 Undergraduate
M rvlastcrsand above
XV. Please tick the most appropriate option below that indicates your employment status.
I am currently employed
1:1 Top executive or manager
I am not emoloved
M Student
11 1louse wi Ic
171Retired
0 Worker
13 Civil servant
Craftsman
Other (Please state)
...........................................
XV]. Please state which of the followings you (or your fiamily) own.
C] Apartment
C1 Summer house
Personal computer
11 Lap top
C73I lome theatre
171pet
171Dish washer
171Car
13 Freezer
171DVD/VCD player
11 Microwave oven
171Credit card
Savings account
269
270
11.How would you describe Renault-Mais as a company? Please write a couple ofsentences and/or adjectivcs about
what this company reminds you.
Ill. Please state your general impression about Renault-Mais by ticking the most appropriate option below.
Very unfavourable
12345
Unfavourable
Neutral
Favourable
Very favourable
IV. What do you think about what impression other people have about Renault-Mais? Plcasc tick the most
appropriate option below.
Very unfavourable
12345
V.
Neutral
Unfavourable
Favourable
Very favourable
Please state your impression about Renault-Mais compared to other companies in the same sector and tick (fie
most appropriate option below.
Very unfavourable
Neutral
Unfavourable
2345
Favourable
VI. II ave you or anyone in your farni I owned a RcnauIt car belore and/or do YOU01'allyOIIC in YOUrIIIIIIi I) 11MCa
Renault car now'?
El Yes
11 No
A little bit
Some
Neither/Nor
Pretty inuch
Very much
(1111plefel)
7
12
Considerate
12
EATective
12
Innovative
12
Sensitive
12
Expert
Trustworthy
Caring
12
12
12
I'Aciting
12
Leader
12
ResponsibIc
12
Fun
12
Compctitivc
12
Young
12
Dynamic
12
Modem
12
Welcoming
12
Successful
12
Progressive
12
Capable
12
Creative
12
I lonest
7
7
271
Vill. The section below is prepared to understand your impression about Renault-Mais various visual expressions.
Please evaluate Renault-Mais according to the fiollowing items by ticking the most appropriate option below for
each statement.
Strongly
disagree
Strongly
Disagree
Neutral
Agree
2
2
3
3
4
4
5
5
2
2
3
3
4
4
5
5
2
2
3
3
4
4
5
5
agree
Don't
know
IX. The section below is prepared to understand your impression about the appearanceof'Rcnault-N/lais buildings. Please
evaluate Renault-Mais according to the following items by ticking the most appropriate option below l6r cacti
statement.
Strongly
disagree
Disagree
Neutral
Agree
Strongly
agree
Don't
knoA
345
12345
12345
12345
X. The section below is prepared to understand your impression about the appearance of Renault-Mais staff. Please
evaluate Renault-Mais according to the fiollowing items by ticking the most appropriate option hclm I*or each
statement.
Agree
SIrongI N'
agree
123
123
Strongly
disagree
Disagree
23
Neutral
272
Don't
knim
X1. This section below is prepared to understand your impression about Renault-Mais attitude and behaviour towards
various social issues. Please evaluate Renault-Mais according to the f'ollowing items by ticking the most
appropriate option below for each statement.
Strongly
disagree
The employees
of the company
such as salesmen,
after
Disagree
Neutral
Agree
Strongly
agree
Don't
know
12345
12345
12345
12345
12345
12345
12345
XII. In this section below, there is a paragraph which explains what image Rcnault-Mais wants to CoIlVey to the JIL)b]iC.
Please read the paragraph carefully and state your opinion about how much the l'ollowing communication acti ities
reflect the defined image in the paragraph. Please tick the most appropriate option below lor cach statcnictit.
Renault-Mais is the leader company in Turkish automobile manullacturing sector, which produces visual. bold,
daring, warm, comfortable, safe and economical cars which are innovative in design. provides good after sales
service; cares for environmental issues and aims to contribute to the public and theTurkish economy.
Not at all
renect
Somewhat
not renect
34
34
34
34
Neither/Nor
reflec(
Solnewhat
I-VnCCI
Coloplefel)
rellec(
Don't
know
The section below is prepared to understand what you receive from external communicators about Rcnault-MaisPlease state your opinion about the fiollowing items by ticki ng the most appropriate option below I*or each statement.
Slrongl)
disagree
My close friends and relatives talk about this compan)
frequently
I hear positive things about this company from my close
friends and relatives
N/1)close friends and rclati%cs recommend this cornpan to
me
My close friends and relatives encourage me to make
purchases from this company
The media. opinion leaders. go\ ernment institutions and
NGOs etc. talk about this cornpan frcquentl
I hear positive things about this company from the media,
opinion leaders, government institutions and NGOs etc.
The inedia. opinion leaders. w\ ernment institutions and
N(; Os etc. arc proud ofthis Lompan)
The media, opinion leaders, government institutions and
NGOs etc. recommend this company
Disagree
234
23
Nentral
Agree
Strongb
Don't
agree
kno'A
5
5
23
23
23
23
23
'173
XIV. In the section below, there are statements to understand how much knowledgeable you are about Rcnault-Mais.
Please state your degree of agreement with the following items by ticking the most appropriate option below for
each statement.
Strongly
disagree
Disagree
Neutral
3
3
3
12
12
12
Agree
Strongly
Don't
agree
knoA
5
45
45
XV. In the section below, there are some statements to understand how you 1'celabout Renault-Mais, Please statc your
degree ofagreement with the following items by ticking the most appropriate option below f'or cach statcnient.
Strongh,
disagree
Disagree
Strongly
Neutral
Agree
12
12
12
3
3
3
4
4
4
5
5
5
12
12
12
12
12
12
agree
Don't
knim
5
4
3
3
-1
4
5
5
XVI. The section below is prepared to understand your impression about what the int eraction and relationship level of'
the communication activities of'Renault-Mais with each other are. Please state y our degree o fagi-ccin crit \%illi tile
following items by ticking the most appropriate option below l6r each statement.
Strongl I,
*
tlisagree
Disagree
Sirongly
agree
Neutral
Agree
12
12
12
12
12
XVII. In the section below, there are some adjectives to understand the personality traits ol'a person. Pleasc state how
much these adjectives describe your personality by ticking the most appropriate o ption below for each ;idjectivc.
Not at all
A little bit
Sonic
Neither/Nor
l1resty mucli
C omplefeki
I lonest
12345
Considerate
F.T1'ective
12345
1234S
Innovative
12345
Sensitive
12345
Expert
12345
Trustworthy
12345
Caring
12345
Exciting
12345
Leader
12345
Responsible
1234S
Fun
12345
74
Don't
knoA
A little bit
Not at all
Some
Neither/Nor
Pretty much
Very much
Completely
12
Young
Dynamic
Modem
12
12
12
Welcoming
12
Successful
12
Progressive
12
Capable
12
Creative
12
Competitive
13 Male
11 Living Nvithpartner
173Single
13 Widow
11 Divorccd
-I
billion'll
billion
TL
1.5
-
1,501 billion'll,
billion'll,
2
--
2,501 billion'll,
3 billion'll
1,001 billionTL
01 ligh School
11 Undcrgraduate
C, mastcrsand
above
XV. Pleasetick the most appropriate option below that indicates your employment status.
I am currently employed
11 Top executive or manager
13 Owner ot'a large or medium size company
Cl Lawyer, dentist, architect etc.
171Office/Clerical staff
11 Worker
171Civil servant
El Craftsman
11 Other (Please state)
...........................................
I am
empluccl
-no-I
171Student
1:1 1louse wi Ic
171Retired
XVI. Please state which ofthe followings you (or your family) own.
13 Apartment
Summer house
Personal computer
El Savings account
0 Lap top
Cl I lorne theatre
171Jct
13 Dish washer
11 Car
171Freezer
CDDVD/V('D pla) cr
Cl Micro%lo"Covc1l
171Credit card
275
This document is related to the doctoral research project on corporate communications and corporate
image formation carried by Flif Karaosmanoglu under the supervision of' Dr. T. C. Mcle\%ar in the
Warwick Business School, UK. The following scale is developed to understand how the vic\., of'
consumers' about global companies' images match with what the companies want to comc). Please
evaluate your company's identity by answering the 11ollowingquestion. Thank you \,cry much Ior our
interest and cooperation.
P111)Warwick
1-] i I'Karaosmanoglu
Busiticss School. t IK
P.S. Please do not answer this questionnaire unless you are a member ofthe dccision-making process in
terms of corporate identity or corporate communication management Of) OLIrCOMP,
111)'.
Please state how much these adjectives describe the idcntity (hat your company wants to com cy to the
consumers and tick the most appropriate option below f6r each adjcctive (Pleasc cillicr hold Ilic option,
change the colour or put the letter X).
Not at all
A little bit
Sonic
Neither/Nor
Prcity much
Ver) much
Completcl)
I lonest
12
Considerate
12
Effective
12
Innovative
12
Sensitive
12
Expert
12
Trustworthy
12
Caring
12
Exciting
12
Leader
12
Rcsponsible
12
Fun
12
Competitive
12
Young
12
Dynamic
12
Modern
12
6
6
Wc1coming
12
Successful
12
-5
5
6
6
Progressive
12
Capable
12
Creative
12
'176
References
277
278
279
Creating
Strategic
Advantage",
Corporate
280
Communications,CassellEducationalLtd., London.
Berry, L. L. and Parasuraman,A. (1991) Marketing Services: Competingthrough
Quality, The FreePress,New York: USA.
Bevis, J. C. (1967) "How CorporateImageResearchIs Used", in the Proceedingsof
the ESOMAR Wapor Congress,Vienna.
Bhattacharya,C. B. and Sen, S. (2003) "Consumer-CompanyIdentification: A
Framework for UnderstandingConsumers'Relationshipswith Companies",
Journal ofMarketing, Vol. 67, No. 2, pp. 76-88.
Bick, G., Jacobson,M. C., Abratt, R. (2003) "The CorporateIdentity Management
ProcessRevisited",Journal ofMarkeling Management,Vol. 19,pp. 835- 855.
281
283
284
285
286
287
288
Approaches",
Joumal ofBusinessResearnkVol. 17,pp. 27-34.
Dowling, G. PL (1994) Corporate Reputations: Strategies for Developing the
Corporate Brand, Kogan Page,London.
Dowling,
Do%%mey,
S. (1986) "The RelationshipbetweenCorporateCulture and Corporate
Identity", Public RelationsQuarterly,Vol. 3 1,No. 4, pp. 7-12.
Drurnwright, M. (1996) "Company Advertising with a Social Dimension: The Role
of Non-cconomic Criteria", Journal ofAfarketing, Vol. 60, No. 4, pp. 71-87.
Duhan, D. F., Johnson, S. D., Wilcox, J. B., Herrell, G. D. (1997) "Influences on
Consumer Use of Word-of-mouth RecommendationSources",Journal ofthe
4cademyofAfarketing
Science,
283-295.
Vol.
25,
No.
4,
pp.
.
Duncan, T. and Everett, S. (1993) "Client Perceptions of Integrated Marketing
Communications", Journal qfAdwrtising Research, Vol. 33, No. 3, pp. 3039.
Duncan, T. and Moriarty, S. (1998) "A Communication-BasedMarketing Model for
Mark-cting Rclationship-, Joumal ofAfarketing, Vol. 62, No. 2, pp. 1-13.
289
and
Multinational
Perspectives, Lexington
Books,
Branding -
290
England.
Fill, C. and Diminopolu, E.'(1999) "Shaping CorporateImages:Attributes Used to
Form Impressions of PharmaceuticalCompanies", Corporate Reputation
Review,Vol. 2, No. 3, pp. 202-213.
FitzGerald, M. and Amott, D. (Eds.) (2000) Marketing CommunicationsClassics,
ThomsonLearning,London.
Folkes, V. S. (1984) "Consumer Reactionsto Product Failure: An Attributional
Approach,Journal of ConsumerResearch,Vol. 10,No. 4, pp. 398409.
Folkes, V. S., Koletsky, S., Graham, J. L. (1987) "A Field Study of Causal
inferences and ConsumerReaction: The View from the Airport, Journal of
ConsumerResearch,Vol. 13,No. 4, pp. 534-539.
Fombrun, C. (1996) Reputation: Realizing Valuefrom Corporate Image, Harvard
Business School Press,Boston.
291
292
Garten,J. E. (1997) The Big Ten: The Big Emerging Markets and How 7hey Will
ChangeOur Lives,Basic Books, New York.
Garver, M. and Mentzer, J, (1999) "Logistics Research Metohds: Employing
Structural Equation to the Test for Construct Validity", Journal of Business
Logistics,Vol. 20, No.1, pp. 33-57.
Gerbing, D. and Anderson,J. (1988) "An UpdatedParadigmfor ScaleDevelopment
Incorporating Unidimensionality and Its Assessment",Journal of Marketing
Research,Vol. 25, No. 2, pp. 186-192.
Gillham, B. (2000) TheResearchInterview, Continuum,London.
Gioia, D. A., Schultz,M., Corley, K. G. (2000) "OrganisationaI Identity, Image and
Adaptive Instability", 4cademyofManagementReview,Vol. 25, No. 1, pp. 6381.
Gioia, D. A. and Thomas,J. B. (1996) "Identity, Image, and Issue Interpretation:
Sense-makingDuring StrategicChangein Academia",.4dininistrative Science
Quarterly, Vol. 4 1,No. 3, pp. 370-403.
Glaser, B. G. and Strauss,A. L. (1967) The Discovery of the Grounded Theory:
Strategiesfor QualitativeResearch,Aldine, Chicago.
Glynn, M. A. and Abzug, R. (2002) "Institutional Identity: Symbolic Isomorphism
and OrganisationalNames",AcademyofManagementJournal, Vol. 45, No. 1,
pp. 267-280.
Goodman, M. B. (2000) "Corporate Communication: The American Picture",
Corporate Communication:An International Communication,Vol. 5, No. 2,
pp. 69-74.
293
Identification",AdministrativeScienceQuarterly,Vol. 15,pp.
Organisational
176-190.
Halstead, D. (2002) "Negative WOM: Substitute for Supplementto Consumer
Complaints". Journal of
295
Harrison-Walker, L.
J.
(2001)
"The
Measurement of
Word-of-Mouth
296
Herr, P. M. ', Kardes, F. R., Kim, J. (1991) "Effects of Word-of-Mouth and Product
Attribute
-,
Information on
Persuasion: An
Accessibility-Diagnosticity
297
298
299
300
301
302
303
304
305
306
Pratt, WG.
Distributors,"
Administrative Science
Mobility: Why Finns Leavethe NASDAQ and Join the New York Stock
Exchange",AdministrativeScienceQuarterly, Vol. 45, No. 2, pp. 268-292.
Reingen, R. H. and Kernan, J. B. (1994) "Social Perccption and Intcrpersonal
Influence: SomeConsequences
of the PhysicalAttractivenessStereotypein a
PersonalSelling Setting,Journal of ConsumerPsychology,Vol. 2, No. 1, pp.
25-38.
Richins, M. (1983) "Negative Word-of-mouth by Dissatisfied Consumers:A Pilot
Study",Journal ofMarkeling, Vol. 47, No. 1, pp. 68-78.
Ries, A. and Trout, J. (1981) Positioning: The Battle for the Mind, Warner Books,
New York, NY.
307
308
309
Simocs, C. and Dibb, S. (2001) "Rethinking the Brand Concept: New Brand
Orientation", Corporate Communications:An International Journal, Vol. 6,
No. 4, pp. 217-224.
Shoes, C., Dibb, S., Fisk, R. P. (2005) "Managing CorporateIdentity: An Internal
Perspective",
-Journal of theAcademyof Marketing Science,Vol. 33, No.2, pp.
153-168.
Simpson, N. (1979) Corporate Identity: Name and Perception, The Conference
Board,New York, NY.
Sirgy, J. M. (1981) "Testing A Self-ConceptModel Using A Tangible Product", in
Proceedings of the American Psychological Association -
Consumer
310
311
312
Wakefield, R. L. and Whitten, D. (2006) "Examining User Perceptionsof ThirdParty Organization Credibility and Trust in an E-retailer", Journal of
Organizationaland End User Computing,Vol. 18,No. 2, pp. 1-9.
Weimann, G. (1983) "The Strength of Weak ConversationalTies in the Flow of
Information and Influence", Social Networks,Vol. 5, September,pp. 245-267.
Weiner, B. (1974) Achievement Motivation and Altrihution Theory, General
LearningPress,Morristown, NJ.
Weiner, B. (1985) "An Attributional Theory of Achievement Motivation and
Emotion", PsychologicalReview,Vol. 92, No. 4, pp. 548-573.
Weiner, B. (1986) An Altributional Theory of Motivation and Emotion, SpringerVerlag, New York, NY.
Weiner, B. (1992) Human Motivation: Metaphors, Yheoriesand Research,Sage
Publications,Newbury Park,CA.
Weiss,A. M., Anderson,E., Maclnnis, D. J. (1999) "Reputation Managementas A
Motivation for SalesStructureDecisions",Journal of Marketing, Vol. 63, No.
4, pp. 74-89.
Wells, B. and Spinks, N. (1999) "Developing A Community Image Program:An
Essential Function of Business Communication", Management Decision,
Vol. 37, No. 3, pp. 289-294.
Westberg, S. J. (1994) Understanding Corporate Image: Implications for
CommunicationStrategy,UnpublishedPhD Thesis,The University of Texasat
Dallas,Dallas, USA.
Whctten, D. A. and Godfrey, P. C. (Eds) (1998) Identity in Organisallons:Building
Theorythrough Conversations,ThousandOaks,CA: Sagc.
314
Study
Organisational
for
Its
Implications
the
of
Identity
Organisational
and
Reputation",Business& Society,Vol. 41, December,pp. 393414.
Wiedmann,K. P. (2002) "Analysing the GermanCorporateReputationLandscape',
Corporate Reputation Review, Vol. 4, No. 4, pp. 337-353.
315