03-08-2017 AMR PROJECT:

Dinshaw’s Ice Cream

GROUP - 8
Prateek Awasthi – P16003
Ataul Karim Baig – P16005
Sameer Bhajni – P16008
Kurian T R – P16054
Vinod Valecha – P16055
Sameer Kalamkar – P15018
Sameer Kalamkar – P15018

0|Page
Contents

Executive Summary...................................................................................................................... 2
Problem Definition....................................................................................................................... 3
Background to the Problem ............................................................................................................... 3
Statement of the Problem ................................................................................................................. 3
Initial Approach to the Problem ................................................................................................... 3
Research Design........................................................................................................................... 4
Type of Research Design .................................................................................................................... 4
Information Captured ........................................................................................................................ 4
Data Collection from Secondary Sources .......................................................................................... 4
Data Collection from Primary Sources............................................................................................... 5
Scaling Techniques ............................................................................................................................. 6
Questionnaire Development and Pretesting..................................................................................... 6
Sampling Techniques.......................................................................................................................... 6
Fieldwork ............................................................................................................................................ 7
Data Analysis ............................................................................................................................... 7
Results......................................................................................................................................... 9
Limitations................................................................................................................................. 15
Conclusion and Recommendations ............................................................................................. 15
Exhibits...................................................................................................................................... 17
References................................................................................................................................. 31
Individual group member contribution ....................................................................................... 32

1|Page
 Executive Summary

Dinshaw’s ice cream is one of the major ice cream manufacturers in Central India and is based
out of Nagpur. The managerial decision problem was to increase the company’s market share
in ice cream segment. Based on this, the marketing research problem was formulated to
determine brand attributes which influences the consumer’s choice for ice cream purchase.
Secondary data was obtained through literature review and by conducting FGDs. A
questionnaire was floated to respondents in age group 16 to 45 years across different
geographies. Following statistical techniques were used to analyse the data obtained through
the survey:
1. Factor Analysis (To group features)
2. Discriminant Analysis (For Attribute based perceptual mapping and to understand the
factors that influence price sensitivity, recommendation, repurchase behaviour)
3. Multidimensional Scaling (To understand positioning of Dinshaw’s in comparison to other
brands)
4. ANOVA (To understand the relationship between demographics v/s evaluation of
Dinshaw’s ice cream, price sensitivity, repurchase behaviour, satisfaction level x, and
recommendation each individually v/s evaluation of Dinshaw’s ice cream)
From our analysis, we recommend the following among others:
1. Adequate Serving Volume is a significant variable in predicting the price sensitivity of the
consumers.
2. Good Taste is a significant variable in predicting the likelihood of people recommending
Dinshaw’s ice cream to others.
3. Good Taste is a significant variable in predicting the likelihood of people repurchasing
Dinshaw’s ice cream.

2|Page
 Problem Definition
Background to the Problem
Dinshaw’s Dairy Foods Ltd. is one of the best-known ice creams brands in Central India. The
management is concerned about the growing competition from brands that are well known in
the country and about the perception of the consumers towards Dinshaw’s Ice cream.

Statement of the Problem

Management Decision Problem: What should Dinshaw’s do to increase its market share?
Marketing Research Problem: To determine brand attributes that influence consumer choices
for ice cream purchase.

Initial Approach to the Problem

Based on the marketing research problem defined above, the following research questions and
their associated hypotheses are formulated:
Research Questions and hypotheses:
RQ1: Do consumers exhibit loyalty towards Dinshaw’s Ice cream?
H1: Consumers of Dinshaw’s Ice cream are loyal.
RQ2: When do people prefer eating ice cream?
H2.1: People prefer eating ice creams during festivities.
H2.2: People prefer eating ice creams during summers only.
RQ3: Are consumers satisfied with Dinshaw’s ice cream quality?
H3.1: Consumers are satisfied with Dinshaw’s ice cream quality.
H3.2: Consumer satisfaction are not dependent on quality.
RQ4: Do consumers find Dinshaw’s ice cream affordable?
H4: Dinshaw’s ice is affordable.
RQ5: Whether wide variety in ice cream products matters to the consumers?
H5: Wide variety of ice creams matter to the consumers
RQ6: Does price influence the purchasing decision of ice cream?
H6: Ice cream purchase decision is influenced by price

These were a few broad level questions which we formed prior to undertaking primary research
design. These questions where further expanded based on the findings from FGDs and
literature review. Post this, the questionnaire was designed, pretested and floated.

3|Page
 Research Design
Type of Research Design
The Research Design used is Conclusive in nature to assist Dinshaw’s to select the best course
of action to take in the current situation. This research is quantitative in nature that will be used
in the managerial decision making process. The conclusive research design is of descriptive
type and single cross-sectional in nature.

Information Captured
The following information was captured:
a) Demographics (Age, Gender, Disposable Income, City, Leisure Time, Education Level)
b) Brand Satisfaction & Loyalty
c) Price Sensitivity
d) Consumer Preferences in terms of brand, packaging, spending, time, place of purchase.
e) Extent of influence of various ice cream attributes on consumers
f) Consumer perception across popular brands

Data Collection from Secondary Sources

Literature Review Findings:
a. Significant sales happen during summer months of April-July while sales dip during
winter months of November-February. Other factors that affect the sales is big way are
the events like marriages, festivals, birthdays, etc.
b. Purchases are driven by impulses; hence, visibility forms a significant part of purchase.
Other factors that influence purchase on outlets are the brand, variety, packaging,
flavours and availability. Bars are the most favoured type of servings followed by
scoops, tubs, and cones.
c. Willingness to pay varies from Rs. 30 for Swirl ice cream to up to Rs. 150 for premium
ice cream.
d. Segmentation: The market is segmented based:
 Age
o Kids & Babies: those aged 0-9 years old
o Tweens and Teens: those aged 10-15 years old
o Early young adults: those aged 16-24 years old
o Older young Adults: those aged 25-34 years old
o Pre-Mid-Lifers: those aged 35-44 years old

4|Page
 o Mid-Lifers: those aged 45-54 years old
o Older Consumers: those aged 55+
 Busy Lives
o Time Rich: average leisure time of more than 7 hours per day
o Time to spare: average leisure time between 5 and 7 hours per day
o Time pressed: average leisure time between 3 and 5 hours per day
o Time poor: average leisure time between 1 and 3 hours per day
o No Time: average leisure time less than 1 hour per day
 Education Level
o Pre-Primary: Kids starting Schooling
o Primary: Kids between 5 to 11 years old
o Lower Secondary: basic formal education
o Upper Secondary: More specialized education - diploma
o Post-Secondary-Non-Tertiary: Equivalent to college
o Tertiary (1st Stage): Under Graduate
o Tertiray (2nd stage): Post Graduate
 Gender
o Male
o Female
 Urban or Rural Dweller
o Urban
o Rural
 Wealth
o Highly Affluent: Annual income from 93rd percentile onwards
o Better off: Household with income from 50th centile to 93rd centile
o Moderate Income: Household with income from 22.5th centile to 50th centile

Data Collection from Primary Sources

FGD Findings:
Two FGDs were conducted consisting of six members each. One group consisted of all male
members (with one member from Nagpur region) while the other group had two female and
four male members with nobody from Nagpur region. From the FGDs, below are our findings:

5|Page
 a. Most of the participants considered ice cream as a post dinner treat.
b. Top of the mind keywords that the participants associated with were ‘made of fresh
milk’, ‘natural’, ‘fresh’.
c. Amul, Kwality Walls, Baskin Robbins, Dinshaw’s were the brands mentioned by the
participants
d. The maximum price that the group was willing to pay in a single visit was Rs. 150.
e. The group unanimously considered ice cream purchase to be an impulsive decision.
f. Cones were the most preferred medium followed by cups. Cones were perceived to be
value for money.
g. Association of color with the flavour was observed
h. There was no clear consensus on the consumption pattern with respect to season.
Participants preferred having ice cream all the year round.

Scaling Techniques
The scaling technique used to value Dinshaw’s for various variables, a 5-point Likert scale has
been used, 1 being “strongly disagree” to 5 being “strongly agree”. To calculate the distance
between brands based on consumer perception, a 9-point Likert scale has been used, with 1
being close and 9 being far apart. To calculate the attributes of the different brands, a 7-point
Likert scale has been used with 1 being the worst and 7 being the best.

Questionnaire Development and Pretesting

The questionnaire was formed after going through the FGD findings and the secondary
research, taking into consideration all the findings as well as the problem statement provided
by the company, covering all the parameters.
The questionnaire was then pretested and edited based on the problems faced during pretesting.

Sampling Techniques
1. Target Population
Element: Ice cream buyers above 16 years of age
Sampling Unit: Individuals
Extent: Nagpur Region (in particular) and other cities
Time: July 2017
2. Sampling frame
Randomly select participants

6|Page
 3. Sampling Technique
Simple Random Sampling Technique
4. Sampling Size
The sample size for the survey was 143.
5. Execution
Sampling was done via online survey which was sent to the respondents in Nagpur
region.

Fieldwork
The questionnaire was prepared on QuestionPro and after pretesting and making suitable
changes, the same was floated to the potential respondents.

Data Analysis
The techniques used for data analysis are as mentioned below:
1. Factor Analysis

Factor analysis has been used to evaluate Dinshaw’s ice cream based on features by grouping
them into factors. We determined a reduced set of factors which influence the consumer’s
evaluation of Dinshaw’s ice-cream and thus tried to suggest improvements in variables within
the factors which, if changed, would influence the evaluation of Dinshaw’s cream for
consumers.

As part of factor analysis, the variables mentioned below were considered:

a) Ease of ordering
b) Flavour
c) Taste
d) Hygiene
e) Experience
f) Availability
g) Price
h) Customization
i) Packaging
j) Serving volume

7|Page
k) Variety
l) Freshness

The variables are clubbed together to form factors. The rules used to reduce the factors are:

a) Eigen Value should be greater than 1

b) KMO measure of sample adequacy should be in the range of 0.5 to 1
c) Bartlett’s Test of Sphericity : p-value < 0.05
d) To measure the reliability in terms of measuring the factors, we used Cronbach’s Alpha
whose values must be greater than 0.7.

2. Discriminant Analysis
Discriminant analysis is used to predict whether a new customer would be satisfied with
Dinshaw’s and to predict if a price increase in Dinshaw’s ice creams would affect the buying
behaviour of the consumer (price sensitivity). Discriminant analysis would help answer
question relating, how in terms of characteristics the customers who repurchase Dinshaw’s ice
cream differ from that of non-repurchase buyers (brand loyalty).

The rules used to verify the statistical validity of the attribute scores are:

a) Eigen Value should be greater than 1 for the functions.

b) Wilks’ Lambda should be less than 0.5.

3. Multidimensional Scaling
To derive the perceptual map for the brands and attributes, multidimensional scaling has been
used. With this, we tried to analyse where the brands lie and on what attributes the brands are
positioned in the minds of consumers.

The rules used to verify the statistical validity of the number of dimensions are:

1. Kruskal’s Stress Value should be less than 0.15.

2. Goodness of Fit (R Squared) should be greater than 0.7.

8|Page
 Results

The results from the analysis of data is as follows:

1. Factor Analysis
The result of factor analysis is as shown in the table below:
Factors Variables Criteria
Customization
Readily Available
Eigen Value = 4.212 (>1) [Exhibit 4]
Wide Range of Packaging
Store KMO  0.677 ( 0.5 – 1) [Exhibit 1]
Hygienic Store
Experience Bartlett’s Test of Sphericity = 0.000 ( < 0.05)
Flavour
Cronbach Alpha = 0.711 (>0.7) [Exhibit 2]
Ease of Ordering
Experience
Eigen Value = 1.504 (>1) [Exhibit 4]
Value for Adequate Volume KMO  0.677 ( 0.5 – 1) [Exhibit 1]
Money Reasonably Priced Bartlett’s Test of Sphericity = 0.000 ( < 0.05)
Cronbach Alpha = 0.711 (>0.7) [Exhibit 2]

Eigen Value = 1.365 (>1) [Exhibit 4]

Good Taste
KMO  0.677 ( 0.5 – 1) [Exhibit 1]
Quality Fresh Ice cream
Bartlett’s Test of Sphericity = 0.000 ( < 0.05)
Attractive Packaging
Cronbach Alpha = 0.711 (>0.7) [Exhibit 2]

The result implies that there are three underlying dimensions or factors that explain the
correlation among the set of variables. These underlying variables were grouped into three
factors Store Experience, Value for Money, Quality which had emphasis for consumers when
they evaluated Dinshaw’s Ice cream.

2. Discriminant Analysis
The multiple discriminant analysis, used to predict the association of member to the group. The
first function generated by the discriminant analysis accounted for the greatest proportion of
differences among the groups while the second function that is generated accounted for a
decreasing proportion of the differences among the groups.

a. Discriminant Analysis for Price Sensitivity

This was done to identify the parameters that influences the price sensitivity. Following were
the results obtained:
Following were the results obtained:
 Two functions were derived for which Wilks’ Lambda values were 0.439 for function 1
and 0.877 for function 2 [Exhibit 9].
 Also, Eigen Values were less than 1 (0.997 and 0.141) [Exhibit 10]

9|Page
Discriminant Model (unstandardized) for the Function1 for Price Sensitivity [Exhibit 11] is:

Z1 = 0.187*wide range of flavors -0.012*Ease of ordering + 0.938*Good taste –

0.524*hygienic store + 0.346*good experience +0.241*not fresh ice cream – 0.262*readily
available + 0.108*customization of toppings + 0.223*wide range of packaging – 0.211*not
attractive packaging + 1.351*adequate volume per serving – 0.764*reasonably priced – 5.716

81.2% of the cases were correctly classified by this model [Exhibit 12].

Function 1 discriminates 1.0 from 2.0 & 3.0.

Where:
1.0 is the group of consumers that would definitely purchase Dinshaw’s ice cream after the
price increase.
2.0 is the group of consumers that might or might not purchase Dinshaw’s ice cream after the
price increase.
3.0 is the group of consumers that would definitely not purchase Dinshaw’s ice cream after the
price increase.
1.0 scored at the positive end on the bipolar function and 2.0 and 3.0 at the negative end of the
function.

Discriminant Model (standardized) to find the best predictor variable [Exhibit 13] is

Z1 = 0.167*wide range of flavors -0.009*Ease of ordering + 0.713*Good taste –

0.413*hygienic store + 0.276*good experience + 0.232*not fresh ice cream – 0.232*readily
available + 0.099*customization of toppings + 0.210*wide range of packaging – 0.222*not
attractive packaging + 1.07*adequate volume per serving – 0.647*reasonably priced

From the above model the factor with the highest loading is the best predictor variable i.e. it
has the highest contribution in predicting the outcome.
Therefore, Adequate volume per serving in this model significantly explains the prediction for
price sensitivity.

Discriminant Model (unstandardized) for the Function2 for Price Sensitivity [Exhibit 11] is:

Z2 = 0.267*wide range of flavors + 0.146*Ease of ordering - 0.0838*Good taste –

0.295*hygienic store + 0.528*good experience +0.021*not fresh ice cream – 0.562*readily
available + 0.335*customization of toppings + 0.658*wide range of packaging + 0.052*not
attractive packaging - 0.117*adequate volume per serving – 0.011*reasonably priced – 0.229

10 | P a g e
Function 2 discriminates 2.0 from 1.0 & 3.0.
Where:
1.0 is the group of consumers that would definitely purchase Dinshaw’s ice cream after the
price increase.
2.0 is the group of consumers that might or might not purchase Dinshaw’s ice cream after the
price increase.
3.0 is the group of consumers that would definitely not purchase Dinshaw’s ice cream after the
price increase.
1.0 & 3.0 scored at the positive end on the bipolar function and 2.0 at the negative end of the
function.

b. Discriminant Analysis for People’s Recommendation

This was done to identify the parameters that influences the people who will recommend
Dinshaw’s to others. Following were the results obtained:
Following were the results obtained:
 Two functions were derived for which Wilks’ Lambda values were 0.261 for function 1
and 0.695 for function 2 [Exhibit 14].
 Also, Eigen Values were greater than 1 for function 1 (1.663) and less than 1 for function
2 (0.439) [Exhibit 15]
Discriminant Model (un-standardized) for Function 1 of People’s Recommendation [Exhibit
16] is

Z1 = -0.072*wide range of flavors - 0.305*Ease of ordering + 1.495*Good taste –

.008*hygienic store + 0.322*good experience -0.184*not fresh ice cream – 0.199*readily
available -0.016*customization of toppings – 0.518*wide range of packaging – 0.13*not
attractive packaging + 1.53*adequate volume per serving + 0.179*reasonably priced – 5.871

84.1% of the cases were correctly classified by this model [Exhibit 17].

Function 1 discriminates 3.0 from 1.0 & 2.0.

Where:
1.0 is the group of consumers that would definitely recommend Dinshaw’s ice cream to others.
2.0 is the group of consumers that might or might not recommend Dinshaw’s ice cream to
others.
3.0 is the group of consumers that would definitely not recommend Dinshaw’s ice cream to
others.
1.0 & 2.0 scored at the positive end on the bipolar function and 3.0 at the negative end of the
function.

11 | P a g e
 Discriminant Model (standardized) to find the best predictor variable [Exhibit 18] is

Z1 = -0.064*wide range of flavors +0.228*Ease of ordering + 0.915*Good taste +

0.006*hygienic store + 0.256*good experience – 0.168*not fresh ice cream – 0.65*readily
available – 0.015*customization of toppings – 0.496*wide range of packaging – 0.013*not
attractive packaging + 0.127*adequate volume per serving + 0.144*reasonably priced

From the above model the factor with the highest loading is the best predictor variable i.e. it
has the highest contribution in predicting the outcome.
Therefore, Good Taste in this model significantly explains the prediction if people are likely to
recommend Dinshaw’s.

Discriminant Model (un-standardized) for Function 2 of People’s Recommendation [Exhibit

16] is

Z2 = -0.436*wide range of flavors + 0.601*Ease of ordering - 0.456*Good taste +

.129*hygienic store - 0.006*good experience -0.211*not fresh ice cream + 0.882*readily
available -0.165*customization of toppings + 0.332*wide range of packaging – 0.392*not
attractive packaging + 1.00*adequate volume per serving – 1.076*reasonably priced – 1.361

Function 2 discriminates 2.0 from 1.0 & 3.0.

Where:
1.0 is the group of consumers that would definitely recommend Dinshaw’s ice cream to others.
2.0 is the group of consumers that might or might not recommend Dinshaw’s ice cream to
others.
3.0 is the group of consumers that would definitely not recommend Dinshaw’s ice cream to
others.
1.0 & 3.0 scored at the positive end on the bipolar function and 2.0 at the negative end of the
function.

c. Discriminant Analysis for Repurchase Behaviour

This was done to identify the parameters that influences the people to repurchase Dinshaws
Ice Cream
Following were the results obtained:
 Two functions were derived for which Wilks’ Lambda values were 0.294 for function 1
and 0.797 for function 2 [Exhibit 19].
 Also, Eigen Values were greater than 1 for function 1 (1.714) and less than 1 for function
2 (0.254) [Exhibit 20]

12 | P a g e
Discriminant Model (un-standardized) for Function 1 of Repurchase behaviour [Exhibit 21] is

Z1 = 0.492*wide range of flavors + 0.186*Ease of ordering + 1.377*Good taste –

0.240*hygienic store + 0.418*good experience -0.118*not fresh ice cream – 0.834*readily
available -0.230*customization of toppings – 0.527*wide range of packaging – 0.199*not
attractive packaging – 0.311*adequate volume per serving + 0.687*reasonably priced – 4.285

84.1% of the cases were correctly classified by this model [Exhibit 22].

Function 1 discriminates 3.0 from 1.0 & 2.0.

Where:
1.0 is the group of consumers that would definitely repurchase Dinshaw’s ice cream.
2.0 is the group of consumers that might or might not repurchase Dinshaw’s ice cream.
3.0 is the group of consumers that would definitely not repurchase Dinshaw’s ice cream.
1.0 & 2.0 scored at the positive end on the bipolar function and 3.0 at the negative end of the
function.

Discriminant Model (standardized) to find the best predictor variable [Exhibit 23] is

Z= 0.431*wide range of flavors + 0.149*Ease of ordering + 0.920*Good taste –

0.187*hygienic store + 0.341*good experience – 0.112*not fresh ice cream – 0.710*readily
available – 0.220*customization of toppings – 0.498*wide range of packaging + 0.214*not
attractive packaging - 0.264*adequate volume per serving + 0.556*reasonably priced

From the above model the factor with the highest loading is the best predictor variable i.e. it
has the highest contribution in predicting the outcome.
Therefore, Good Taste in this model significantly explains the prediction for consumers
repurchase behaviour.

Discriminant Model (un-standardized) for Function 2 of Repurchase behaviour [Exhibit 21] is

Z2 = -0.508*wide range of flavors + 0.033*Ease of ordering + 0.527*Good taste +

0.172*hygienic store - 0.047*good experience -0.175*not fresh ice cream + 0.832*readily
available -0.094*customization of toppings + 0.396*wide range of packaging – 0.071*not
attractive packaging + 0.974*adequate volume per serving – 1.012*reasonably priced – 4.26

Function 2 discriminates 2.0 from 1.0 & 3.0.

Where:
1.0 is the group of consumers that would definitely repurchase Dinshaw’s ice cream.
2.0 is the group of consumers that might or might not repurchase Dinshaw’s ice cream.
3.0 is the group of consumers that would definitely not repurchase Dinshaw’s ice cream.

1.0 & 3.0 scored at the positive end on the bipolar function and 2.0 at the negative end of the
function.
13 | P a g e
d. Discriminant Analysis for Attribute Based Perceptual Mapping
As part of Attribute Based Perceptual Mapping, the respondents were asked to rate five
brands (Amul, Kwality Walls, Vadilal, Dinshaw’s, and Baskin Robins) against four
attributes (Price, Quality, Variety, Availability).
Following were the results obtained:
 Four functions were derived for which Wilks’ Lambda values were 0.701, 0.908, 0.971,
and 0.997 [Exhibit 24]. Since, the values were greater than 0.5, the results are not
mathematically significant for Attribute Based Perceptual Mapping
 Also, Eigen Values were less than 1 (0.295, 0.069, 0.027, 0.003) [Exhibit 25]
The test was then conducted for the respondents only from Nagpur region and for different
combinations of brands, the results were still inconclusive.

3. Multidimensional Scaling
Kruscal’s Stress = 0.3454 (>0.15) and R Squared = 0.59006 (<0.7). Hence, it is not statistically
acceptable solution for Multidimensional Scaling. [Exhibit 26]

4. ANOVA
ANOVA results for following set of combinations showed no significance against the
Demographic parameters:
The demographic included Age, Income, Gender and educational level of the participant. While
doing the ANOVA our
H0: The means across the demographic group for the behaviour to be tested is the same.
HA: The means across are not the same.
There were 3 levels: recommendation to others, repurchase behaviour and price sensitivity.
a) Demographic vs recommendation to others
b) Demographic vs repurchase behaviour
c) Demographic vs price sensitivity
Since significance value was greater than 5%, we couldn’t reject null hypothesis that there was
no impact of demographic against each of the above parameters. [Exhibit 27]
ANOVA results for following set of combinations showed no significance against the
Demographic parameters:
To check for the significance of the group means across the group, ANOVA for
1. Evaluation of Dinshaw’s Icecream vs. recommend to other
2. Evaluation of Dinshaw’s Icecream vs. repurchase behaviour
3. Evaluation of Dinshaw’s Icecream vs. price sensitivity
4. Evaluation of Dinshaw’s Icecream vs. overall satisfaction

14 | P a g e
There were 3 groups in each of the following: recommend to other, repurchase behaviour, price
sensitivity and 2 groups for overall satisfaction while there were 12 questions for evaluation of
Dinshaw’s Ice cream.
Impacting Variables
Recommend to other Range of Flavours, Ease of Ordering, Good
Taste, Store Experience, Readily Available,
Serving Volume and Reasonably Priced
Repurchase Behaviour Range of Flavours, Good Taste, Store
Experience, Serving Volume and
Reasonably Priced
Price Sensitivity Range of Flavours, Good Taste, Store
Experience, Customization, Serving Volume
Overall Satisfaction Range of Flavours, Good Taste

For the rest of the variables, p-value >0.05 so we failed to reject the H0, i.e. we cannot say that
the group means are statistically significantly different. [Exhibit 28]
Thus focusing on the Impacting variable that drives Dinshaw’s ice cream performance in that
specific region of question is recommended.

Limitations
The limitation during this research were as follows:
1. Time Constraint
2. Drop Out Rate
3. Inaccessibility to company-owned outlets
4. Organizational Constraint : Inability to get appointment for expert interview

Conclusion and Recommendations

1. Adequate Serving Volume is a significant variable in predicting the price sensitivity of

the consumers.
2. Good Taste is a significant variable in predicting the likelihood of people
recommending Dinshaw’s ice cream to others.
3. Good Taste is a significant variable in predicting the likelihood of people repurchasing
Dinshaw’s ice cream.

Thus, our recommendation for Dinshaw’s to drive repurchase decision of consumers and
recommending others to consume Dinshaw’s is to focus on the Taste aspect of the ice cream.
Also, even if Dinshaw’s ice cream increases its price by 10% and maintains an adequate serving
volume, the consumers will still be willing to buy Dinshaw’s ice cream.

15 | P a g e
From factor analysis, Factor Store experience had the highest Eigen value, thus it contributed
the most towards evaluation that the customer gave towards ice-cream followed by Value for
Money and Quality. In store experience, the variable which had high loading was the
customization of topping for the ice cream. Thus, it is recommended that for enhancing the
store experience with more variety of topping would enhance customers future evaluation of
Dinshaw’s.

16 | P a g e
 Exhibits

A. FACTOR ANALYSIS EXHIBITS

Exhibit 1: KMO and Bartlett’s Test

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .677

Bartlett's Test of Sphericity Approx. Chi-Square 288.944

df 66

Sig. .000

Exhibit 2: Cronbach Alpha Values

Reliability Statistics

Cronbach's
Alpha Based on
Cronbach's Standardized
Alpha Items N of Items

.711 .732 12

Exhibit 3: Scree Plot

17 | P a g e
Exhibit 4 : Total Variance Explained

Total Variance Explained

Extraction Sums of Squared Rotation Sums of Squared

Initial Eigenvalues Loadings Loadings

% of Cumulative % of Cumulative % of Cumulative

Component Total Variance % Total Variance % Total Variance %

1 4.212 35.103 35.103 4.212 35.103 35.103 3.262 27.187 27.187

2 1.504 12.537 47.640 1.504 12.537 47.640 2.136 17.804 44.991
3 1.365 11.378 59.018 1.365 11.378 59.018 1.683 14.027 59.018
4 .957 7.978 66.996
5 .775 6.458 73.454
6 .691 5.758 79.213
7 .606 5.047 84.259
8 .596 4.966 89.225
9 .517 4.312 93.537
10 .357 2.975 96.512
11 .304 2.531 99.043
12 .115 .957 100.000

Extraction Method: Principal Component Analysis.

Exhibit 5: Rotated Component Matrix

Rotated Component Matrixa

Component

1 2 3

Wide Range of Flavor .671 .158 -.234

Ease of Ordering .665 .279 .061
Good Taste .422 .499 -.532
Hygenic Store .625 .075 .212
Good experience .553 .066 -.234
Not fresh ice cream -.011 -.342 .715
Readily Available .681 .112 -.167
Customization of topping and
.719 .005 -.241
mix
wide range of packaging .652 .151 -.013
not attractive packaging -.131 .196 .801
one serving is of adequate
.103 .905 -.042
volume
reasonably priced .252 .875 -.044
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.a

18 | P a g e
a. Rotation converged in 5 iterations.

Exhibit 6: Component Plot in Rotated Space

B. DETERMINANT ANALYSIS EXHIBITS

Exhibit 7: Eigen Values

Eigenvalues
Function Eigenvalue % of Variance Cumulative % Canonical Correlation
1 .295a 74.8 74.8 .477
2 .069a 17.5 92.3 .254
3 .027a 6.9 99.3 .163
4 .003a .7 100.0 .053
a. First 4 canonical discriminant functions were used in the analysis.

Exhibit 8: Wilks’ Lambda

Wilks' Lambda

19 | P a g e
Test of Function(s) Wilks' Lambda Chi-square df Sig.
1 through 4 .701 113.437 16 .000
2 through 4 .908 30.869 9 .000
3 through 4 .971 9.517 4 .049
4 .997 .913 1 .339

C. DISCRIMINANT ANALYSIS FOR PRICE SENSITIVITY

Exhibit 9: Wilks’ Lambda

Wilks' Lambda

Test of Function(s) Wilks' Lambda Chi-square df Sig.

1 through 2 .439 49.805 24 .001

2 .877 7.966 11 .716

Exhibit 10: Eigenvalues

Eigenvalues

Canonical
Function Eigenvalue % of Variance Cumulative % Correlation

1 .997a 87.6 87.6 .707

2 .141a 12.4 100.0 .351

a. First 2 canonical discriminant functions were used in the analysis.

Exhibit 11: Canonical Discriminant Function Coefficients

Canonical Discriminant Function Coefficients

Function

1 2

Wide Range of Flavor .187 .267

Ease of Ordering -.012 .146
Good Taste .938 -.838
Hygenic Store -.524 -.295
Good experience .346 .528
Not fresh ice cream .241 .021
Readily Available -.262 -.562
Customization of topping
.108 .335
and mix
wide range of packaging .223 .658
not attractive packaging -.211 .052
one serving is of adequate
1.351 -.117
volume

20 | P a g e
reasonably priced -.764 -.011
(Constant) -5.716 -.229

Unstandardized coefficients

Exhibit 12: Classification Results

Classification Resultsa

If the price of Dinshaw's ice Predicted Group Membership

cream had been 10%
higher, how likely are you to
purchase the ice cream? 1.0 2.0 3.0 Total

Original Count 1.0 35 1 2 38

2.0 3 5 2 10

3.0 3 2 16 21

% 1.0 92.1 2.6 5.3 100.0

2.0 30.0 50.0 20.0 100.0

3.0 14.3 9.5 76.2 100.0

a. 81.2% of original grouped cases correctly classified.

Exhibit 13: Standardized Canonical Discriminant Function Coefficients

Standardized Canonical Discriminant Function
Coefficients

Function

1 2

Wide Range of Flavor .167 .240

Ease of Ordering -.009 .117
Good Taste .713 -.637
Hygenic Store -.413 -.232
Good experience .276 .422
Not fresh ice cream .232 .021
Readily Available -.232 -.497
Customization of topping
.099 .308
and mix
wide range of packaging .210 .619
not attractive packaging -.222 .055
one serving is of adequate
1.070 -.092
volume
reasonably priced -.647 -.009

21 | P a g e
 D. DISCRIMINANT ANALYSIS FOR USER’s RECOMMENDATION

Exhibit 14: Wilks’ Lambda

Wilks' Lambda

Wilks'
Test of Function(s) Lambda Chi-square df Sig.

1 through 2 .261 81.267 24 .000

2 .695 22.011 11 .024

Exhibit 15: Eigenvalues

Eigenvalues

% of Cumulative Canonical
Function Eigenvalue Variance % Correlation

1 1.663a 79.1 79.1 .790

2 .439a 20.9 100.0 .552

a. First 2 canonical discriminant functions were used in the

analysis.

Exhibit 16: Canonical Discriminant Function Coefficients

Canonical Discriminant Function

Coefficients

Function

1 2

Wide Range of
-.072 -.436
Flavor
Ease of Ordering .305 .601
Good Taste 1.495 -.456
Hygenic Store .008 .129
Good experience .322 -.006
Not fresh ice cream -.184 -.211
Readily Available -.199 .882
Customization of
-.016 -.165
topping and mix
wide range of
-.518 .332
packaging

22 | P a g e
not attractive
-.013 -.392
packaging
one serving is of
.153 1.000
adequate volume
reasonably priced .179 -1.076
(Constant) -5.871 -1.361

Unstandardized coefficients

Exhibit 17: Classification Results

Classification Resultsa

How likely would you Predicted Group Membership

recommend
Dinshaw's Ice-cream
to a friend or
relative? 1.0 2.0 3.0 Total

Original Count 1.0 38 4 0 42

2.0 5 14 1 20

3.0 0 1 6 7

% 1.0 90.5 9.5 .0 100.0

2.0 25.0 70.0 5.0 100.0

3.0 .0 14.3 85.7 100.0

a. 84.1% of original grouped cases correctly classified.

Exhibit 18: Standardized Canonical Discriminant Function Coefficients

Standardized Canonical Discriminant

Function Coefficients

Function

1 2

Wide Range of
-.064 -.387
Flavor
Ease of Ordering .228 .450
Good Taste .915 -.279
Hygenic Store .006 .099
Good experience .256 -.005
Not fresh ice cream -.168 -.193
Readily Available -.165 .735

23 | P a g e
Customization of
-.015 -.157
topping and mix
wide range of
-.496 .319
packaging
not attractive
-.013 -.406
packaging
one serving is of
.127 .831
adequate volume
reasonably priced .144 -.867

E. DISCRIMINANT ANALYSIS FOR REPURCHASE

Exhibit 19: Wilks’ Lambda

Wilks' Lambda

Wilks'
Test of Function(s) Lambda Chi-square df Sig.

1 through 2 .294 74.105 24 .000

2 .797 13.690 11 .251

Exhibit 20: Eigenvalues

Eigenvalues

% of Cumulative Canonical
Function Eigenvalue Variance % Correlation

1 1.714a 87.1 87.1 .795

2 .254a 12.9 100.0 .450

a. First 2 canonical discriminant functions were used in the

analysis.

24 | P a g e
Exhibit 21: Canonical Discriminant Function Coefficients

Canonical Discriminant Function

Coefficients

Function

1 2

Wide Range of
.492 -.508
Flavor
Ease of Ordering .186 .033
Good Taste 1.377 .527
Hygenic Store -.240 .172
Good experience .418 -.047
Not fresh ice cream -.118 -.175
Readily Available -.834 .832
Customization of
-.230 -.094
topping and mix
wide range of
-.527 .396
packaging
not attractive
.199 -.071
packaging
one serving is of
-.311 .974
adequate volume
reasonably priced .687 -1.012
(Constant) -4.285 -4.260

Unstandardized coefficients

Exhibit 22: Classification Results

Classification Resultsa

How likely are you to Predicted Group Membership

purchase Dinshaw's
Ice cream again? 1.0 2.0 3.0 Total

Original Count 1.0 48 2 0 50

2.0 9 3 0 12

3.0 0 0 7 7

% 1.0 96.0 4.0 .0 100.0

2.0 75.0 25.0 .0 100.0

3.0 .0 .0 100.0 100.0

25 | P a g e
a. 84.1% of original grouped cases correctly classified.

Exhibit 23: Standardized Canonical Discriminant Function Coefficients

Standardized Canonical Discriminant

Function Coefficients

Function

1 2

Wide Range of
.431 -.445
Flavor
Ease of Ordering .149 .027
Good Taste .920 .352
Hygenic Store -.187 .134
Good experience .341 -.038
Not fresh ice cream -.112 -.166
Readily Available -.710 .708
Customization of
-.220 -.090
topping and mix
wide range of
-.498 .373
packaging
not attractive
.214 -.076
packaging
one serving is of
-.264 .826
adequate volume
reasonably priced .556 -.818
F. DISCRIMINANT ANALYSIS FOR ATTRIBUTE BASED PERCEPTUAL
MAPPING
Exhibit 24: Wilks’ Lambda

Wilks' Lambda
Test of Function(s) Wilks' Lambda Chi-square df Sig.
1 through 4 .701 113.437 16 .000
2 through 4 .908 30.869 9 .000
3 through 4 .971 9.517 4 .049
4 .997 .913 1 .339

26 | P a g e
Exhibit 25: Eigenvalues

Eigenvalues
Canonical
Function Eigenvalue % of Variance Cumulative % Correlation
1 .295a 74.8 74.8 .477
2 .069a 17.5 92.3 .254
3 .027a 6.9 99.3 .163
4 .003a .7 100.0 .053
a. First 4 canonical discriminant functions were used in the analysis.

G. MULTIDIMENSIONAL SCALING EXHIBITS

Exhibit 26 : Kruskal’s Stress and R-squared

Alscal Procedure Options

Data Options-

Number of Rows (Observations/Matrix). 5

Number of Columns (Variables) . . . 5
Number of Matrices . . . . . . 1
Measurement Level . . . . . . . Ratio
Data Matrix Shape . . . . . . . Symmetric
Type . . . . . . . . . . . Dissimilarity
Approach to Ties . . . . . . . Leave Tied
Conditionality . . . . . . . . Matrix
Data Cutoff at . . . . . . . . .000000

Model Options-

Model . . . . . . . . . . . Euclid
Maximum Dimensionality . . . . . 2
Minimum Dimensionality . . . . . 1
Negative Weights . . . . . . . Not Permitted

Output Options-

Job Option Header . . . . . . . Printed

Data Matrices . . . . . . . . Printed
Configurations and Transformations . Plotted
Output Dataset . . . . . . . . Not Created

27 | P a g e
Initial Stimulus Coordinates . . . Computed

For matrix (2- Dimensional)

Stress = .34540 RSQ = .59006

For matrix (1 – Dimensional)

Stress = .55533 RSQ = .30282

A. ANOVA TEST

Exhibit 27: Demographics

ANOVA
How likely would you recommend Dinshaw's Ice-cream to a friend or
relative based of Demographics?
Sum of Mean
Squares df Square F Sig.
Between Groups 2.277 4 .569 1.257 .296
Within Groups 28.970 64 .453
Total 31.246 68

ANOVA
How likely are you to purchase Dinshaw's Ice cream again based of
Demographics?
Sum of Mean
Squares df Square F Sig.
Between Groups .029 1 .029 .063 .802
Within Groups 30.174 67 .450
Total 30.203 68

ANOVA
Overall how satisfied are you with Dinshaw's Ice-Cream based of
Demographics?
Sum of Mean
Squares df Square F Sig.
Between Groups .101 2 .050 .732 .485
Within Groups 4.537 66 .069
Total 4.638 68

28 | P a g e
Exhibit 28

ANOVA
Repurchase Decision
Sig.
Wide Range of Flavor .008
Ease of Ordering .093
Good Taste .000
Hygienic Store .435
Good experience .039
Readily Available .055
Customization of topping and mix .269
wide range of packaging .130
not attractive packaging .470
one serving is of adequate volume .004
reasonably priced .000

ANOVA
Price Sensitivity
Sig.
Wide Range of Flavor .032
Ease of Ordering .137
Good Taste .000
Hygienic Store .993
Good experience .008
Readily Available .687
Customization of topping and mix .017
wide range of packaging .103
not attractive packaging .186
one serving is of adequate volume .000
reasonably priced .008

ANOVA
Overall Satisfaction
Sig.
Wide Range of Flavor .003
Ease of Ordering .563
Good Taste .000

29 | P a g e
Hygienic Store .070
Good experience .052
Readily Available .515
Customization of topping and mix .338
wide range of packaging .249
not attractive packaging .087
one serving is of adequate volume .052
reasonably priced .066

ANOVA
Recommend Dinshaw's to other
Sig.
Wide Range of Flavor .018
Ease of Ordering .001
Good Taste .000
Hygenic Store .219
Good experience .006
Readily Available .014
Customization of topping and mix .149
wide range of packaging .346
not attractive packaging .051
one serving is of adequate volume .001
reasonably priced .000

30 | P a g e
References

https://www.marketresearch.com/product/sample-7857570.pdf

http://izt.ciens.ucv.ve/ecologia/Archivos/ECOLOGIA_DE%20_POBLACIONES_Hasta%202004/ECOL_P
OBLAC_Hasta%202004_(O-Z)/Tinsley%20y%20Brown%202000_X.pdf

31 | P a g e
Individual group member contribution

Karim FGD1, Questionnaire Design, Presentation prep, Data Preparation, SPSS

Kurian FGD1, Literature Review, Report Preparation, SPSS
Prateek FGD1, Questionnaire Design, Presentation prep, Data Preparation, SPSS
Sameer K FGD2, Literature Review, Report Preparation, Data Preparation, SPSS
Vinod FGD2, Questionnaire Design, Report Preparation, SPSS
Sameer B FGD2, Literature Review, Report Preparation, SPSS

32 | P a g e