Você está na página 1de 9

Journal of Retailing and Consumer Services 21 (2014) 220228

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services

journal homepage: www.elsevier.com/locate/jretconser

Anatomy of shopping experience for malls in Mumbai:

A conrmatory factor analysis approach
Harvinder Singh a,n, Sanjeev Prashar b,1
Institute of Management Technology (IMT), Raj Nagar, Ghaziabad 201001, Uttar Pradesh, India
Indian Institute of Management (IIM), Raipur, Chhattisgarh, India

art ic l e i nf o a b s t r a c t

Article history: The city of Mumbai has not yet exhausted its potential in terms of availability of mall space. Yet shoppers
Received 5 August 2013 of Mumbai seem to prefer high-street locations even if the product or format is operating from inside a
Accepted 5 August 2013 shopping mall also. It means that shoppers look beyond the basic chore of shopping and experience plays
Available online 22 August 2013
a vital role. This study was undertaken to understand the composition of shopping experience so that
Keywords: mall developers and managers succeed in generating exciting among shoppers by orchestration of
Shopping malls shopping experience using components as identied at the end of this research. It is based on empirical
Shopping experience investigation of 400 respondents selected from four shopping malls in Mumbai. The study identies
Mall management ambience, convenience, marketing focus, safety & security and physical infrastructure as factors vital in
Conrmatory factor analysis
dening shopping experience. Results were conrmed using conrmatory factor analysis.
Customer preference
& 2013 Elsevier Ltd. All rights reserved.

1. Introduction It means that the city is not saturated in terms of mall potential
but the shoppers are showing preference for high-street locations,
Mumbai, the nancial capital of India, is one of the seven maybe because they nd them more appealing. However with
metropolitan cities of India. The city accommodates 18.4 million the rapid expansion of Indian middle class having rising income
people and has per capita income of Rs. 1, 41,000 which is nearly levels and mounting brand consciousness, demand for a superior
thrice the per capita income of India (Department of Economics shopping experience is evident in the metropolitan cities (Mayank
and Statistic, Government of Maharashtra, 2012). Due to its rened et al., 2010).
demographics the city was expected to be a fertile ground for In recent past it has been observed that Indian retailers are
development of modern retail and shopping malls. In fact Indias showing a preference for high-street locations over mall locations.
rst shopping mall Crossroads did come up in Mumbai in the year Apart from reasons relating to operational and nancial efciency,
1998. However the potential seems to have zzled out in recent it is often said that high-street locations offer better footfalls (Nair,
years. Taking per capita mall space (PCMS) for Indias big cities to 2010; Kamath, 2012). If retailers that operate the same format
be 1.2 square feet per person, Mumbai can easily and protable from both these locations say so it mean that reason for customer
accommodate 22 million square feet of mall space (ASIPAC, 2011). s preference to high-street goes much beyond the merchandise or
Total available mall space in Mumbai was about 13 million square price that remain the same in both the cases. This search leads the
feet as on 2011 (ASIPAC, 2011). Even then malls in Mumbai are researcher towards experiential aspects of shopping.
facing reduced footfalls, spiralling vacancy rates and reducing mall
rentals. When compared with traditional, high-street markets like
Linking Road, picture is rather gloomy. Real estate rentals at place 2. Literature review
like Linking Road are pegged at somewhere close to Rs. 850 per
square feet where highest mall rentals in Mumbai are constrained Shopping malls are of very recent origin in India and hence
at Rs. 480 per square feet (Cushman and Wakeeld, 2012). there is a shortage of India-specic studies in this eld. Most of the
studies available on management aspects of shopping malls have
been conducted in the US and Europe where the phenomenon
called shopping mall has matured.
Corresponding author. Tel.: +91 120 3004369, Mob.: +919810280586; Looking at the Indian studies the most popular theme of
fax: +91 120 2827895.
E-mail addresses: hsingh@imt.edu, singhharvinder1974@gmail.com,
research on shopping malls provides a descriptive analysis of the
hsingh@imt.edu (H. Singh), dr.sanjeev.prashar@gmail.com (S. Prashar). status at present. It includes studies done by Kuruvilla and Ganguli
Tel.: +91 9039039499. (2008) in which they discussed development and operations of

0969-6989/$ - see front matter & 2013 Elsevier Ltd. All rights reserved.
H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228 221

shopping malls in Indian context; a study by Singh and Bose for studying entertainment in context of shopping malls. The study
(2008) on comparison of Indian and the US shopping malls and revealed that mall developers and managers have indeed been using
another study by Swaminathan and Vani (2008) checking custo- entertainment component of a mall as a signicant method to
mer attitude towards shopping malls. A study that comes closer to differentiate the mall and increase their market share. Research
exploring experiential aspects of shopping is the one carried out developed an instrument to tap into multiple dimensions of what
by Tripathi and Sidddiqui (2008) in which the authors study the makes for an entertaining mall visit.
relationship between service environment and patronage inten-
tions. However in this work authors stopped at establishing 2.2. Mall attributes and customer expectations
relationship between an already existing theoretical construct
and customer behaviour. Study that comes closest to the present Many researchers tried to explore the attributes and features that
work is the one by Singh and Sahay (2012) in which the authors would make a mall more entertaining or enjoyable for shoppers and
explored the composition of shopping experience for shoppers of each came up with a different set of attributes as answer. It conrms
Delhi NCR. However the researchers did not validate the results the notion that mall shopping and retail patronage is a relative choice
using an established statistical procedure. phenomenon (LeHew and Fairhurst, 2000; Lowry, 1997).
These studies indicate towards the huge gap in terms of studies Relationship between retail patronage and physical attributes
exploring the psychology of shoppers and extracting the signi- like size and distance was studied by Huff and Rust (1984). Their
cance of experiential aspects of mall shopping. Outcome of the study explored all aspects like denition, estimation and measure-
research should not only be distinct but should also stand the test ment of congruence of retail trading area and propounded retail
of statistical prowess and validity. gravity model which helped in predicting mall patronage based on
the principle of cost (accessibility) versus utility (size). The model
2.1. Shopping experience tried to predict and explain retail patronage as a function of store
size and distance from the consumer. Square footage of selling
Numerous studies have been conducted in past highlighting space was used as a surrogate measure for the attraction of
the signicance of shopping experience albeit most of these are in shopping area. Craig et al. (1984) used central place theory and
context of retail stores and not shopping malls. Explaining shop- viewed shopping areas as commerce centres to which consumer
ping as a pursuit for product acquisition explanation does not households must travel to obtain needed goods and services.
reect its total value as consumers shop not only for goods and Research based on central place theory employed economic utility
services but also for experiential and emotional reasons. This models that incorporated factors such as distance/travel time and
aspect of shopping has been explored by many researchers the size of a centre to express the relationship between costs and
(Jones, 1999). Signicance of experiential aspects of shopping benets of shopping area choice.
was rst highlighted by Holbrook and Hirschman (1982) when Other spatial aspects like proportion of shopping area indoors,
they enriched traditional information processing approach with an proportion of shopping area reserved for pedestrians and com-
admixture of symbolic, hedonic, and aesthetic nature of consump- pactness were used by Oppewal and Timmermans (1999) in their
tion. It was also found that perceived store shopping experience is experimental study consumer perception of public spaces using
relatively more important than merchandise price or quality conjoint analysis.
perceptions in explaining consumers value perceptions of a retail Tenant-mix is another area that holds relevance for shopping
store (Kerin et al., 1992). In an experimental study carried out by malls. In a study based in UK, Kirkup and Raq (1994) studied
Swinyard (1993), shopping experience emerged as a relevant force management of tenant-mix in new shopping centre. They
along with customer mood and involvement level as he studies explored major occupancy problems faced by owners/managers
their impact on shopping intentions. The same point was estab- of shopping centres in handling tenants and found that develop-
lished by Bellenger and Korgaonkar (1980) with the help of ment of strong, distinctive and consistent tenant-mix is crucial for
shopper proling. Worked on proling of people who enjoyed the success of a shopping centre. Malls with more and larger
shopping as leisure-time activity, they found that 69% of respon- anchors were posited as better locations for specialty chains of
dents fell in recreational shopper category supporting the idea that small stores. In a study relating to new shopping malls during rst
recreational shoppers are a signicant force in retail shopping and few months of their launch, Brito (2009) established that it is the
recreation or entertainment is a major deliverable in shopping. store selection and retail-mix that holds the key to making image
Mall-specic discussion of shopping experience was carried out inuence patronage effective.
by Csaba and Askergaard (1999) in an article covering historical A numbers of studies have examined the effect of physical
development of shopping malls in the US. They emphasized that environment on consumers emotions and patronage intentions
shopping experience in a mall is the outcome of interplay between (Ward et al., 1992). Mattila and Wirtz (2001) studied impact of
two different sets of forces/factors. One set of factors operate at the scent and music with the help of an experimentation study. They
front stage whereas the second set operated at the backstage. manipulated scent and music in a three by three factorial design in
Shopping experience in a mall is interplay of both these sets. a eld setting and results indicated a denitive impact. Donovan
An extension of the same concept is total shopping experience. and Rossiter (1982) worked on impact of store atmospherics. Apart
It is an overall assessment of subjective worth considering all from these softer elements physical and tangible elements like
relevant evaluative criterions. Here value is all factors, quantita- decorations and furnishing and greenery in and around a mall also
tive and qualitative, subjective and objective, that make up the affect consumer perception of shopping centres (Oppewal and
complete (total) shopping experience. Babin et al. (1994) gave it Timmermans, 1999).
more signicance by mentioning that value is provided by the Developing a relationship between mall attributes (tenant variety,
complete (total) shopping experience; not simply by product mall environment and shopping environment) and shoppers excite-
acquisition. Entertaining shopping experience is supposed to ment and desire to stay at the mall, Wakeeld and Baker (1998)
increase the likelihood of store patronage by the shoppers. tested a number of variables. In their validated research instrument,
Baker and Haytko (2000) made a signicant comment on the important attributes like music, lighting, temperature, design, archi-
composition of shopping experience when they empirically explored tecture, stores, restrooms and entertainment were accommodated.
the emotional content of shopping experience in context of teen girls. These attributed were clubbed into four constructs: ambient factors,
In recent years Kim et al. (2005) used graphical modelling approach design factors, layout and variety. Apart from other variables,
222 H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228

researches could establish denite relationship between specic address this issue. It aims at identifying the factors dening
environmental elements and excitement and desire to stay in shopping experience of an Indian shopper. It explores the follow-
the mall. ing issues:
Convenience as a vital attribute has been studied by Reimers
and Clulow (2009) in an Australian setting. The study established 1. Which factors dene shopping experience in a mall?
the eminence of convenience irrespective of the format (shopping 2. Are all the factors equally relevant for the mall shoppers in
mall or shopping strip). Elements of convenience taken up in the Mumbai?
study were: Spatial convenience, temporal convenience, car con- 3. How shopping experience can be improved in light of the
venience, hedonic attributes and shopping services. ndings of this research?
Collating the work on mall attributes done by previous
researchers, Kim et al. (2005) summarized vital mall attributes
into a comprehensive list of 51 items. It was later pruned to 38.
These 38 items were falling in following nine groups: layout, 4. Research methodology
design & architecture, extra facilities, mood, courtesy, getting out,
exhibition, music & food and hanging around. Output of this The study included both exploratory as well as conclusive
research was an instrument to tap into multiple dimensions of phases. Whereas exploratory phase was used primarily for back-
what makes for an entertaining mall visit. A similar list of ground study and questionnaire development, conclusive study
attributes was used by Venkateswarulu and Uniyal (2007) with dealt with data collection from actual respondents through a
reference to shopping malls in the city of Mumbai. El-Adly (2007) structured questionnaire.
identied six factors dening attractiveness of a shopping mall.
These were: Comfort, entertainment, diversity, mall essence,
convenience and luxury. 4.1. Designing of research instrument
Safety and security issue also emerge as a relevant theme
affecting shopping. Frasquet et al., 2001) included personal security Background of the study included exploration into what consti-
as a variable while studying perceived value of shopping malls. tutes shopping experience. Review of literature as discussed pre-
In a study focussed on safety and security concerns of shopping viously resulted in generation of a list of 34 attributes that were
centre customers Overstreet and Clodfelter (1995). El-Adly (2007) possible candidates for inclusion in intended model for shopping
also considered security as a mall attribute though he has men- experience. This checklist was then consulted with an expert panel
tioned it as a constituent of dimension Comfort. Jackson et al. comprising of two mall managers of the level of General Manager,
(2011) also used safety of mall as an attribute in their study on tow professors who teach and publish on this theme and one
impact of gender and generational cohorts on mall attribute research scholar pursuing doctoral research of mall management.
importance and shopping value. Panel recommended deletion as well as addition of attributes to the
Frasquet et al. (2001) used attractive sales and promotion as a initial list. For instance statements related to entertainment as
suggested by El-Adly (2007) were deleted as entertainment options
variable affecting perceived value of a shopping mall. Signicance of
marketing activities was also highlighted by a study done on malls in were considered a part of tenant-mix by many researchers like
Wakeeld and baker (1998). On the other hand security was
Kolkata by Chattopadhyaya and Sengupta (2006). They posited that
shopping centres which could achieve distinctive positioning by expanded to four statements each representing a different source
of potential threat. These were: safety from acts of terror, safety from
developing appropriate marketing strategies enjoyed increased cus-
tomer patronage. Warnaby and Yip (2005) did a similar study on accidents (like re), safety from crimes and safety while shopping
(like slippery oor). The panel rened the list to twenty-two mall
promotional planning for shopping centres in UK.
attributes that could possibly dene shopping experience for malls in
the city of Mumbai. Initial draft of questionnaire was test-checked by
3. Research objectives administering to a sample of 20 respondents to checks appropriate-
ness in terms of wording and sentence structure. Final outcome of
Research conducted during last few decades presents a long list this phase was the research instrument containing 22 statements
worded carefully to check a shoppers response (agreement/disagree-
of attributes and factors shaping shopping experience in different
contexts. Indian shopping malls in general and shopping malls of ment) on vital attributes of a shopping mall. Responses to these
statements were to be recorded on a 5-point Likerts scale in which a
Mumbai in particular score very well in terms of these factors
(Singh and Bose, 2008; Kuruvilla and Joshi, 2010). But customer score of 1 indicated strong disagreement whereas score of 5 indicated
response to shopping malls has been luke-warm despite tremen- strong level of agreement with the given statement. All the state-
dous untapped potential as discussed in the introduction section. ments were positively worded with no reverse coding.
Since high-street markets still manage to attract decent footfalls,
does it mean that high-street markets offer something more 4.2. Sampling and mode of contact
relevant to shoppers in terms of shopping experience? Since
Indian malls are more or less a replica of western malls (in some A sample of 400 respondents was used for this study. Population
case a more ornate version) does it mean that Indian shoppers for this research was dened as people who visit malls for the
denes dene shopping in a different manner as compared to its purpose of shopping and actually carry out shopping. Sampling
western counterparts? It would be wrong to say that shoppers of element was individual shopper visiting the malls. To ensure data
Mumbai are unenthusiastic towards the concept of shopping mall collection only from genuine shoppers it was decided to contact
since malls like Inorbit Mall and Oberoi Mall consistently attract persons carrying minimum of one shopping bag while coming out of
footfalls. Maybe given a choice, shoppers decide in favour of one a mall. Quota sampling was used for this study. Four malls from the
offering superior shopping experience. Such an experience can be city of Mumbai were selected for data collection. One mall was taken
ensured by carefully aligning management philosophy and prac- from South Mumbai, One from Malad and two malls were from
tices of the mall with the expectations of shoppers. However, Mulund. Hundred respondents were selected from each mall. Malls
a determined action on this front requires estimation of what selected for this research were similar in terms of size, age and
includes shopping experience. This research was conceived to tenant-mix. Mall intercept method was chosen to contact the
H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228 223

Table 1
Tabulated factor output.

Factor no. Factor title Variables included Average Factor

F1 20.324% Ambience Temperature control (3.19) 3.14 .874

Background music (2.82) .874
Landscaping (3.14) .860
Ambient odor (3.02) .856
General hygiene (3.55) .853
Illumination (3.11) .846
F2 14.504% Convenience to shoppers Floor plan (3.72) 3.35 .883
Utilities (3.27) .878
Distance of mall from home (3.31) .853
Lifts & escalators (3.11) .851
F3 12.895% Marketing focus Promotional events (2.64) 3.08 .853
Promotional schemes (2.96) .830
Mall faade (3.16) .825
Tenant-mix (3.56) .816
F4 11.163% Safety & security Security against acts of terror (2.560) 2.35 .827
Safety from accidents (2.35) .811
Safety from crimes (2.21) .758
Safety while shopping (2.28) .676
F5 9.811% Physical infrastructure Size of the mall (2.85) 2.82 .783
Size of the atrium (2.66) .740
Open space (2.79) .714
Parking space (2.97) .649

Table 2 Table 3
Cronbachs alpha values for factors. Variance extracted by each factor.

S. no Factor no. and label Cronbachs alpha S. no Factor no. and label Variance

1 F1 (Ambience) 0.931 1 F1 (Ambience) 0.741

2 F2 (Convenience to shoppers) 0.901 2 F2 (Convenience to shoppers) 0.751
3 F3 (Marketing focus) 0.858 3 F3 (Marketing focus) 0.691
4 F4 (Safety & security) 0.780 4 F4 (Safety & security) 0.593
5 F5 (Physical infrastructure) 0.705 5 F5 (Physical infrastructure) 0.523

shoppers. Second stage involved selecting respondents on the basis 5.2. Examining the initial output
of convenience and judgment.
Data was subjected to data reduction using principal compo-
nent analysis. Initial output was received along with communality
5. Data analysis and extraction of factors values (Annexure A2). All the communality values were greater
than 0.5 except for one (parking space; 0.488). Since variable-wise
Data collected from 400 respondents was subjected to data MSA for the same was reasonable hence the variable was not
reduction using exploratory factor analysis (EFA) with the help of deleted. Data was subsequently subjected to Varimax rotation.
statistical package for social science (SPSS). This research used Rotated solution explains 68.697% of the variance associated
principal component analysis. Data was checked and found with the problem (Annexure A3) and comprised of ve factors
suitable for factor analysis in terms of vital parameters. Initial (Annexure A4).
unrotated solution was received and subjected to Varimax rotation
for further renement. 5.3. Listing and labelling of factors

Exploratory factor analysis condensed 22 statements into a ve

5.1. Checking suitability of data for factor analysis
distinct bunch of statements. Statements within a single bunch
had something in common as indicated by signicantly higher
Data set was checked for suitability by examining the output of
factor loadings whereas, statements placed in different bunches
correlation analysis, variable-wise measure of sampling adequacy,
were dissimilar to each other in terms of shoppers response. Each
KMO test of sampling adequacy and Bartletts test of sphericity
bunch or bundle represented a factor. For assigning a label (name)
(Boyd et al., 2002; Malhotra, 2004). Presence of an underlying
to each factor, nature and wording of all the statements com-
structure in the data-set was indicated by signicant correlation
prising that factor were examined and considered. The exercise
between a large numbers of variables. The observation was corro-
resulted in ve distinct names (Table 1).
borated by Anti-image correlation matrix showing adequate variable-
wise measure sampling adequacy (diagonal values greater than
0.5; Annexure A1)). KMO test of sampling adequacy gave a value of 5.3.1. Factor 1
0.845 which is higher than bench-mark value of 0.5. Calculated First factor extracted out of this research explained 20.324% of
value of 4451.756 at 231 degrees of freedom and 5% level of total variance. It comprised of six mall attributes (variables)
signicance in Bartletts test of sphericity also indicated the same namely temperature control, cleanliness, pleasant odour, back-
(Annexure A1). All these results indicated that it was a t case for ground music, illumination and landscaping. Each of these vari-
factor analysis. ables was strongly correlated with the extracted factor 1 as factor
224 H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228

ambience to be a potent factor as its average score was above 3

(3.14) indicating agreement with the statements constituting this
factor. Variable-wise average score ranged between 2.82 and 3.55.
Background music was the only exception with score below 3
(2.82) whereas general hygiene was considered most signicant
(average score 3.55).

5.3.2. Factor 2
Second factor extracted in this analysis explained 14.504% of
the total variance and comprised of four variables or mall
attributes. These were: oor-plan, utilities, distance of mall from
home and lifts & escalators. These variables were highly correlated
with the extracted factor and the factor loadings ranged between
0.883 and 0.851. Each of these variables was a source of conve-
nience or ease for the shoppers. Hence it was labelled as Conveni-
ence to shoppers. It emerged as the most signicant factor in
customers opinion with highest average score i.e. 3.35. Another
feature of this factor was that each attribute (variable) constituting
this factor was individually considered important by the respon-
dents since average score for each variable was above 3.

5.3.3. Factor 3
Third factor comprising of four variables (attributes) explained
12.895% of the total variance. Attributes comprising this factor
were: promotional events, promotional schemes, mall facade and
tenant-mix. Factor loadings of these four variables on factor 3 were
healthy and ranged from 0.853 to 0.816. Labelling this factor
was quite problematical. Events and promotional schemes are an
attempt by the mall to induce shoppers and may be classied as
marketing activity. Facade of a mall also attracts people from over
a distance. However, nature of impact made by tenant-mix is
different. Diversity of tenant-mix extends a rational benet to
the shoppers to induce them for a mall visit. Though impact of

Table A1
KMO and Bartletts test.

KaiserMeyerOlkin measure of sampling adequacy .845

Bartletts test of sphericity Approx. Chi-square 4451.756
Df 231
Sig. .000

Table A2
Communalities for 22 variable problem.

Initial Extraction

Distance of mall from home 1.000 .747

Mall faade 1.000 .709
Temperature control 1.000 .765
General hygiene 1.000 .735
Fig. 1. Proposed model for mall shopping experience. Size of the atrium 1.000 .586
Floor plan 1.000 .807
Tenant-mix 1.000 .673
Table 4 Parking space 1.000 .488
Model summary. Ambient odor 1.000 .740
Promotional events 1.000 .739
Weights Covariances Variances Means Intercepts Total Utilities 1.000 .800
Security against acts of terror 1.000 .714
Fixed 27 0 0 0 0 27 Background music 1.000 .772
Labeled 0 0 0 0 0 0 Illumination 1.000 .724
Unlabeled 17 10 27 0 0 54 Safety from accidents 1.000 .670
Total 44 10 27 0 0 81 Size of the mall 1.000 .632
Promotional schemes 1.000 .702
Safety while shopping 1.000 .528
loadings for each of these were signicantly high (range from Lifts & escalators 1.000 .732
0.874 to 0.846). Each of these variables represented an element Open space 1.000 .514
Landscaping 1.000 .749
that adds to aesthetics and magnicence offered by the mall. This Safety from crimes 1.000 .587
factor was labelled as ambience. Respondents considered
H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228 225

Table A3
Total variance explained for the given problem.

Comp Initial Eigen values Extraction sums of squared loadings Rotation sums of squared loadings

Total % of Cumulative Total % of Cumulative Total % Of Cumulative

1 4.691 21.324 21.324 4.691 21.324 21.324 4.471 20.324 20.324

2 3.741 17.007 38.330 3.741 17.007 38.330 3.191 14.504 34.828
3 2.689 12.221 50.551 2.689 12.221 50.551 2.837 12.895 47.723
4 2.072 9.420 59.971 2.072 9.420 59.971 2.456 11.163 58.886
5 1.920 8.726 68.697 1.920 8.726 68.697 2.158 9.811 68.697

Table A4
Rotated component matrix for 20 variable problem.


1 2 3 4 5

Temperature control .874 .021 .005  .043 .005

Background music .874  .075 .061 1.628E-5 .006
Landscaping .860  .003 .082  .037 .032
Ambient odor .856 .056 .048  .041  .016
General hygiene .853  .002 .040  .065  .044
Illumination .846 .020 .033  .037 .076
Floor plan .003 .883  .051 .118 .105
Utilities .044 .878  .072 .121 .089
Distance of mall from home  .021 .853 .054 .029 .124
Lifts & escalators  .001 .851  .072 .040  .017
Promotional events .057  .067 .853  .011  .056
Promotional schemes .013  .057 .830  .076  .065
Mall faade .096  .010 .825  .132 .033
Tenant-mix .056 .003 .816  .059 .037
Security against acts of terror  .069 .072  .140 .827 .006
Safety from accidents  .043 .073  .076 .811  .011
Safety from crimes  .001  .056  .093 .758  .021
Safety while shopping  .073 .227 .043 .676 .107
Size of the mall .005  .098 .050  .078 .783
Size of the atrium .000 .182  .017 .069 .740
Open space .036  .017  .025 .043 .714
Parking space .000 .251  .055 .031 .649

Table A5
List of labels used in Fig. 1.

Label Description Label Description

F1 Ambience V9 Ambient odor

F2 Convenience to shoppers V10 Promotional events
F3 Marketing focus V11 Utilities
F4 Safety & security V12 Security against acts of terror
F5 Physical infrastructure V13 Background music
V14 Illumination
V1 Distance of mall from home V15 Safety from accidents
V2 Mall faade V16 Size of the mall
V3 Temperature control V17 Promotional schemes
V4 General hygiene V18 Safety while shopping
V5 Size of the atrium V19 Lifts & escalators
V6 Floor plan V20 Open space
V7 Tenant-mix V21 Landscaping
V8 Parking space V22 Safety from crimes

tenant-mix is not as obvious as other three variables, it does variable with the highest variable-wise score in the entire research
include the basic marketing functions of need assessment and (3.56).
fullment. Collating these four variables, the most appropriate
label seemed to be marketing focus. Overall average score for
this factor was 3.08. However all the variables (attributes) were 5.3.4. Factor 4
not equally important as promotional events and schemes had Factor number four comprised of four variables (attributes) and
variable-wise average scores below 3 (2.64 and 2.96, respectively). explained 11.163% of total variance. Factor loading for all four
On the contrary tenant-mix turned out to be the most relevant variables was high (ranging between 0.827 and 0.676) indication a
226 H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228

high level of correlation between these variables and the factor signicant with factor loadings ranging from 0.783 to 0.649.
concerned. Variables included in this factor were: security against However this factor was not very signicant as indicated by its
acts of terror, safety from accidents, safety from crimes and safety average score (2.82). Average score of each of the constituent
while shopping. Since all these variables pointed at shoppers need variables was also less than 3.
to feel immune from mishaps, it was labelled as safety &
security. It was found to be the least signicant with average
score of 2.35. Individually, each of these variables commanded a 5.4. Validity and reliability of factor output
luke-warm response from shoppers as average score for each of
these variables was less than 3. Validity and reliability of the factor output was checked
statistically. Reliability was established by estimating Cronbachs
5.3.5. Factor 5 alpha for each of the factors, as given in Table 2. Alpha value for all
This factor also comprised of four variables and explained the factors was above 0.7 indicating that the output is reliable.
9.811% of total variance. Variables included were: size of the mall, Convergent validity for a factor indicates that all variables
size of the atrium, open space in the mall and parking space. constituting a single factor are actually coherent. i.e. they share
Extent of correlation between the factor and the variables was a high proportion of variance in common. Convergent validity
was checked with the help of variance extracted (VE). VE was
Table A6 calculated by adding squared factor loadings for all variables
Factor loadings for proposed model. constituting a factor and dividing total sum by the number of
variables (Table 3).
Variance extracted for all ve factors was greater than 0.5.
PARK INFR .596 Discriminant validity establishes that the factors extracted are
OPSP INFR .547 truly distinct from each other. It was established by observing
ATRM INFR .706 factor loading for each variable under all factors. Final factor
output showed absence of any cross loadings as there was only
CRIM SAFT .636 one signicant factor loading per variable.
6. Conrmation of factor structure
LIFT CONV .776 Factor structure as indicated by exploratory factor analysis is
DIST CONV .806 represented by Fig. 1. It represents the measurement model for
UTIL CONV .866 shopping experience in malls in the city of Mumbai. The model
comprises of ve factors (F1, F2, F3, F4 and F5). It depicts bunch of
LAND AMB .834 variables associated with respective factors, error variance asso-
MUSC AMB .848 ciated with each variable and co-variance between the factors. List
TEMP AMB .848 of labels used in Fig. 1 is given in Table A5. This model was
subjected to a conrmatory test in the form of conrmatory factor
analysis (CFA).

Table A7
Parameter estimatesregression weights.

Estimate S.E. C.R. P Label


OPSP INFR .852 .110 7.713
ATRM INFR 1.137 .133 8.568
SIZE INFR .993 .123 8.099
CRIM SAFT 1.234 .132 9.330
ACCD SAFT 1.235 .121 10.177
TERR SAFT 1.424 .136 10.451
FCDE MKTG .984 .068 14.373
SCHM MKTG 1.050 .072 14.520
EVNT MKTG 1.069 .072 14.925
DIST CONV 1.112 .065 17.009
UTIL CONV 1.223 .066 18.483
FLRP CONV 1.207 .064 18.828
ODOR AMB 1.000
LAND AMB 1.002 .050 19.902
MUSC AMB 1.038 .051 20.425
TEMP AMB .995 .049 20.417
HYGN AMB .923 .047 19.520
ILLU AMB .891 .046 19.218
H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228 227

6.1. CFA results and model t ambient issues? The questions remain unanswered and point
towards a series of researches to be conducted in future. Despite
Proposed model was evaluated with the help of CFA. Evaluation acts of terrorist violence in Mumbai in the year 2011 and repeated
was done to check adequacy of parameter estimates as well as the bomb-blasts in Pune, safety & security has drawn the least
model as a whole (Table 4). attention of shoppers. It is understandable since no act of terrorist
Model given in Fig. 1 had 49 variables out of which 27 were violence has happened in any of the shopping malls in India. Same
exogenous and 22 were endogenous. Data was subjected to holds true for accidents and crimes in Indian and Mumbai malls.
conrmatory factor analysis using AMOS. It resulted in achieve- Mall developers and managers should take an integrated view of
ment of a minimum. There were 253 distinct sample moments and the situations and work on multi-pronged strategies to improve
54 distinct parameters to be estimated resulting in 199 degrees of shopping experience. It is evident that all the constituent factors do
freedom. All the parameters were feasible and standard errors not contribute equally to shopping experience. It is imperative for
reasonable. Statistical signicance of parameter estimates was decision makers to identify the key factors and decision areas where
established as test-statistic in each case was greater than 1.96 suitable changes can yield more than proportionate dividends.
(Annexure A6A8). Observations and suggestions presented above come with
The proposed model was found adequate as indicated by a note of caution. These are derived from responses given by
goodness-of-t statistics. Minimum discrepancy/degree of free- a sample of respondents from Mumbai over a certain period of
dom (CMIN/DF) ratio was 1.651 which indicated adequate t. time. These results need to be supplemented by follow up
Goodness-of-t indices like goodness-of-t Index (GFI), adjusted researches on similar pattern before arriving at a rm hypothesis
goodness-of-t Index (AGFI) and incremental t index (IFI) had about composition of shopping experience. These results are
values closer to 1.000 (GFI 0.931, AGFI 0.912, IFI 0.970). com- applicable for Mumbai and can be extrapolated to the whole of
parative t index (CFI), relative t index (RFI) and TuckerLewis India only after validation by further researches. Future studies
index (TLI) had values close to 0.95 (CFI0.970, RFI 0.916, may focus on collecting similar data from shopping malls across
TLI 0.965). Root mean square error of approximation (RMSEA) different cities, at different points of time and from different
value was less than 0.04 (0.040). Both.05 and.01 values of
Hoelters Critical N for the hypothesized model exceeded 200 Table A8
(.05 283, .01 301). All these indicators were a sign of good t Parameter estimatescovariances.
and the model stood validated.
Estimate S.E. C.R. P Label

6.2. Modication indices (MIs) INFR 2 AMB .014 .036 .389 .697
INFR 2 CONV .143 .036 3.956
The evaluation process got over since all goodness-of-t para- INFR 2 MKTG  .025 .031  .793 .428
INFR 2 SAFT .027 .024 1.105 .269
meters conrmed the adequacy of the. However modication indices
SAFT 2 AMB  .087 .041  2.125 .034
were checked for the sake of completeness. It was found that no MI SAFT 2 CONV .139 .039 3.533 nnn

value was signicantly large to indicate scope for renement SAFT 2 MKTG  .132 .037  3.555 nnn

(Annexure A9). It meant there is no substantial evidence of model MKTG 2 AMB .124 .053 2.348 .019
mist. These observations further validated the hypothesized model. MKTG 2 CONV  .091 .048  1.884 .060
CONV 2 AMB .006 .055 .109 .913

7. Conclusion
Table A9
Shopping malls are capital intensive Goliaths who are unfortu- Modication indices.
nately getting beaten by the Davids of high-street. It is happening
M.I. Par change M.I. Par change
because shopping malls in Mumbai are more competition centric
than customer-centric. Malls developers are investing heavily in ILLU INFR 4.189 .157 SSHP INFR 7.071 .278
making them more ornate, ignoring the real expectations of ILLU PARK 4.800 .085 SSHP FLRP 13.551 .138
customers. For the similar merchandise Mumbai shoppers seem MUSC CONV 6.620  .106 SSHP UTIL 12.081 .126
MUSC FLRP 6.729  .075 SSHP DIST 10.101 .118
to prefer high-street as it offers the appropriate experience. Since MUSC DIST 4.855  .063 SSHP LIFT 6.703 .103
shopping experience is an orchestration of a number of factors, it MUSC LIFT 8.560  .089 SSHP TNTX 9.054 .124
is important to identify and understand the factors before taking MUSC EVNT 4.022 .065 SSHP ATRM 6.922 .145
any decision. A higher average score assigned to Ambience UTIL PARK 4.249 .088 SIZE CONV 17.814  .186
DIST MKTG 9.277 .161 SIZE SAFT 9.551  .211
indicates a higher preference for this among shoppers in the city
DIST SAFT 4.064  .147 SIZE FLRP 11.795  .106
of Mumbai. It seems to belie the notion that shopping malls offer DIST SCHM 7.272 .099 SIZE UTIL 21.758  .139
better ambience. It is pointer for future research into India-specic DIST FCDE 11.663 .132 SIZE DIST 6.026  .075
variables that would make ambience more appealing. The other DIST TNTX 4.608 .078 SIZE LIFT 17.317  .136
point to ponder is dominance convenience over factors like DIST TERR 5.186  .086 SIZE FCDE 7.842 .101
DIST CRIM 4.135  .069 SIZE TERR 11.236  .118
marketing focus and physical infrastructure. It would be interest- DIST OPSP 4.363 .105 SIZE ACCD 4.154  .076
ing to nd what makes a retail store more convenient to a shopper. LIFT INFR 5.139  .203 SIZE SSHP 4.185  .072
Convenience needs to be looked from the perspective that as most LIFT SIZE 6.041  .112 SIZE OPSP 5.623 .111
of the malls in Mumbai (and in India) and already situated close to LIFT OPSP 7.124  .131 ATRM SSHP 4.631 .070
EVNT ACCD 8.835 .108 OPSP LIFT 5.467  .073
the residential clusters, at a very convenient distance. Dimensions
FCDE ACCD 5.888  .088 OPSP SIZE 5.002 .093
of convenience associated with Indian malls could be radically FCDE SIZE 7.145 .115 PARK CONV 11.101 .147
different from what it is in case of shopping malls in the developed TNTX SSHP 5.965 .094 PARK FLRP 8.004 .087
world. Interestingly physical infrastructure seems least important CRIM CONV 5.207  .131 PARK UTIL 14.858 .115
for the Mumbai shoppers. Does it mean that malls developers CRIM DIST 5.795  .095 PARK DIST 4.511 .065
CRIM LIFT 5.231  .097 PARK LIFT 7.164 .088
need to emphasise less on physical infrastructure and should focus SSHP CONV 14.215 .202
on small, neighbourhood malls that take care of convenience and
228 H. Singh, S. Prashar / Journal of Retailing and Consumer Services 21 (2014) 220228

segments of shoppers. It would be interesting to nd out whether Kerin, R.A., Jain, A., Howard, D.J., 1992. Store shopping experience and consumer
composition of shopping experience varies with place, time and price-quality-value-perceptions. Journal of Retailing 68 (4), 376397.
Kim, Y., Jikeyong, K., Minsung, K., 2005. The relationship among family and social
shopper demographics. interaction, loneliness, mall shopping motivation & mall spending of older
consumers. Psychology & Marketing 22 (12), 4753.
Kirkup, M.H., Raq, M., 1994. Tenancy development in new shopping centres.
Appendix A International Review of Retail, Distribution and Consumer Research 4 (3),
Kuruvilla, S.J., Ganguli, J., 2008. Mall development & operations: an Indian
See Tables A1A9.
perspective. Journal of Retail & Leisure Property 7 (3), 204215.
Kuruvilla, S.J., Joshi, N., 2010. Inuence of demographics, psychographics, shopping
References orientation, mall shopping attitude and purchase patterns on mall patronage in
India. Journal of Retailing and Consumer Services 17 (4), 259269.
LeHew, M.L.A., Fairhurst, A.E., 2000. US shopping mall attributes: an exploratory
ASIPAC. 2011. Mumbai Metropolitan Region: Demand and Supply of Mall Space in investigation of their relationship to retail productivity. International Journal of
India. India: Bangalore.
Retail & Physical Distribution 28 (6), 261279.
Babin, B.J., Darden, W.R., Grifn, M., 1994. Work and/or fun: measuring hedonic and
Lowry, J.R., 1997. Life cycle of shopping centres. Business Horizons, January
utilitarian shopping value. Journal of Consumer Research 20 (March), 644656.
February, 7786.
Baker, J., Haytko, D., 2000. The mall as entertainment: exploring teen girls total
Malhotra, N., 2004. Marketing ResearchAn Applied Orientation, fourth ed.
shopping experiences. Journal of Shopping Centre Research 7 (1), 2958.
Pearson Education, New Delhi.
Bellenger, D.N., Korgaonkar, P.K., 1980. Proling the recreational shopper. Journal of
Mattila, A.S., Wirtz, J., 2001. Congruence of scent and music as a driver of in-store
Retailing 56 (3), 7792.
Boyd, H.W., Westfall, R., Stasch, S.F., 2002. Marketing Research-Text & Cases, evaluations and behaviour. Journal of Retailing 77 (2), 273289.
seventh ed. AITBS, New Delhi. Mayank, H., Mirchandani, A., Gupta, A.K., 2010. India Retail: Opportunities in a
Brito, P.Q., 2009. Shopping centre image dynamics of a new entrant. International Revolution. Jones Lang LaSalle, Mumbai.
Journal of Retail and Distribution Management 37 (7), 580599. Nair, A. 2010, August, 28. Ahmedabad, NCR Malls Hit as Retail Cos Prefer High-street
Chattopadhyaya, A., Sengupta, K., 2006. Perception of mall shoppers in Kolkata: Stores. The Economic Times. Retrieved from http://articles.economictimes.
a study. The ICFAI Journal of Consumer Behaviour 1 (3), 3443. indiatimes.com/2010-08-28/news/27602872_1_mall-vacancy-high-street-lar
Craig, S., Ghosh, A., McLafferty, S., 1984. Models of retail location process: a review. ge-retail-stores#.UCYBeBWOGcs.email.
Journal of Retailing 60 (1), 536. Oppewal, H., Timmermans, H., 1999. Modelling consumer perception of public
Csaba, F.F., Askergaard, S., 1999. Malls and the orchestration of the shopping space in shopping centres. Environment and Behaviour 31 (1), 4565.
experience in a historical perspective. Advances in Consumer Research 26, Overstreet, J., Clodfelter, R., 1995. Safety and security concerns of shopping centre
3440. consumers and the effect of these concerns on shopping behaviour. Journal of
Cushman and Wakeeld. 2012. Market Beat Retail Snapshot: Mumbai, India. India: Shopping Centre Research 2 (1), 91109.
Mumbai. Reimers, V., Clulow, V., 2009. Retail centres: Its time to make them convenient.
Department of Economics and Statistic, Government of Maharashtra. 2012. Eco- International Journal of Retail and Distribution Management 37 (7), 541562.
nomic Survey of Maharashtra 20112012 Retrieved August 10, 2012, from Singh, H., Bose, S.K., 2008. My American cousin: a comparison between Indian and
http://mahades.maharashtra.gov.in/les/publication/esm_2011-12_eng.pdf. the US shopping malls. Journal of Asia-Pacic Business 9 (4), 358372.
Donovan, R.J., Rossiter, J.R., 1982. Store atmospherics: an environmental psychology Singh, H., Sahay, V., 2012. Determinants of shopping experience: exploring the mall
approach. Journal of Retailing 58 (1), 3457. shoppers of national capital region (NCR) of India. International Journal of
El-Adly, M.I., 2007. Shopping malls attractiveness: a segmentation approach. Retail and Distribution Management 40 (3), 235248.
International Journal of Retail and Distribution Management 35 (11), 936950. Swaminathan, F.J., Vani, V., 2008. Consumer attitude colour growth of malls: Delhi
Frasquet, M., Gil, I., Molla, A., 2001. Shopping centre selection modelling: a NCR. Abhigyan 26 (2), 4451.
segmentation approach. International Review of Retail, Distribution and Con- Swinyard, W.R., 1993. The effects of mood, involvement and quality of shopping
sumer Research 11 (1), 2338.
experience on shopping intentions. Journal of Consumer Research 20 (2),
Holbrook, M.B., Hirschman, E.C., 1982. The experiential aspects of consumption
consumer fantasies, feelings, and fun. Journal of Consumer Research 9 (Sep-
Tripathi, S.N., Sidddiqui, M.H., 2008. Servicescape of a shopping mall: a hierarchical
tember), 132240.
framework. AIMS International Journal of Management 2 (3), 215235.
Huff, D.L., Rust, R.T., 1984. Measuring the congruence of a trading area. Journal of
Venkateswarulu, A., Uniyal, D.P., 2007. Concept of a mall: measuring attitude and
Marketing 48 (4), 6874.
Jackson, V., Stoel, L., Brantley, A., 2011. Mall attributes and shopping value: perception of shoppers towards malls of Mumbai. Indian Retail Review 1 (1),
differences by gender and generational cohort. Journal of Retailing and 716.
Consumer Services 18 (1), 19. Wakeeld, K.L., Baker, J., 1998. Excitement at the mall: determinants and effects on
Jones, M.A., 1999. Entertaining shopping experiences: an exploratory investigation. shopping response. Journal of Retailing 74 (4), 515539.
Journal of Retailing & Consumer Services 6 (3), 129139. Ward, J.C., John, B., Mary, J.B., 1992. Measuring the prototypically and meaning of
Kamath, R. 2012, July, 27. Retailers Seem to Prefer Standalone Stores to Malls. retail environments. Journal of Retailing 68 (2), 194220.
Business Standard. Retrieved from http://www.business-standard.com/india/ Warnaby, G., Yip, K.M., 2005. Promotional planning in UK shopping centres: an
news/retailers-seem-to-prefer-standalone-stores-to-malls/481551/. exploratory study. Marketing Intelligence and Planning 23 (1), 4357.