Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private….. 361 7 COMPARITIVE ASSESSMENT OF TRIVENI SUPERMARKET, MARGIN- FREE MARKETS AND PRIVATE SUPERMARKETS USING SELECTED RETAIL VARIABLES 7.1 Measures 7.2 Validity and Reliability of Data 7.3 Retail factor variables considered to compare Triveni Supermarkets with Margin-Free Markets and other Private Supermarkets 7.4 Factor Analysis Using the Selected Retail Variables of Private Supermarkets 7.5 Factor Analysis Using the Selected Retail Variables of Triveni Supermarkets 7.6 Factor Analysis Using the Selected Retail Variables of Margin- Free Supermarkets 7.7 Comparative Assessment of Triveni Supermarkets, Margin- Free Markets and Private Supermarkets Using Selected Retail Factors 7.8 Statistical Tests for Comparing the Outlets Consumer goods retailing has become a highly competitive form of business as many players offer the same categories of products in the same areas. At the same time, because of a number of reasons, the retailing of consumer goods has been an attractive sector of business. As a result, different types of players have established ventures in almost all places of the country with the objective of making maximum returns. Large scale retail outlets like supermarkets or hypermarkets or value chain stores or margin free consumer stores offer products to the public in almost all towns of Kerala and in rural areas. Retailing is a customer -driven marketing strategy comprising the art C o n t e n t s
55
Embed
COMPARITIVE ASSESSMENT OF TRIVENI SUPERMARKET, …shodhganga.inflibnet.ac.in/bitstream/10603/28490/17/17_chapter7.pdf · Comparative Assessment of Triveni Supermarket, Margin –
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
7.1 Measures 7.2 Validity and Reliability of Data 7.3 Retail factor variables considered to compare Triveni
Supermarkets with Margin-Free Markets and other Private Supermarkets
7.4 Factor Analysis Using the Selected Retail Variables of Private Supermarkets
7.5 Factor Analysis Using the Selected Retail Variables of Triveni Supermarkets
7.6 Factor Analysis Using the Selected Retail Variables of Margin- Free Supermarkets
7.7 Comparative Assessment of Triveni Supermarkets, Margin- Free Markets and Private Supermarkets Using Selected Retail Factors
7.8 Statistical Tests for Comparing the Outlets
Consumer goods retailing has become a highly competitive form of
business as many players offer the same categories of products in the same
areas. At the same time, because of a number of reasons, the retailing of
consumer goods has been an attractive sector of business. As a result, different
types of players have established ventures in almost all places of the country
with the objective of making maximum returns. Large scale retail outlets like
supermarkets or hypermarkets or value chain stores or margin free consumer
stores offer products to the public in almost all towns of Kerala and in rural
areas. Retailing is a customer -driven marketing strategy comprising the art
Co
nt
en
ts
Chapter -7
362
and science of choosing target markets and building profitable relations with
customers. To design a winning marketing strategy, the marketing manager
must answer two important questions: what customers to serve (what is our
target market)? And how can we serve these customers best (what is our value
proposition?)? (Philip Kotler. 2008)1.
Supermarkets offer numerous consumer products under one roof and
provide the convenience of self service. They normally display a wide range of
packed products in racks and the space layout may suit the free movement of
customers from rack to rack to select their preferred brands. They incepted the
concept of self service from full service where the customers enjoy the
advantages of seeing, feeling, touch and assurance (Prathibha & Sharma.
2011)2. Modern consumers seek improved availability of goods and quality,
reasonable price, pleasant shopping experience, services, etc. All these retail
variables are inevitable for the long run success of a large consumer store. The
present study intends to compare the Triveni Supermarkets, Margin- Free
Markets and Private Supermarkets on the basis of the responses of common
customers of these three types of outlets on selected retail variables by the factor
analysis method. Triveni Supermarkets are in the cooperative sector, managed
and controlled by the CONSUMERFED. Margin- Free Supermarkets are retail
outlets supplying consumer goods and work under a charitable society known as
the Consumer Protection and Guidance Society. Both types of retail outlets are
found in all parts of the state of Kerala. Likewise, hundreds of single and chain
supermarkets exist in all parts of Kerala managed by individuals and corporate
entities. All these three types of consumer stores compete with one another in
the supply of food grains and other consumer goods. In the light of stiff
competition between these types of stores success depends upon the consumers’
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
363
response to the stores based on their perception on pricing strategy, quality of
goods, service attractiveness, facilities offered, etc.
7.1 Measures
For measuring the opinion of customers for the purpose of comparing
the outlets such as Triveni Supermarkets, Private Supermarkets and Margin-
Free Markets 18 variables are used. Some of the attributes in the study were
selected taking in to account the Asian Shoppers Pyramid as given in the
research conducted by IBM for Coca Cola Research Council. According to this
pyramid, the attributes selected varied from basic to value attributes. The
attributes comprise price, quality, choice, location, offers and discounts, service
level and store size. Other attributes applicable to supermarkets such as
attractiveness in display, space layout in the store, quality of packing, complaint
redress mechanism, speed in billing, cleanliness in the shop, parking facilities
and the availability of fresh stock are also included in the present study.
In this part of the study, the consumers of Triveni Supermarkets who use
Margin-Free Markets and other Private Supermarkets (outlets offering food
and grocery) were asked to rate their perception on eighteen retail variables
applicable to these three types of outlets separately and these data were used to
compare the stores. The seven point scale (7 means ‘highest’ and 1 means
‘lowest’)-one of the most commonly used scales among marketing researchers
to assess psychographic variables-is used for collecting responses from
selected common customers (Hair Jr.2004)3. Statistical tools ‘Factor Analysis’,
One way ANOVA and ‘Repeated Measure MANOVA’ are used in the study.
Exploratory Factor Analysis Method is used to reduce the large number
of variables in to classes for drawing meaningful conclusions. Repeated
Measure MANOVA is a suitable tool as the data is collected from Triveni
customers who also frequently use other types of retail outlets. The validity
Chapter -7
364
and reliability of the scales are assessed to identify and preclude any unreliable
and/or invalid measure that could arise from using multiple items
(Field.A.2000)4. Initially, a content validity was assessed to ensure that the
instrument contained the representative sample of the universe of the subject
matter of interest (J.Stevans.1998)5. This was done by examining the questions
for clarity and completeness using feedback from experts.
7.2 Validity and Reliability of Data
Before assessing construct validity, correlation matrix of variables was
drawn to check for multi- colinearity, communalities of all items individually
to understand the common variance within items. Discriminant validity was
assessed through principal component analysis using Verimax rotation, and
factor correlation matrix (Cheng. 2001)6. Only those variables were retained
which had Eigen values greater than unity and only those items were retained
which had a factor loading greater than 0.475. Convergent validity was
assessed through factor loading and item-total correlations (Cooper &
Schindler 2006)7. For factor loading, the commonly accepted value is 0.3 and
above (Hair Jr. 2004)8. This was done to determine the number of factors
needed for the study. Reliability was assessed through Cronbach’s alpha,
alpha if item deleted, and item –correlations are used to assess internal
consistency. All the tests used in this research were done using the statistical
programme ‘Statistical Package for Social Sciences’ (SPSS- 17).
7.3 Retail factor variables considered to compare Triveni Supermarkets with Margin-Free Markets and other Private Supermarkets
Triveni Stores, Margin- Free Markets and Private Supermarkets offer the
same type of consumer goods to the public and their outlets are located in the
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
365
same areas and so they have to compete with one another. All these outlets
claim to sell quality consumer goods at low prices. Essential goods used for
food preparation like food grains, powdered grains and other powders and oil ,
sugar, biscuits, nuts and other packed bakery items, cleaning items, personal
and home care products and other staple items are commonly offered by
these three outlets. The price level and quality of these items as perceived by
the common customers, service attractions, promotions, space layout,
availability of brands, etc offered by the three outlets are considered for the
study. The pilot study has revealed that 18 variables are important in
determining the competency of a consumer retail outlet to satisfy their
customers. The presumed retail variables used to collect data in the study are
18 in number. Common customers of these three types of outlets are asked to
rate each variable applicable to these outlets in the light of their perception.
These eighteen variables are factored in to six factor variables by using Factor
Analysis Method with the help of SPSS. Then the mean scores of the
individual variables as well as factor variables (cumulated mean scores of
variables included in the factor) are computed. The mean scores of factor
variables so computed are used in the comparison of Triveni Stores with
Margin-Free Markets and other Private Supermarkets in Kerala.
Variables used in the study to compare the types of consumer stores are:
1) Price of food and groceries
2) Quality of food and groceries
3) Price of convenience goods
4) Quality of convenience goods
5) Customer personal care
6) Availability of fresh stock
7) Availability of brands
Chapter -7
366
8) Location advantages
9) Attractiveness in display
10) Space layout in store
11) Quality of Packing
12) Parking facilities.
13) Complaint redress mechanism
14) Speediness in billing
15) Discount available
16) Cleanliness in the shop
17) Facilities such as toilet, water ,seating facilities, etc
18) Working hours
Data about the above mentioned eighteen retail variables are collected at
the seven point scale for comparing Triveni Supermarkets with Margin-Free
Markets and other Private Supermarkets.
7.4 Factor Analysis Using the Selected Retail Variables of Private Supermarkets Private Supermarkets are retail outlets offering consumer goods and they
claim that they offer quality goods at lowest prices. In the study, they are
perceived to a type of consumer stores and do not mean a particular private
supermarket. Customers selected (Respondents) for collecting sample data
were common customers of the Private Supermarkets, Margin- Free Markets
and Triveni Supermarkets. The sampled customers claimed that they use the
three types of consumer outlets simultaneously for purchasing essential and
household items. The Exploratory Factor Analysis technique is used in the
study to reduce the number of variables for assessing the customer rating about
the three types of stores. Factor Analysis is a statistical technique designed to
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
367
represent a wide range of variables on a smaller number of dimensions. For
using this technique, Factor Correlation Matrix, Communalities Table, table
provides Eigen Values and Rotated Component Matrix are used. It has been
ensured from correlation matrix that eighteen retail variables are not correlated
each other.
Factor Analysis is a data reduction technique which identifies a small
number of factors that explain most of the variance observed in a large set of
variables. This is generally employed to generate hypothesis regarding causal
mechanism for a problem. Using the correlation structure, factor analysis is
employed to identify underlying hypothecated variables called factors that
explain the correlation pattern explained.
The KMO and Bartlett's Test is used for knowing the suitability of the
collected data for factor analysis.
Table 7.1 KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .710
Approx. Chi-Square 2920.730
Df 190
Bartlett's Test of Sphericity
Sig. .000 Source: Primary data.
Table 7.1 shows the results of KMO and Bartlett's Test. The Kaiser-
Mayer-Olkin Measure of Sampling Adequacy is a statistic that indicates the
proportion of variance in the variables that might be caused by underlying
factors. High values (close to 1.00) generally indicate that generally factor
analysis is useful with the data set and the values; less than 0.50 indicates that
the results of the factor analysis are not very apt in the context. The result of
KMO and Bartlett's Test indicates that the factor analysis is appropriate for
Chapter -7
368
the present data base, as the KMO value is.770 (between 0.50 and 1.00) and
statistical results of Bartlett’s test of sphericity are significant (where p=0.000
with df 190) for the factor correlations with in a correlation matrix. Small
values less than 0.05 of the significance level indicate that a factor analysis is
useful to a particular set of data set. From the KMO and Bartlett's Test, we can
observe that the significance level is 0.000 <0.05, indicating the data is
appropriate to conduct a factor analysis for the study.
The correlation matrix of the variables extracted is given below.
Comparative A
ssessment of Triveni Superm
arket, Margin – free M
arkets and Private…..
369
Chapter -7
370
The correlation matrix would help to assess the discriminant validity of
the exploratory tendency of the variables. The correlation matrix shown in
Table 7.2 depicts that no significant correlation exists between any of the
variables. The correlation co-efficient were far below the 0.85 cut off mark,
(Gerberg & Anderson.2008) 9 which was used as rule of thumb. This implies
that there is no conceptual overlapping among the variables, and discriminant
validity of the sub-scales is proved once again.
The output from analysis is viewed after validating the variables using the communalities.
Table 7.3 Communalities extracted for the retail variables of Private Supermarkets
Variables Initial Extraction Price of food and groceries 1.000 .791 Quality of food and groceries 1.000 .769 Price of convenience goods 1.000 .760 Quality of convenience goods 1.000 .768 Customer personal care 1.000 .523 Availability of fresh stock 1.000 .648 Availability of brands 1.000 .552 Location advantages 1.000 .567 Attractiveness in display 1.000 .606 Space layout in store 1.000 .616 Quality of Packing 1.000 .637 Parking facilities. 1.000 .675 Complaint redress mechanism 1.000 .482 Speediness in billing 1.000 .484 Discount available 1.000 .807 Cleanliness in shop 1.000 .636 Facilities such as toilet, water, seating facilities 1.000 .680 Working hours 1.000 .632 Extraction Method: Principal Component Analysis.
Source: Primary data
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
371
Since all extraction communalities are fairly large (>0.482) it is supposed
to be a good set of variables. The principal component extraction followed by
varimax rotation are reported in the following table.
Table 7.4 Total Variance Explained for retail variables of Private Supermarkets
Initial Eigen values Extraction Sums of Squared Loadings
The mean scores of retail factor variable ‘Price of Goods (5.46)’ of Triveni
Supermarkets and the mean score of ‘Quality of Goods’ (10.51) reveal that they
charge comparatively low prices and the quality of goods rated by the sample
customers is high. The customer rating for discount is also high (5.71) and it
indicates that they supply goods at a low price. The following hypothesis is
generated to recognize the variation of mean scores of factors among regions.
Hypothesis
H0: There is no difference in the means of retail factor variables of Triveni Supermarkets among the regions.
H1: There is difference in the means of retail factor variables of Triveni Supermarkets among the regions.
Chapter -7
378
For testing the hypothesis on mean differences of retail factor variables
one way ANOVA is used. Table 7.10 depicts whether region-wise variation
exists in the mean scores of factor retail variables of Triveni supermarkets.
Table 7.10 ANOVA table : Region-wise Variations of Retail Factor Variables of Triveni Supermarkets
Factors Sum of Squares df Mean
Square F Sig.
Between Groups 277.264 2 138.632 15.942 .000 Within Groups 3730.653 429 8.696
Service
Total 4007.917 431 Between Groups 393.394 2 196.697 24.749 .000 Within Groups 3409.486 429 7.948
Availablity And Ambience
Total 3802.880 431 Between Groups 160.907 2 80.454 26.485 .000 Within Groups 1303.201 429 3.038
Facilities
Total 1464.109 431 Between Groups 88.532 2 44.266 33.474 .000 Within Groups 567.319 429 1.322
Quality of Goods
Total 655.852 431 Between Groups 21.060 2 10.530 18.630 .000 Within Groups 242.486 429 .565
Discount
Total 263.546 431 Between Groups 94.181 2 47.090 28.007 .000 Within Groups 721.299 429 1.681
Price of Goods
Total 815.479 431 Source: Primary data
ANOVA table depicts Mean Square, F- values and the significant
levels (p values) of all the six factor variables considered. The null
hypothesis is rejected as p < .05 and the alternative hypothesis is accepted as
there exists significant difference in the means of all retail factor variables of
Triveni Supermarkets among the three regions. This shows that there is
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
379
significant difference in the performance of Triveni Stores among various
regions based on the responses of the respondents.
7.6 Factor Analysis Using the Selected Retail Variables of Margin- Free Supermarkets As per the opinion given by common customers of the three types of
enterprises, the selected retail variables of Margin- Free Markets are (rated by
customers) factored into six factors using factors generated for Private
Supermarkets as the reference. Retail factor variables of Margin-Free Markets
so obtained are given below and their mean values and standard deviations
ascertained are also given in the table 7.11.
Table 7.11 Descriptives of retail factor variables of Margin-Free Markets
As per the table 7.11, it is seen that the mean value of the retail factor variable
‘Price of Goods’ (10.14) was higher than that of the factor variable ‘Quality of
Goods’ (8.86) and the average rating for discount was also low. This led to the
conclusion that Margin-Free Markets charge comparatively higher prices for the
goods compared to the quality of goods that they offer. The comparative ratings
given by the selected customers for Margin-Free Markets led us to conclude that
the quality of goods and the price of goods offered by this type of stores do not
match. It was also found that these stores offer low discount as well as premium
Chapter -7
380
products to the customers. To identify the variation among the regions are
significant or not, the following hypothesis is framed
Hypothesis
H0: There is no difference in the means of retail factor variables of
Margin-Free Markets among the regions.
H1: There is difference in the means of retail factor variables of
Margin- Free Markets among the regions.
For validating the hypothesis, one way ANOVA is used.
Table 7.12 ANOVA table for testing region-wise variations of factor retail variables of Margin- Free Markets.
Factors Sum of Squares df Mean
Square F Sig.
Between Groups 61.014 6 10.169 1.239 .285 Within Groups 3093.643 420 8.206
Service
Total 3154.656 426 Between Groups 20.052 6 3.342 .467 .833 Within Groups 2699.604 420 7.161
Availability and Ambience
Total 2719.656 426 Between Groups 28.343 6 4.724 1.168 .323 Within Groups 1525.091 420 4.045
Facilities
Total 1553.435 426 Between Groups 9.736 6 1.623 1.205 .303 Within Groups 507.753 420 1.347
Quality of Goods
Total 517.490 426 Between Groups 48.954 6 8.159 3.204 .004 Within Groups 960.168 420 2.547
Price of Goods
Total 1009.122 426 Between Groups 5.672 6 .945 1.100 .362 Within Groups 323.953 420 .859
Discount
Total 329.625 426 Source: Primary data
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
381
As seen from the ANOVA table 7.12, by evaluating the, Mean Squares,
F values and sig. level of factor variables, all factor variables except
‘Price of Goods’, the null hypothesis is accepted as p > .05. It is inferred
that there exists no difference among the means of retail factor variables
‘Service’, ‘Availability and Ambience’, ‘Facilities’, ‘Quality of Goods’ and
‘Discount’ of Margin-free markets among the regions. This shows there is no
significant difference in the responses of customers relating to the five factors
mentioned above among the three regions. In the case of factor variable
‘Price of goods’, the null hypothesis is rejected as P<.05. Here the
alternative hypothesis is accepted as there is significant difference among the
means of this factor variable of Margin-Free Markets among the regions.
7.7 Comparative Assessment of Triveni Supermarkets, Margin- Free Markets and Private Supermarkets Using Selected Retail Factors For comparison, the mean scores and standard deviations of factor
variables of the three types of stores are considered. Mean scores obtained for
factor variables are used for drawing bar diagrams, which will lend clarity to
the comparison of the retail outlets. Repeated measure MANOVA will be
useful in this respect to test whether statistically significant differences
exist among the three types of retail outlets. Here two types of tests are
resorted to: 1) Test for knowing whether significant differences exist among
the means of retail factor variables for the three different types of retail outlets
and 2) Test for knowing whether significant differences exist among the
means of retail factors for the outlets on regions, income levels of customers
and the areas of residence of customers.
Chapter -7
382
Table 7.13 Group Statistics of Retail Factor Variables for the comparison of three types of supermarkets
Discount .945 21.155 2 .000 .948 .958 .500 Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to an identity matrix. a. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table. Source: Primary data
Mauchly's sphericity test examines the form of the common covariance
matrix. A spherical matrix has equal variances and covariances equal to zero.
The common covariance matrix of the transformed within-subject variables
must be spherical, or the F tests and associated p values for the univariate
approach to testing within-subjects hypotheses are invalid. If the Chi-square
approximation has an associated p value less than the alpha level, the
sphericity assumption has been violated. The chi-square approximation for
this test is significant at 5 per cent level for four out of six variables
considered. Since this is less than the alpha level of 0.05, we can be
confident that the data do not meet the sphericity assumption. Hence,
multivariate tests are used.
For practical purposes, these issues are important only in helping to
decide which output to use, and if the output should be adjusted. If we use
Chapter -7
392
the univariate output, we may have more power to reject the null hypothesis
in favor of the alternative hypothesis. However, the univariate approach is
appropriate only when the sphericity assumption is not violated. If the
sphericity assumption is violated, then in most situations it is better off
staying with the multivariate output. (It amounts to using Multivariate
output often).
7.8.1 Test for knowing whether significant difference exists among the means of retail factor variables for three different types of retail outlets Multivariate Approach to Within-Subjects Tests:
As noted above, the multivariate output is still valid even if the
sphericity assumption is not met. SPSS prints the multivariate approach to
testing the within-subjects factors after Mauchly's test of sphericity. The first
multivariate test of a within-subjects effect is the within-subjects main effect
test. It examines changes in factor variables as a function of type of retail
outlets considered. The null hypothesis is that the means of factor variables
do not change across different types of retail outlets considered.
Hypothesis:
H0: There is no difference in the means of factor variables for
different types of retail outlets.
H1: There is difference in the means of factor variables for
different types of retail outlets.
Comparative Assessment of Triveni Supermarket, Margin – free Markets and Private…..
393
Table 7.15 Multivariate Test for Within-Subjects Main Effect
Within Subjects Effect Value F Hypothesis df Error df Sig. Pillai's Trace 1.575 460.684 12.000 1492.000 .000 Wilks' Lambda .027 637.329a 12.000 1490.000 .000 Hotelling's Trace 13.987 867.209 12.000 1488.000 .000
factor1
Roy's Largest Root 12.121 1506.997b 6.000 746.000 .000 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + Region Within Subjects Design: factor1 d. Tests are based on averaged variables. Source: Primary data
There are four tests, each reporting a separate multivariate test statistic
(Pillai’s, Hotelling's, Wilks', and Roy's); among them the Wilks’ test is
commonly used. These multivariate statistics are converted to ‘ F’ values. In
some cases, the converted F and its degrees of freedom are approximations.
When this is not the case, a note at the bottom of the output states that the
statistics are exact.
Since, the F ratio for this hypothesis is very large [F(12,1490) =
637.329, P < .0001], we confidently reject the null hypothesis and conclude
that the means of factor variables change with type of retail outlets in the
population from which the sample was drawn. This is concluded by
stating that there exist significant differences in the means of all six retail
factor variables for the three different types of retail outlets.
7.8.2 Test for the hypothesis that regional differences interact with type of retail outlets considered.
Table 7.16 gives the region-wise mean scores and standard deviations of
six factor variables of Triveni Supermarkets, Margin-Free Markets and Private
Supermarkets.
Chapter -7
394
Table 7.16 Region-wise means and standard deviations of retail factor variables of the retail outlets.
Region
Triveni Supermarkets
Margin-free Supermarkets
Private
Supermarkets Retail factors
Mean
Std. Deviation Mean
Std. Deviation Mean
Std. Deviation
Central 24.0000 1.97183 21.3169 2.69059 25.9504 2.01893