Description
The project focused on finding out the Consumer Behaviour Pattern On Online Buying The Clothes. The stated objective of the study was further broken down to secondary objectives which aimed at finding information regarding the cloths purchased by student,frequency of purchases, average spending, factors affecting online buying decision process etc.
PROJECT REPORT
ON
CONSUMER BEHAVIOUR PATTERN ON ONLINE
BUYING OF CLOTHS
By
V.A.Tripathi
Alok Arya
SALES AND MARKETING
2012
DEPARTMENT OF MANAGEMENT STUDIES
INDIAN INSTITUTE OF TECHNOLOGY
NEW DELHI
Online Buying Behavior Page 2
Table of Contents
Executive Summary ........................................................................................................................... 3
Introduction ...................................................................................................................................... 4
Objectives ...................................................................................................................................... 4
Primary Research Objective ....................................................................................................... 4
Secondary Research Objectives .................................................................................................. 4
Methodology ..................................................................................................................................... 5
Survey Administration .................................................................................................................... 5
Sampling ....................................................................................................................................... 5
Data Reduction .............................................................................................................................. 5
Data Analysis ................................................................................................................................. 6
Findings ............................................................................................................................................. 7
CROSS TABULATIONS ..................................................................................................................... 7
REGRESSION ANALYSIS ................................................................................................................ 15
ANOVA ........................................................................................................................................ 18
CLUSTER ANALYSIS....................................................................................................................... 19
DISCRIMINANT WITH CLUSTER ANALYSIS ..................................................................................... 22
FACTOR ANALYSIS ........................................................................................................................ 24
Conclusions ..................................................................................................................................... 28
Annexures ....................................................................................................................................... 30
Annexure 1a: Agglomeration Schedule for Cluster Analysis ......................................................... 31
Annexure 1b: Correlation Matrix for Factor Analysis .................................................................... 33
Annexure 1c: ANOVA Table for Regression Analysis ..................................................................... 36
Annexure 2a: Questionnaire for Exploratory Research ................................................................. 37
Annexure 2b: Final Questionnaire ................................................................................................ 38
Online Buying Behavior Page 3
Executive Summary
The project focused on finding out the Consumer Behaviour Pattern On Online Buying The
Cloths. The stated objective of the study was further broken down to secondary objectives
which aimed at finding information regarding the cloths purchased by student,frequency of
purchases, average spending, factors affecting online buying decision process etc.
The exploratory research was carried out with student studying in IITD with a set of 10 open
ended questions. The exploratory findings helped us in determining the key factors which
needed to be further explored for research. The secondary research was taken from
sources like Indian Journal of Management Technology, Zinnov LLC, and ACNielson. The
questionnaire designed had 9 questions and was administered to 106 respondents. Each of
the questions was designed to satisfy at least one of the secondary objectives of the
research. The response format was of a mixed variety which also helped in better
determination of outcomes.
Post data reduction, Cross tabulation was used for analyzing the causal relationship
between different pairs of factors. ANOVA was also applied to a pair of factors.
The Regression Analysis between the dependent variable “Average Amount spent per purchase of
cloths online” and the independent variables of Frequency of Purchase of products and services
online, owning a Credit Card, Marital Status, Education and Age, was done. The regression model
did not give any significant correlation between the factors and the Dependent Variable.
Although there is a strong interdependence between a few variables yet when taken
collectively they do not show high correlation.
Then, Cluster Analysis was done on the data and based on the responses; we could divide the
respondents in three clearly distinct groups. We named them: Confident Online Buyer, Unsure surfer
and Mall Shopper. We also performed Discriminant with Cluster Analysis to predict cluster
membership of consumers based on their attitude towards online shopping.
We performed Factor Analysis to find the major factors. We could identify six factors: Value
for Money, Trust, Connected and Up to date, Problems Faced, and Traditionalism.
Online Buying Behavior Page 4
Introduction
India has the world’s 4
th
largest Internet user base, which crossed the 100 million mark recently.
Better connectivity, booming economy and higher spending power helped the Indian e-commerce
market revenues to cross $500 million with a CAGR of 103%over last 4 years. This may not be a
significant number, averaging to only around $5 per user per year.
With the above background in mind, this research has been conducted to gain an insight into the
online buying behaviour of consumers. The objective is to explore the factors which influence online
purchase, the psychographic profile of the consumer groups and understanding the buying decision
process.
Our findings should help an Internet Marketer to determine the product/service categories to be
introduced or to be used for marketing for a specific segment of consumers. This would also allow
them to add or remove services/features which are important in the buying decision process. This
study however does not aim to identify newer areas to introduce new services, nor should it be used
to predict the success or failure of internet ventures.
Objectives
Primary Research Objective
To determine the factors and attributes which influence online buying behavior of consumers
between the age group of 18-30 years.
Secondary Research Objectives
1. To determine the psychographic profile of consumers who purchase over the Internet.
2. To determine the key product or service categories opted for, by consumers depending on
their profile.
3. To determine the factors which influence the buying decision process of a consumer.
4. To determine the average spending and frequency of purchase over the internet by a
consumer.
The exploratory research, conducted on over 12 respondents (Annexure I), focused on further
analysing the research objectives and also determining various factors which would impact the
primary research objective. Through a set of 12 open-ended questions, we could finally conclude on
some of the key factors to be further explored in the research, these included frequency of
purchase, safety issues, amount per purchase, payment methods etc…
Online Buying Behavior Page 5
Secondary Research was based on researches done by Zinnov LLC on Internet Penetration in India,
Changing Consumer Perceptions towards Online shopping in India – IJMT. Both of the researches
stressed on the consumer profiles, popular services and payments methods as important factors.
Methodology
The research was administered both online and in person during a 5 day period in February 2008.
The location of in-person administration was SIC Campus, Pune. Over 81 responses are from the
online survey and the rest 24 from in-person survey conducted.
Survey Administration
The questionnaire comprised of 9 questions (Annexure II) which measured responses for different
factors of frequency of purchase, payment methods, preferred products, average spending, hours
spent on the internet etc…
The questions measuring respondent attitudes used Likert Scale (1-5), 18 statements were given to
respondents to measure their attitudes towards online buying, and a few factual questions had
dichotomous responses.
The methods used for survey was questionnaire administration with respondents filling out the
responses themselves and online survey on SurveyGizmo.com
Sampling
The survey was conducted on 105 respondents; sample was based on affordability criteria especially
on time constraints. Email invitations were sent to invite respondents on the Internet, and students
in SIC Campus were contacted for responses.
Gender
65%
35%
Male
Female
Occupation
40%
60%
Student
Working Professional
Data Reduction
The key steps of data processing which were implemented were Editing, Coding, Transcribing, and
Summarizing statistical calculations.
Online Buying Behavior Page 6
EDITING: For some of the item non-response errors like frequency of purchase, product category or
websites. The data was interpreted and assigned to the known categories wherever possible.
CODING: For questions involving qualitative values the responses were codified using numerical
categories or values. For example; Online shopping is more convenient, the response of “strongly
agree” was coded as 1 and “strongly disagree” was coded as 5.
TRANSCRIBING: The data collected from all 105 questionnaires was edited, codified and finally
transferred on MS Excel on computer.
Data Analysis
Post Data Reduction, the data was further used for analyzing the impact of various factors on each
other as well the correlation amongst them using SPSS. The factors as well as their correlation were
studied with the help of the following techniques:
CROSS-TABS WITH CHI-SQUARE: The factors were grouped into 5 pairs based on the responses from
the questionnaire. These were studied using Chi-Square as that would help us to know the
interdependency between them. Chi-square in general studies causal relationship and thus the
hypotheses were created for each of them was done at 95%significance level. By conducting the
test and interpreting the results through the p-value, we can either accept or not accept the null
hypothesis.
REGRESSION ANALYSIS: In regression analysis, we create a model wherein we determine the
correlation between the dependent variable and multiple independent variables. By conducting the
tests and interpreting the results, we can determine the adjusted R
2
value which tells us how good
the regression model fits to the data. If the value is high, then the model fits well to the data and
that there is a high correlation between the variables. On the other hand, if the value is low, then
the model does not fit very well to the data and there is no significant correlation between the
variables.
ANOVA: Analysis of variance, better known as ANOVA, helps us to group the data into various
population samples and then check their relationship with an independent variable, which we
consider to be significant depending on the responses from the questionnaire. The null hypothesis
for this is also created at a 95%significant variable and then depending on the significant value from
the results, the hypothesis is accepted or not accepted.
CLUSTER ANALYSIS: This technique is used for segmentation of consumers on the basis of
similarities between them. The similarities could be of demographics, buying habits, or
psychographics. Hierarchical clustering is used to find the initial cluster solution, and K Means is later
used to determine cluster membership of respondents, cluster labelling is also done.
DISCRIMINANT ANALYSIS WITH CLUSTERS: A combination of Discriminant Analysis with clusters to
create a model which helps in predicting cluster membership of a consumer on the basis of the input
factors.
Online Buying Behavior Page 7
FACTOR ANALYSIS: This is a technique to reduce data complexity by reducing the number of
variables being studies. It helps identify latent or underlying factors from an array of seemingly
important variables. This procedure helps gaining insight into psychographic variables.
Findings
CROSS TABULATIONS
a) Credit Card- Frequency of Purchase
Null Hypothesis: At 95%significance level, owning a credit card does not have any impact on
the frequency of purchase.
Alternate Hypothesis: At 95%significance level, owning a credit card has an impact on the
frequency of purchase.
OwnCreditCard * FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once
a
Month
2 -3
Times
a
Month
Once in
3
Months
Once in
6
Months
Never
Tried
Once
a
Month
OwnCreditCard
Yes 23 14 28 11 1 77
No 3 2 11 10 5 31
Total 26 16 39 21 6 108
Online Buying Behavior Page 8
As the p-value from the table is lesser than 0.05, which is our assumed level of significance, we do
not accept the null hypothesis, that is, for the sample population, owninga credit card has an
impact on the frequency of purchase.
b) E-banking-Frequency of Purchase
Chi-Square Tests
Value df
Asymp. Sig.
(2-sided)
Pearson Chi-
Square
18.222(a) 4 .001
Likelihood Ratio 17.962 4 .001
Linear-by-Linear
Association
15.313 1 .000
N of Valid Cases 108
a 3 cells (30.0%) have expected count less than
5. The minimum expected count is 1.72.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N Percent
OwnCreditCard
*
FreqofPurchase
108 100.0% 0 .0% 108 100.0%
Online Buying Behavior Page 9
Null Hypothesis: At 95%significance level, e-banking does not have any impact on the
frequency of purchase.
Alternate Hypothesis: At 95%significance level, e-banking has an impact on the frequency
of purchase.
E_banking* FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once
a
month
2-3
times
a
month
Once in
3
months
Once in
6
months
Never
tried
Once
a
month
E_banking
Yes 22 15 32 11 0 80
No 3 1 6 10 7 27
Total 25 16 38 21 7 107
Chi-Square Tests
Value df
Asymp.
Sig. (2-
sided)
Pearson Chi-
Square
33.492(a) 4 .000
Likelihood Ratio 32.845 4 .000
Linear-by-Linear
Association
20.737 1 .000
N of Valid Cases 107
a 2 cells (20.0%) have expected count less
than 5. The minimum expected count is
1.77.
Case ProcessingSummary
Online Buying Behavior Page 10
As the p-value from the table is lesser than
0.05, which is our assumed level of
significance, we do not accept the null
hypothesis, that is, for the sample
population, E-bankinghas an impact on the
frequency of purchase.
c) Gender-Amount Spent
Null Hypothesis: At 95%significance level, gender does not have any impact on the average
amount spent per purchase made online.
Alternate Hypothesis: At 95%significance level, e-banking has an impact on the average
amount spent per purchase made online.
Cases
Valid Missing Total
N Percent N Percent N Percent
E_banking*
FreqofPurchase
107 100.0% 0 .0% 107 100.0%
Chi-Square Tests
Value df
Asymp. Sig.
(2-sided)
Pearson Chi-Square 2.789(a) 4 .594
Likelihood Ratio 2.811 4 .590
Linear-by-Linear
Association
.003 1 .955
N of Valid Cases 106
a 1 cells (10.0%) have expected count less than 5.
The minimum expected count is 4.81.
Gender * AmountSpent Crosstabulation
Count
AmountSpent Total
Less
than
500
500
-
1000
1000
-
2000
2000
-
5000
Greater
than
5000
Less
than
500
Gender
Male 11 13 9 27 12 72
Female 7 4 6 9 8 34
Total 18 17 15 36 20 106
Online Buying Behavior Page 11
As the p-value fromthe table is
greater than 0.05, which is our
assumed level of significance, we
accept the null hypothesis, that is,
for the sample population; gender
does not have any impact on the
average amount spent per purchase
made online.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N
Perc
ent
Gender *
AmountSpent
106 100.0% 0 .0% 106
100.
0%
Online Buying Behavior Page 12
d) Gender-Frequency of Purchase
Null Hypothesis: At 95%significance level, gender does not have any impact on the
frequency of purchase of online products and services.
Alternate Hypothesis: At 95%significance level, gender has an impact on the frequency of
purchase of online products and services.
As the p-value from the table is lesser than 0.05, which is our assumed level of significance, we do
not accept the null hypothesis, that is, for the sample population; gender has an impact on the
frequency of purchase of online products and services.
Gender * FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once
a
Month
2-3
Times
a
Month
Once in
3
Months
Once in
6
Months
Never
Tried
Once
a
Month
Gender
Male 20 14 27 9 3 73
Female 4 2 13 11 3 33
Total 24 16 40 20 6 106
Chi-Square Tests
Value df
Asymp.
Sig. (2-
sided)
Pearson Chi-Square 11.278(a) 4 .024
Likelihood Ratio 11.499 4 .021
Linear-by-Linear
Association
9.084 1 .003
N of Valid Cases 106
a 3 cells (30.0%) have expected count less
than 5. The minimum expected count is 1.87.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N Percent
Gender *
FreqofPurchase
106 100.0% 0 .0% 106 100.0%
Online Buying Behavior Page 13
e) Income-Frequency of Purchase
Null Hypothesis: At 95%significance level, income of respondents does not have any impact
on the frequency of purchase of online products and services.
Alternate Hypothesis: At 95%significance level, income of respondents has an impact on
the frequency of purchase of online products and services.
Income * FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once a
Month
2-3
Times a
Month
Once in 3
Months
Once in 6
Months
Never
Tried
Once a
Month
Income
Less than
10000
2 0 1 0 0 3
10000-
20000
1 0 2 5 1 9
20000-
30000
2 2 11 2 0 17
30000-
50000
5 0 3 1 0 9
50000-
100000
2 1 0 1 0 4
Greater
than
100000
1 1 2 0 0 4
Total 13 4 19 9 1 46
Online Buying Behavior Page 14
As the p-value from the table is greater than 0.05, which is our assumed level of significance, we
do not accept the null hypothesis, that is, for the sample population; income does not have an
impact on the frequency of purchase of online products and services.
Chi-Square Tests
Value df
Asymp. Sig. (2-
sided)
Pearson Chi-Square 28.966(a) 20 .088
Likelihood Ratio 29.758 20 .074
Linear-by-Linear
Association
2.806 1 .094
N of Valid Cases 46
a 29 cells (96.7%) have expected count less than 5. The
minimum expected count is .07.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N Percent
Income *
FreqofPurchase
46 100.0% 0 .0% 46 100.0%
Online Buying Behavior Page 15
REGRESSION ANALYSIS
The Regression Analysis between the dependent variable “Average Amount spent per purchase
made online” and the independent variables of Frequency of Purchase of products and services
online, owning a Credit Card, Marital Status, Education and Age, was done using SPSS. The details
are as below:
Variables Entered/Removed(b)
Model Variables Entered
Variables
Removed
Method
1
MaritalStatus, FreqofPurchase,
Education, CreditCard, Age(a)
. Enter
2
Age
Backward (criterion: Probability of
F-to-remove >=.100).
3 . Education
Backward (criterion: Probability of
F-to-remove >=.100).
4 . MaritalStatus
Backward (criterion: Probability of
F-to-remove >=.100).
a All requested variables entered.
b Dependent Variable: AmtSpent
Online Buying Behavior Page 16
Coefficients(a)
Model
Unstandardized Coefficients Standardized Coefficients t Sig.
B Std. Error Beta B Std. Error
1
(Constant) 1.696 1.954
.868 .388
FreqofPurchase .402 .122 .330 3.305 .001
Age .054 .078 .083 .696 .489
CreditCard -.695 .318 -.234 -2.186 .032
Education -.152 .202 -.076 -.753 .454
MaritalStatus .384 .464 .096 .828 .410
2
(Constant) 2.897 .912
3.178 .002
FreqofPurchase .403 .121 .331 3.323 .001
CreditCard -.755 .305 -.254 -2.477 .015
Education -.134 .199 -.067 -.671 .504
MaritalStatus .534 .409 .134 1.304 .196
3
(Constant) 2.564 .762
3.364 .001
FreqofPurchase .415 .120 .341 3.467 .001
CreditCard -.772 .303 -.259 -2.547 .013
MaritalStatus .561 .406 .140 1.380 .171
4
(Constant) 3.366 .496
6.781 .000
FreqofPurchase .408 .120 .335 3.393 .001
CreditCard -.882 .294 -.297 -3.005 .003
a Dependent Variable: AmtSpent
Online Buying Behavior Page 17
As can be seen from the above table, the independent variables can be gradually removed in the
regression model as they don’t have any significant impact on the value of R
2
. The value of R
2
is quite
low and so it can be said that the regression model does not fit into the data very well. Also, the sum
of squares of regression is lesser than the sum of squares of residuals and this reiterates the findings
of R
2
. This is because if the sum of squares of regression is lesser than the sum of squares of
residuals, then the independent variables do not explain the variation in the dependent variable
well. While cross tabs suggest a positive relationship between multiple pairs of factors, the linear
correlation model, with all factors together, does not fit in with the outcomes.
Excluded Variables(d)
Model
Beta In t Sig. Partial Correlation Collinearity Statistics
Tolerance Tolerance Tolerance Tolerance Tolerance
2 Age .083(a) .696 .489 .077 .682
3
Age .071(b) .606 .546 .067 .694
Education -.067(b) -.671 .504 -.074 .962
4
Age .122(c) 1.164 .248 .127 .874
Education -.080(c) -.798 .427 -.087 .971
MaritalStatus .140(c) 1.380 .171 .150 .926
a Predictors in the Model: (Constant), MaritalStatus, FreqofPurchase, Education, CreditCard
b Predictors in the Model: (Constant), MaritalStatus, FreqofPurchase, CreditCard
c Predictors in the Model: (Constant), FreqofPurchase, CreditCard
d Dependent Variable: AmtSpent
Online Buying Behavior Page 18
ANOVA
Null hypothesis: At 95%confidence interval for the population taken, income does not have any
impact on the frequency of purchase of online products and services.
Alternate Hypothesis: At 95%confidence interval for the population taken, income has an impact on
the frequency of purchase of online products and services.
ANOVA
Frequency
Sum of
Squares df
Mean
Square F Sig.
Between
Groups
5.317 6 .886 .605 .726
Within Groups 149.417 102 1.465
Total 154.734 108
N Mean
Std.
Deviation
Std.
Error
95%Confidence
Interval for Mean
Minim
um
Maxim
um
Lower
Bound
Upper
Bound
.00 64 2.3125 1.29560 .16195 1.9889 2.6361 .00 4.00
Less than
10000
3 2.3333 .57735 .33333 .8991 3.7676 2.00 3.00
10000-20000 7 2.8571 1.46385 .55328 1.5033 4.2110 .00 4.00
20000-30000 16 2.7500 .85635 .21409 2.2937 3.2063 1.00 4.00
30000-50000 7 2.7143 .75593 .28571 2.0152 3.4134 2.00 4.00
50000-100000 5 2.0000 1.22474 .54772 .4793 3.5207 1.00 4.00
Greater than
100000
7 2.4286 1.27242 .48093 1.2518 3.6054 .00 4.00
Total 109 2.4312 1.19696 .11465 2.2039 2.6584 .00 4.00
Online Buying Behavior Page 19
Means Plots
Income
Greater than
100000
50000-
100000
30000-50000 20000-30000 10000-20000 Less than
10000
.00
M
e
a
n
o
f
F
r
e
q
u
e
n
c
y
3.00
2.80
2.60
2.40
2.20
2.00
The p-value from the ANOVA table is greater than the significance value of 0.05 assumed by us.
Thus, at this significance level we accept the null hypothesis. So we can conclude that income does
not have an impact on the frequency of purchase of online products and services for these
respondents. Do remember that the same conclusion was arrived at when Cross tabulation of
location and usage rate was performed earlier.
CLUSTER ANALYSIS
The cluster analysis was run where people were surveyed about their attitudes towards internet
shopping. The preferences indicated by respondents were used to find out the consumer segments
that react differently to different parameters related to online shopping. The segments obtained
would give an understanding as to how the consumers are placed in terms of their attitudes.
Hierarchical clustering was done to determine the initial cluster solution. While the initial cluster
solution by SPSS gives us 2 clusters, we take a difference of coefficients greater than or equal to 2.72
form another cluster. Now we execute the K-Mean cluster to get the final cluster solution and
through ANOVA table we get that all the variables bear significance at 95%confidence level.
Online Buying Behavior Page 20
ANOVA
Clust er Error F Sig.
Mean Square df Mean Square Df
Mean
Square df
InternetvsMall
27.190 2 .515 103 52.814 .000
LatestInfo
1.744 2 .341 103 5.116 .008
Accesibility
6.364 2 .536 103 11.874 .000
Convenience
27.439 2 .367 103 74.819 .000
Savetime
7.485 2 .608 103 12.312 .000
AnywhereAnytime
2.935 2 .559 103 5.255 .007
CreditCardSafe
3.441 2 .619 103 5.562 .005
SpecificDateTime
6.393 2 .553 103 11.554 .000
GuaranteedQuality
4.378 2 .469 103 9.334 .000
Discounts
9.581 2 .710 103 13.500 .000
Hasslefree
8.649 2 .691 103 12.518 .000
CashonDelivery
1.011 2 .791 103 1.278 .283
EasyFind
5.585 2 .838 103 6.668 .002
FacedProblems
.866 2 .730 103 1.186 .309
Continue
6.682 2 .823 103 8.121 .001
TouchandFeel
5.330 2 .728 103 7.320 .001
DeliveryProcess
8.284 2 .499 103 16.612 .000
NoCreditCard
12.382 2 .950 103 13.035 .000
The F tests should be used only for descriptive purposes because the clusters have been chosen to maximize
the differences among cases in different clusters. The observed significance levels are not corrected for this and
thus cannot be interpreted as tests of the hypothesis that the cluster means are equal.
Final Clust er Cent ers
Clust er
1 2 3
InternetvsMall
3.38 2.09 4.00
LatestInfo
1.58 1.78 2.05
Accesibility
1.37 1.94 2.18
Convenience
2.62 1.81 3.86
Savetime
2.33 1.59 2.55
AnywhereAnytime
1.88 2.09 2.50
CreditCardSafe
2.52 2.78 3.18
SpecificDateTime
2.10 2.28 3.00
GuaranteedQuality
2.98 3.56 3.55
Discounts
2.15 2.72 3.23
Hasslefree
2.54 2.03 3.18
CashonDelivery
2.15 2.34 2.50
EasyFind
2.00 2.63 2.68
FacedProblems
2.52 2.81 2.59
Continue
2.40 2.94 3.27
TouchandFeel
1.81 2.53 1.95
DeliveryProcess
2.42 2.66 3.45
NoCreditCard
4.08 3.81 2.82
Online Buying Behavior Page 21
Variable Description Cluster 1 (52) Cluster 2(32) Cluster 3(22)
I prefer making a purchase from internet than using local
malls or stores
Disagree Strongly Agree Strongly
Disagree
I can get the latest information from the Internet regarding
different products/ services that is not available in the
market.
Strongly Agree NAND Strongly
Disagree
I have sufficient internet accessibility to shop online. Strongly Agree Mildly Disagree Strongly
Disagree
Online shopping is more convenient than in-store
shopping.
Mildly Agree Strongly Agree Strongly
Disagree
Online shopping saves time over in-store shopping. Disagree Strongly Agree Strongly
Disagree
Online shopping allows me to shop anywhere and at
anytime.
Strongly Agree Mildly Agree Strongly
Disagree
It is safe to use a credit card while shopping on the
Internet.
Strongly Agree NAND Strongly
Disagree
Online shopping provides me with the opportunity to get
the products delivered on specific date and time anywhere
as required.
Strongly Agree Moderately Agree Strongly
Disagree
Products purchased through the Internet are with
guaranteed quality.
Strongly Agree Strongly Disagree Strongly
Disagree
Internet provides regular discounts and promotional
offers to me.
Strongly Agree NAND Strongly
Disagree
Internet helps me avoid hassles of shopping in stores. NAND Strongly Agree Strongly
Disagree
Cash on Delivery is a better way to pay while shopping on
the Internet.
Strongly Agree NAND Strongly
Disagree
Sometimes, I can find products online which I may not find
in-stores.
Strongly Agree Strongly Disagree Strongly
Disagree
I have faced problems while shopping online. Strongly Agree Strongly Disagree Agree
I continue shopping online despite facing problems on
some occasions.
Strongly Agree Mildly Disagree Strongly
Disagree
It is important for me to touch and feel certain products
before I purchase them. So I cannot buy them online.
Strongly Agree Strongly Disagree Agree
I trust the delivery process of the shopping websites. Strongly Agree Agree Strongly
Disagree
I do not shop online only because I do not own a credit
card.
Strongly Disagree Disagree Strongly Agree
Online Buying Behavior Page 22
On the basis of the above scales, obtained by rating the relative results, we can name our clusters as:
? Cluster 1 : Confident Online Buyer
? Cluster 2 : Unsure surfer
? Cluster 3 : Mall Shopper
DISCRIMINANT WITH CLUSTER ANALYSIS
By combining cluster analysis with Discriminant analysis, we could derive a model which could
predict cluster membership of the respondents on the basis of the variables analysed.
The cluster solution is changed with K Means cluster>Save option selected for Cluster Membership.
This gives a new column in the Data Sheet, showing the cluster membership of the responses. Using
this data sheet, Discriminant analysis is done over the cluster membership column as the dependent
variables.
Eigenvalues
a First 2 canonical discriminant functions were used in the
analysis.
Wilks' Lambda
Test of
Function(s)
Wilks'
Lambda
Chi-
square df Sig.
1 through 2 .072 248.627 36 .000
2
.289 117.402 17 .000
Canonical Discriminant Function Coeff icients
Unstandardized coefficients
The Eigenvalues are greater than 1, and the Wilk’s Lamba
is below 0.5, this indicates that the Discriminant Model is
able to explain the data fairly well.
The Canonical Discriminant Function Coefficients table contains Group-1 number of functions; the
equations as derived from above are:
F
1
=.901(InternetvsMall) - .090 LatestInfo - .122 Accesibility -.989 Convenience +.370 Savetime - .228
AnywhereAnytime – 0.080 CreditCardSafe +.028 SpecificDateTime - .621 GuarenteeQuality +.045
Discounts - .097 Hasslefree +.193 CoD - .038 EasyFind +0.029 FacedProblems - .198 Continue - .447
TouchandFeel +.308 DeliveryProcess +.008 NoCreditCard – 3.372
Functio
n
Eigenvalu
e
% of
Variance
Cumulative
%
Canonical
Correlation
1 3.009(a) 55.0 55.0 .866
2 2.464(a) 45.0 100.0 .843
Function
1 2
InternetvsMall
.901 -.372
LatestInfo
-.090 .216
Accesibility
-.122 .346
Convenience
.989 .750
Savetime
.370 -.747
AnywhereAnytime
-.228 .185
CreditCardSafe
-.080 .241
SpecificDateTime
.028 .510
GuaranteedQuality
-.621 .543
Discounts
.045 .232
Hasslefree
.097 .180
CashonDelivery
.193 .262
EasyFind
-.038 .162
FacedProblems
.029 .272
Continue
-.198 .179
TouchandFeel
-.447 .623
DeliveryProcess
.308 .262
NoCreditCard
.008 -.609
(Constant)
-3.372 -7.122
Online Buying Behavior Page 23
F
2
=-.372 InternetvsMall +.216 LatestInfo +.346 Accesibility +.750 Convenience - .747 Savetime +
.185 AnywhereAnytime +.241 CreditCardSafe +.510 SpecificDateTime +.543 GuarenteeQuality -
.232 Discounts +.180 HassleFree +.262 CoD +.162 EasyFind +.272 FacedProblems +.179 Continue +
.623 TouchnFeel +.262 DeliveryProcess - .609 NoCreditCard – 7.122
Classif ication Function Coeff icients
Cluster Number of Case
1 2 3
InternetvsMall
6.099 2.479 5.961
LatestInfo
10.113 10.871 10.813
Accesibility
-4.203 -3.056 -3.047
Convenience
5.202 3.798 9.499
Savetime
.077 -2.731 -2.261
AnywhereAnytime
1.323 2.440 1.705
CreditCardSafe
5.904 6.687 6.715
SpecificDateTime
4.090 5.135 6.087
GuaranteedQuality
4.190 7.322 5.385
Discounts
1.330 1.707 2.285
Hasslefree
2.116 2.215 2.944
CashonDelivery
8.338 8.321 9.619
EasyFind
1.519 1.999 2.087
FacedProblems
5.910 6.424 6.994
Continue
-.681 .331 -.279
TouchandFeel
6.076 8.846 7.825
DeliveryProcess
2.462 2.087 3.909
NoCreditCard
3.878 2.501 1.551
(Constant)
-79.869 -87.731 -116.575
Fisher's linear discriminant functions
The above table provides the linear Discriminant function which can be used to judge the membership of each
data set by putting the value in each of them, and choosing the one which provides the highest value.
Classif ication Result s(a)
Cluster Number of Case
Predicted Group Membership Total
1 2 3 1
Original Count 1
51 1 0 52
2
0 32 0 32
3
1 0 21 22
% 1
98.1 1.9 .0 100.0
2
.0 100.0 .0 100.0
3
4.5 .0 95.5 100.0
a 98.1% of original grouped cases correctly classified.
Our Discriminant Model is able to explain 98.1%of the data set correctly, which is significantly high,
and indicates that this is a dependable Discriminant Model for prediction.
Online Buying Behavior Page 24
FACTOR ANALYSIS
The responses are put into SPSS for data reduction through Factor Analysis. The details of the above
are provided below:
Total Variance Explained
Component
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings
Total %of Variance Cumulative % Total %of Variance Cumulative % Total %of Variance Cumulative %
1 3.854 21.410 21.410 3.854 21.410 21.410 2.559 14.219 14.219
2 2.175 12.082 33.491 2.175 12.082 33.491 2.318 12.879 27.098
3 1.652 9.175 42.666 1.652 9.175 42.666 2.160 11.997 39.095
4 1.514 8.413 51.080 1.514 8.413 51.080 1.727 9.593 48.688
5 1.277 7.096 58.176 1.277 7.096 58.176 1.422 7.901 56.589
6 1.098 6.101 64.276 1.098 6.101 64.276 1.243 6.903 63.492
7 1.066 5.924 70.200 1.066 5.924 70.200 1.207 6.708 70.200
8 .875 4.859 75.059
9 .771 4.283 79.342
10 .737 4.095 83.438
11 .545 3.030 86.467
12 .529 2.939 89.407
13 .384 2.134 91.541
14 .377 2.092 93.633
15 .358 1.988 95.621
16 .293 1.626 97.247
17 .278 1.546 98.793
18 .217 1.207 100.000
Extraction Method: Principal Component Analysis.
Online Buying Behavior Page 25
We see from the Cumulative Percentage column that there have been seven components or factors
extracted which explain 70.2%of the total variance (information contained in the original 18
variables). This is an acceptable solution as generally, 70%of the total variance should be explained
by the factors for the solution to be accepted.
Rotated Component Matrix(a)
Component
1 2 3 4 5 6 7
InternetvsMall .714 -.279 .149 .056 .046 -.197 -.241
LatestInfo .099 .322 -.146 .804 -.072 -.125 -.105
Accesibility -.043 .157 .158 .859 .105 .084 .020
Convenience .796 .045 .202 .206 .030 -.078 -.109
Savetime .726 .270 -.040 -.015 .017 -.120 .065
AnywhereAnytime .314 .571 .153 .109 .251 .025 -.034
CreditCardSafe .023 -.046 .657 -.047 .090 -.271 -.416
SpecificDateTime .272 -.198 .466 .404 -.197 .321 .107
GuaranteedQuality .023 .412 .647 .016 .071 -.124 .309
Discounts .069 .714 .208 .233 .095 -.095 -.158
Hasslefree .694 .225 -.022 -.153 -.129 .320 -.017
CashonDelivery -.121 -.024 -.022 .008 .123 .882 -.126
EasyFind .013 .869 .018 .143 -.043 .055 .012
FacedProblems -.203 .020 -.091 .084 .788 -.025 .049
Continue .071 .379 .518 -.040 .555 .015 -.077
TouchandFeel -.351 -.060 -.006 .079 -.546 -.193 -.065
DeliveryProcess .112 .135 .796 .065 -.101 .126 -.059
NoCreditCard -.142 -.130 -.047 -.053 .084 -.147 .885
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 8 iterations.
Online Buying Behavior Page 26
Looking at the rotated component matrix, we see that the loadings of variables-Internet preferred
over mall, Convenience and Saves Time, on factor 1 are high (greater than 0.7) and thus factor 1 is
made up of these variables. Similarly, we can interpret for the other factors as well and a table for
the same has been provided at the end of this section.
Factor Variables Label for the Factor
1 Internet over Mall
Convenience
Saves Time
Time-bound and comfort seeking
2 Discounts
Easy Find
Value for Money
3 Delivery Process
Credit Card Safe to Use
Guaranteed Quality
Trust
4 Latest Information
Accessibility
Connected and Up to date
5 Faced Problems Problems Faced
6 Cash on Delivery
Don’t Own a credit card
Traditionalism
7 Don’t Own a credit card N/A
We have combined the sixth and the seventh factor as they go hand-in-hand. A person who does not
own a credit card but shops online would invariably prefer to pay by cash on delivery.
PERCEPTUAL MAPS
Online Buying Behavior Page 27
Online Buying Behavior Page 28
Others:
Conclusions
We found a strong inter-dependence between a few variables affecting online buying behaviour. For
example, we found that owning a credit card has a significant impact on the frequency of online
purchases as credit card is the most popular mode of payment on the Internet. Apart from the credit
card, E-Banking is also slowly becoming a popular mode of payment and we found a relationship
between people who use E-Banking and their frequency of online purchases too.
Interestingly, we found that gender does not have any major impact on the average amount spent
over the Internet in a month, but it does have a relationship with the frequency of purchases. Also,
the income of an individual does not have show any significant relationship with the frequency of
purchases. These findings are starkly similar to the findings of Changing Consumer Perceptions
towards Online shopping in India – IJMT, which was a part of our secondary data.
Based on the responses, we could divide the respondents in three clearly distinct groups. We named
them: Confident Online Buyer, Unsure surfer and Mall Shopper. We were also able to successfully
able to create a discriminant model which could predict cluster membership of users.
We could also arrive at six factors which can explain the data with 70%significance, these factors
could be categorised into Time-bound and comfort seeking Value for Money, Trust, Connected and
Up to date, Problems Faced, and Traditionalism.
We also found that the most popular product category sold online is Air/Rail Tickets. This forms a
major chunk of the average amount spent by our respondents on the internet, Books come a close
second. It must be noted that both the above products have a relatively low touch-and-feel need.
Online Buying Behavior Page 29
These findings depict almost the same ranking as found by ACNielson on popular services on the
Internet.
The most popular websites for these were found to be Makemytrip.com and Yatra.com. Apart from
Air Tickets, Books, Gifts and Electronic Products are also very popular with the Online Shoppers and
they are spending, on an average, Rs2000-Rs5000 per month on online purchases.
Online Buying Behavior Page 30
Annexures
Online Buying Behavior Page 31
Annexure 1a: Agglomeration Schedule for Cluster Analysis
Agglomeration Schedule
Stage
Cluster
Combined
Coefficients
Stage Cluster
First Appears
Next
Stage
Cluster
1
Cluster
2
Cluster
2
Cluster
1
1
105 106 0.000 0 0 2
2 16 105 0.000 0 1 4
3
103 104 0.000 0 0 4
4
16 103 0.000 2 3 6
5
101 102 0.000 0 0 6
6
16 101 0.000 4 5 8
7 99 100 0.000 0 0 8
8
16 99 0.000 6 7 10
9
97 98 0.000 0 0 10
10
16 97 0.000 8 9 72
11
19 85 3.000 0 0 31
12 1 91 4.000 0 0 53
13 84 86 5.000 0 0 32
14
31 49 5.000 0 0 44
15
92 94 6.000 0 0 20
16
6 93 6.000 0 0 26
17
25 89 6.000 0 0 36
18 43 81 6.000 0 0 45
19
21 65 6.000 0 0 32
20
24 92 7.000 0 15 55
21
52 90 7.000 0 0 48
22
45 46 7.000 0 0 44
23 27 44 7.000 0 0 49
24
70 87 8.000 0 0 35
25
39 82 8.000 0 0 46
26
6 69 8.000 16 0 37
27
42 64 8.000 0 0 70
28 26 58 8.000 0 0 57
29 2 37 8.000 0 0 49
30
15 33 8.000 0 0 61
31
19 47 8.500 11 0 36
32
21 84 9.000 19 13 54
33
30 83 9.000 0 0 64
34 73 74 9.000 0 0 63
35
34 70 9.000 0 24 54
36
19 25 9.333 31 17 55
37
6 88 9.667 26 0 50
38
4 78 10.000 0 0 53
39 54 55 10.000 0 0 67
40
12 53 10.000 0 0 68
41
5 41 10.000 0 0 45
42
7 20 10.000 0 0 65
43
76 79 11.000 0 0 83
44 31 45 11.000 14 22 52
45 5 43 11.000 41 18 73
Online Buying Behavior Page 32
46
22 39 11.000 0 25 58
47 3 23 11.000 0 0 79
48
52 60 11.500 21 0 63
49
2 27 11.500 29 23 62
50
6 32 11.750 37 0 58
51
18 71 12.000 0 0 94
52 31 61 12.000 44 0 62
53 1 4 12.000 12 38 67
54
21 34 12.250 32 35 61
55
19 24 12.333 36 20 68
56
17 67 13.000 0 0 87
57 26 56 13.000 28 0 64
58 6 22 13.933 50 46 73
59
50 63 14.000 0 0 66
60
35 48 14.000 0 0 70
61
15 21 14.286 30 54 69
62
2 31 14.450 49 52 71
63 52 73 14.500 48 34 72
64
26 30 14.500 57 33 74
65
7 59 15.000 42 0 81
66
36 50 15.000 0 59 86
67
1 54 15.250 53 39 78
68 12 19 15.250 40 55 71
69 15 38 15.889 61 0 77
70
35 42 16.000 60 27 85
71
2 12 16.522 62 68 74
72
16 52 17.200 10 63 84
73 5 6 17.375 45 58 77
74 2 26 17.653 71 64 78
75
28 75 18.000 0 0 95
76
29 40 18.000 0 0 88
77
5 15 18.017 73 69 82
78
1 2 18.458 67 74 82
79 3 72 18.500 47 0 90
80
11 57 19.000 0 0 100
81
7 62 19.333 65 0 83
82
1 5 20.361 78 77 85
83
7 76 20.750 81 43 87
84 16 51 21.750 72 0 89
85 1 35 22.029 82 70 88
86
9 36 22.333 0 66 92
87
7 17 22.667 83 56 93
88
1 29 23.786 85 76 89
89 1 16 24.603 88 84 92
90 3 80 24.667 79 0 93
91
14 96 26.000 0 0 94
92
1 9 27.223 89 86 95
93
3 7 27.250 90 87 97
94
14 18 28.000 91 51 96
95 1 28 29.506 92 75 96
96
1 14 30.753 95 94 97
97
1 3 32.106 96 93 98
98
1 10 34.825 97 0 100
99
8 68 36.000 0 0 103
100 1 11 37.082 98 80 101
Online Buying Behavior Page 33
101
1 95 38.550 100 0 102
102 1 13 43.089 101 0 103
103
1 8 43.637 102 99 104
104
1 77 48.423 103 0 105
105
1 66 66.914 104 0 0
Annexure 1b: Correlation Matrix for Factor Analysis
Correlation Matrix
Inter
netvs
Mall
Lat
estI
nfo
Acc
esib
ility
Con
veni
ence
Sav
eti
me
Anywh
ereAn
ytime
Credi
tCard
Safe
Specif
icDate
Time
Guara
nteed
Qualit
y
Dis
cou
nts
Has
slef
ree
Cash
onDe
livery
Ea
syF
ind
Face
dPro
blem
s
Co
nti
nu
e
Touc
hand
Feel
Deliv
eryPr
ocess
NoCr
edit
Card
Cor
rela
tion
Intern
etvsM
all
1.000
.05
3
-
.049
.569
.28
5
.134 .189 .192 -.020
-
.00
1
.34
1
-.176
-
.14
9
-.111
.02
3
-.174 .208 -.205
LatestI
nfo
.053
1.0
00
.602 .212
.13
2
.164 -.046 .125 .078
.37
4
.02
6
-.089
.37
9
-.047
.03
0
-.025 .016 -.151
Accesi
bility
-.049
.60
2
1.00
0
.159
.05
1
.234 .113 .255 .233
.27
7
-
.06
3
.116
.23
6
.101
.12
4
-.003 .161 -.057
Conve
nience
.569
.21
2
.159
1.00
0
.54
1
.260 .182 .264 .133
.25
2
.43
3
-.091
.09
3
-.115
.18
9
-.200 .297 -.190
Saveti
me
.285
.13
2
.051 .541
1.0
00
.325 .098 .089 .130
.16
8
.44
3
-.118
.18
1
-.103
.17
3
-.158 .023 -.080
Anywh
ereAny
time
.134
.16
4
.234 .260
.32
5
1.000 .136 .170 .263
.41
3
.21
2
-.031
.47
5
.086
.40
6
-.210 .142 -.135
Credit
CardSa
fe
.189
-
.04
6
.113 .182
.09
8
.136 1.000 .093 .310
.10
3
-
.03
1
-.092
-
.03
3
-.062
.33
5
-.005 .373 -.241
Specifi
cDateT
ime
.192
.12
5
.255 .264
.08
9
.170 .093 1.000 .113
.02
9
.17
0
.063
-
.04
7
-.142
.13
5
.051 .342 -.103
Guara
nteed
Qualit
y
-.020
.07
8
.233 .133
.13
0
.263 .310 .113 1.000
.33
3
.09
8
-.119
.31
4
.014
.44
6
-.109 .425 .104
Online Buying Behavior Page 34
Discou
nts
-.001
.37
4
.277 .252
.16
8
.413 .103 .029 .333
1.0
00
.14
3
-.073
.53
6
.129
.39
7
-.048 .341 -.185
Hasslef
ree
.341
.02
6
-
.063
.433
.44
3
.212 -.031 .170 .098
.14
3
1.0
00
.121
.14
9
-.112
.06
5
-.139 .104 -.187
Casho
nDeliv
ery
-.176
-
.08
9
.116
-
.091
-
.11
8
-.031 -.092 .063 -.119
-
.07
3
.12
1
1.000
.01
2
.066
.02
9
-.115 .050 -.119
EasyFi
nd
-.149
.37
9
.236 .093
.18
1
.475 -.033 -.047 .314
.53
6
.14
9
.012
1.0
00
.014
.28
5
-.076 .187 -.098
FacedP
roblem
s
-.111
-
.04
7
.101
-
.115
-
.10
3
.086 -.062 -.142 .014
.12
9
-
.11
2
.066
.01
4
1.000
.30
9
-.071 -.107 .118
Contin
ue
.023
.03
0
.124 .189
.17
3
.406 .335 .135 .446
.39
7
.06
5
.029
.28
5
.309
1.0
00
-.267 .344 -.120
Toucha
ndFeel
-.174
-
.02
5
-
.003
-
.200
-
.15
8
-.210 -.005 .051 -.109
-
.04
8
-
.13
9
-.115
-
.07
6
-.071
-
.26
7
1.00
0
-.069 .015
Deliver
yProce
ss
.208
.01
6
.161 .297
.02
3
.142 .373 .342 .425
.34
1
.10
4
.050
.18
7
-.107
.34
4
-.069 1.000 -.123
NoCre
ditCar
d
-.205
-
.15
1
-
.057
-
.190
-
.08
0
-.135 -.241 -.103 .104
-
.18
5
-
.18
7
-.119
-
.09
8
.118
-
.12
0
.015 -.123
1.00
0
Sig.
(1-
tail
ed)
Intern
etvsM
all
.29
4
.308 .000
.00
2
.086 .026 .025 .420
.49
5
.00
0
.035
.06
3
.129
.40
9
.037 .016 .017
LatestI
nfo
.294 .000 .014
.08
9
.046 .320 .100 .213
.00
0
.39
5
.181
.00
0
.316
.37
9
.401 .436 .061
Accesi
bility
.308
.00
0
.052
.30
3
.008 .125 .004 .008
.00
2
.25
9
.117
.00
7
.151
.10
3
.487 .050 .281
Conve
nience
.000
.01
4
.052
.00
0
.004 .031 .003 .087
.00
5
.00
0
.178
.17
0
.120
.02
6
.020 .001 .025
Saveti
me
.002
.08
9
.303 .000 .000 .160 .181 .093
.04
2
.00
0
.113
.03
2
.147
.03
8
.053 .407 .208
Anywh
ereAny
.086
.04
.008 .004
.00
.082 .041 .003
.00 .01
.376
.00
.190
.00
.015 .073 .084
Online Buying Behavior Page 35
time 6 0 0 5 0 0
Credit
CardSa
fe
.026
.32
0
.125 .031
.16
0
.082 .172 .001
.14
6
.37
5
.174
.36
8
.264
.00
0
.478 .000 .006
Specifi
cDateT
ime
.025
.10
0
.004 .003
.18
1
.041 .172 .124
.38
5
.04
1
.260
.31
6
.073
.08
4
.301 .000 .148
Guara
nteed
Qualit
y
.420
.21
3
.008 .087
.09
3
.003 .001 .124
.00
0
.15
9
.112
.00
1
.442
.00
0
.132 .000 .145
Discou
nts
.495
.00
0
.002 .005
.04
2
.000 .146 .385 .000
.07
2
.228
.00
0
.093
.00
0
.312 .000 .029
Hasslef
ree
.000
.39
5
.259 .000
.00
0
.015 .375 .041 .159
.07
2
.108
.06
4
.126
.25
6
.077 .144 .027
Casho
nDeliv
ery
.035
.18
1
.117 .178
.11
3
.376 .174 .260 .112
.22
8
.10
8
.45
1
.249
.38
3
.121 .305 .112
EasyFi
nd
.063
.00
0
.007 .170
.03
2
.000 .368 .316 .001
.00
0
.06
4
.451 .444
.00
2
.218 .027 .158
FacedP
roblem
s
.129
.31
6
.151 .120
.14
7
.190 .264 .073 .442
.09
3
.12
6
.249
.44
4
.00
1
.236 .139 .115
Contin
ue
.409
.37
9
.103 .026
.03
8
.000 .000 .084 .000
.00
0
.25
6
.383
.00
2
.001 .003 .000 .111
Toucha
ndFeel
.037
.40
1
.487 .020
.05
3
.015 .478 .301 .132
.31
2
.07
7
.121
.21
8
.236
.00
3
.242 .437
Deliver
yProce
ss
.016
.43
6
.050 .001
.40
7
.073 .000 .000 .000
.00
0
.14
4
.305
.02
7
.139
.00
0
.242 .105
NoCre
ditCar
d
.017
.06
1
.281 .025
.20
8
.084 .006 .148 .145
.02
9
.02
7
.112
.15
8
.115
.11
1
.437 .105
Online Buying Behavior Page 36
Annexure 1c: ANOVA Table for Regression Analysis
ANOVA(e)
Model
Sum of Squares df Mean Square F Sig.
1
Regression 35.202 5 7.040 4.396 .001(a)
Residual 129.717 81 1.601
Total 164.920 86
2
Regression 34.427 4 8.607 5.408 .001(b)
Residual 130.493 82 1.591
Total 164.920 86
3
Regression 33.711 3 11.237 7.108 .000(c)
Residual 131.209 83 1.581
Total 164.920 86
4
Regression 30.700 2 15.350 9.607 .000(d)
Residual 134.219 84 1.598
Total 164.920 86
a Predictors: (Constant), MaritalStatus, FreqofPurchase, Education, CreditCard, Age
b Predictors: (Constant), MaritalStatus, FreqofPurchase, Education, CreditCard
c Predictors: (Constant), MaritalStatus, FreqofPurchase, CreditCard
d Predictors: (Constant), FreqofPurchase, CreditCard
e Dependent Variable: AmtSpent
Online Buying Behavior Page 37
Annexure 2a: Questionnaire for Exploratory Research
Hi,
We are conducting a research on consumer behavior on the internet. As a part of our exploratory
research we request you to spare a few minutes in answering the questions below. Do feel free to
ask us for any clarifications or doubts.
Your responses would serve as a base to our research; and would guide us in discovering important
facts about the buying process.
Thanks.
1. What are the online services that you generally use while surfing the net (answer as many)?
2. Which are the websites that you visit frequently for the following purposes:
a. E-mail:
b. Chat:
c. Shopping:
d. Job Search:
e. Social Networking:
f. Banking and other Financial Services(Stock Trading):
g. Education:
h. General Browsing:
2. How frequently do you shop online?
3. If you do not shop online, then what are the reasons behind not shopping online?
4. What are the occasions when you buy online?
5. While purchasing online, what are the goods and services that you generally buy?
6. What are the payment methods you generally use for online purchases?
7. What are the different types of payment options that you have come across?
Online Buying Behavior Page 38
8. Do you feel it is safe to buy online?
9. What are the features that make a website more attractive than the others while buying online?
10. On an average, how much do you spend while buying online?
11. What is your general experience of buying online as compared to conventional shopping?
12. What additional features, do you feel, would enhance your buying experience on the internet?
Any additional insights:
Name: Age:
Profession: Sex: M/F
Years since you started using the Internet:
Do you own a credit card? Y/N
Do you use Internet Banking? Y/N
Annexure 2b: Final Questionnaire
QUESTIONNAIRE
We would be thankful for your cooperation if you spare a few minutes to answer the
following questions:
1. Do you like to purchase clothes via E-Shopping?
Yes No
Online Buying Behavior Page 39
2. On an average, how much time (per week) do you spend while surfing the Net?
a) 0-2 hours b) 2-6 hours
c) 6-10 hours d) 10-15 hours
e) Greater than 15 hours
3. I am more comfortable for purchasing clothes via
Offline clothes purchasing Online Clothes Purchasing
4. How much you spend money for purchasing clothes on monthly basis?
Yes No
5. I am/ would not comfortable buying clothes online because
a) have no trust b) Qaulity issue
c) fitting issue d) Fake
Any other product, please specify
6. If you already buying cloths online How frequently do you purchase cloths online?
a) Once a month b) 2-3 times a month
c) Once in 3 months d) Once in 6 months
Any other, please specify
7. What is the average amount that you spend per purchase while shopping online?
a) Rs 5000
8. Which of the following web sites do you shop at?
a)
Any other, please specify
9. Recall your earlier online buying/ shopping experience and please indicate you degree of
agreement with the following statements:
Strongly
Agree
Agree Neither
Agree nor
Disagree
Disagree Strongly
Disagree
I prefer making a purchase from
internet than using local malls or
stores
I can get the latest information
from the Internet regarding
different products/ services that is
not available in the market.
I have sufficient internet
accessibility to shop online.
Online shopping is more
Online Buying Behavior Page 40
convenient than in-store shopping.
Online shopping saves time over
in-store shopping.
Online shopping allows me to shop
anywhere and at anytime.
It is safe to use a credit card while
shopping on the Internet.
Online shopping provides me with
the opportunity to get the
products delivered on specific date
and time anywhere as required.
Products purchased through the
Internet are with guaranteed
quality.
Internet provides regular
discounts and promotional offers
to me.
Internet helps me avoid hassles of
shopping in stores.
Cash on Delivery is a better way to
pay while shopping on the
Internet.
Sometimes, I can find products
online which I may not find in-
stores.
I have faced problems while
shopping online.
I continue shopping online despite
facing problems on some
occasions.
It is important for me to touch and
feel certain products before I
purchase them. So I cannot buy
them online.
I trust the delivery process of the
shopping websites.
I do not shop online only because I
do not own a credit card.
Name: Gender: Occupation:
Online Buying Behavior Page 41
Monthly Income: Education: Marital Status:
doc_198233867.pdf
The project focused on finding out the Consumer Behaviour Pattern On Online Buying The Clothes. The stated objective of the study was further broken down to secondary objectives which aimed at finding information regarding the cloths purchased by student,frequency of purchases, average spending, factors affecting online buying decision process etc.
PROJECT REPORT
ON
CONSUMER BEHAVIOUR PATTERN ON ONLINE
BUYING OF CLOTHS
By
V.A.Tripathi
Alok Arya
SALES AND MARKETING
2012
DEPARTMENT OF MANAGEMENT STUDIES
INDIAN INSTITUTE OF TECHNOLOGY
NEW DELHI
Online Buying Behavior Page 2
Table of Contents
Executive Summary ........................................................................................................................... 3
Introduction ...................................................................................................................................... 4
Objectives ...................................................................................................................................... 4
Primary Research Objective ....................................................................................................... 4
Secondary Research Objectives .................................................................................................. 4
Methodology ..................................................................................................................................... 5
Survey Administration .................................................................................................................... 5
Sampling ....................................................................................................................................... 5
Data Reduction .............................................................................................................................. 5
Data Analysis ................................................................................................................................. 6
Findings ............................................................................................................................................. 7
CROSS TABULATIONS ..................................................................................................................... 7
REGRESSION ANALYSIS ................................................................................................................ 15
ANOVA ........................................................................................................................................ 18
CLUSTER ANALYSIS....................................................................................................................... 19
DISCRIMINANT WITH CLUSTER ANALYSIS ..................................................................................... 22
FACTOR ANALYSIS ........................................................................................................................ 24
Conclusions ..................................................................................................................................... 28
Annexures ....................................................................................................................................... 30
Annexure 1a: Agglomeration Schedule for Cluster Analysis ......................................................... 31
Annexure 1b: Correlation Matrix for Factor Analysis .................................................................... 33
Annexure 1c: ANOVA Table for Regression Analysis ..................................................................... 36
Annexure 2a: Questionnaire for Exploratory Research ................................................................. 37
Annexure 2b: Final Questionnaire ................................................................................................ 38
Online Buying Behavior Page 3
Executive Summary
The project focused on finding out the Consumer Behaviour Pattern On Online Buying The
Cloths. The stated objective of the study was further broken down to secondary objectives
which aimed at finding information regarding the cloths purchased by student,frequency of
purchases, average spending, factors affecting online buying decision process etc.
The exploratory research was carried out with student studying in IITD with a set of 10 open
ended questions. The exploratory findings helped us in determining the key factors which
needed to be further explored for research. The secondary research was taken from
sources like Indian Journal of Management Technology, Zinnov LLC, and ACNielson. The
questionnaire designed had 9 questions and was administered to 106 respondents. Each of
the questions was designed to satisfy at least one of the secondary objectives of the
research. The response format was of a mixed variety which also helped in better
determination of outcomes.
Post data reduction, Cross tabulation was used for analyzing the causal relationship
between different pairs of factors. ANOVA was also applied to a pair of factors.
The Regression Analysis between the dependent variable “Average Amount spent per purchase of
cloths online” and the independent variables of Frequency of Purchase of products and services
online, owning a Credit Card, Marital Status, Education and Age, was done. The regression model
did not give any significant correlation between the factors and the Dependent Variable.
Although there is a strong interdependence between a few variables yet when taken
collectively they do not show high correlation.
Then, Cluster Analysis was done on the data and based on the responses; we could divide the
respondents in three clearly distinct groups. We named them: Confident Online Buyer, Unsure surfer
and Mall Shopper. We also performed Discriminant with Cluster Analysis to predict cluster
membership of consumers based on their attitude towards online shopping.
We performed Factor Analysis to find the major factors. We could identify six factors: Value
for Money, Trust, Connected and Up to date, Problems Faced, and Traditionalism.
Online Buying Behavior Page 4
Introduction
India has the world’s 4
th
largest Internet user base, which crossed the 100 million mark recently.
Better connectivity, booming economy and higher spending power helped the Indian e-commerce
market revenues to cross $500 million with a CAGR of 103%over last 4 years. This may not be a
significant number, averaging to only around $5 per user per year.
With the above background in mind, this research has been conducted to gain an insight into the
online buying behaviour of consumers. The objective is to explore the factors which influence online
purchase, the psychographic profile of the consumer groups and understanding the buying decision
process.
Our findings should help an Internet Marketer to determine the product/service categories to be
introduced or to be used for marketing for a specific segment of consumers. This would also allow
them to add or remove services/features which are important in the buying decision process. This
study however does not aim to identify newer areas to introduce new services, nor should it be used
to predict the success or failure of internet ventures.
Objectives
Primary Research Objective
To determine the factors and attributes which influence online buying behavior of consumers
between the age group of 18-30 years.
Secondary Research Objectives
1. To determine the psychographic profile of consumers who purchase over the Internet.
2. To determine the key product or service categories opted for, by consumers depending on
their profile.
3. To determine the factors which influence the buying decision process of a consumer.
4. To determine the average spending and frequency of purchase over the internet by a
consumer.
The exploratory research, conducted on over 12 respondents (Annexure I), focused on further
analysing the research objectives and also determining various factors which would impact the
primary research objective. Through a set of 12 open-ended questions, we could finally conclude on
some of the key factors to be further explored in the research, these included frequency of
purchase, safety issues, amount per purchase, payment methods etc…
Online Buying Behavior Page 5
Secondary Research was based on researches done by Zinnov LLC on Internet Penetration in India,
Changing Consumer Perceptions towards Online shopping in India – IJMT. Both of the researches
stressed on the consumer profiles, popular services and payments methods as important factors.
Methodology
The research was administered both online and in person during a 5 day period in February 2008.
The location of in-person administration was SIC Campus, Pune. Over 81 responses are from the
online survey and the rest 24 from in-person survey conducted.
Survey Administration
The questionnaire comprised of 9 questions (Annexure II) which measured responses for different
factors of frequency of purchase, payment methods, preferred products, average spending, hours
spent on the internet etc…
The questions measuring respondent attitudes used Likert Scale (1-5), 18 statements were given to
respondents to measure their attitudes towards online buying, and a few factual questions had
dichotomous responses.
The methods used for survey was questionnaire administration with respondents filling out the
responses themselves and online survey on SurveyGizmo.com
Sampling
The survey was conducted on 105 respondents; sample was based on affordability criteria especially
on time constraints. Email invitations were sent to invite respondents on the Internet, and students
in SIC Campus were contacted for responses.
Gender
65%
35%
Male
Female
Occupation
40%
60%
Student
Working Professional
Data Reduction
The key steps of data processing which were implemented were Editing, Coding, Transcribing, and
Summarizing statistical calculations.
Online Buying Behavior Page 6
EDITING: For some of the item non-response errors like frequency of purchase, product category or
websites. The data was interpreted and assigned to the known categories wherever possible.
CODING: For questions involving qualitative values the responses were codified using numerical
categories or values. For example; Online shopping is more convenient, the response of “strongly
agree” was coded as 1 and “strongly disagree” was coded as 5.
TRANSCRIBING: The data collected from all 105 questionnaires was edited, codified and finally
transferred on MS Excel on computer.
Data Analysis
Post Data Reduction, the data was further used for analyzing the impact of various factors on each
other as well the correlation amongst them using SPSS. The factors as well as their correlation were
studied with the help of the following techniques:
CROSS-TABS WITH CHI-SQUARE: The factors were grouped into 5 pairs based on the responses from
the questionnaire. These were studied using Chi-Square as that would help us to know the
interdependency between them. Chi-square in general studies causal relationship and thus the
hypotheses were created for each of them was done at 95%significance level. By conducting the
test and interpreting the results through the p-value, we can either accept or not accept the null
hypothesis.
REGRESSION ANALYSIS: In regression analysis, we create a model wherein we determine the
correlation between the dependent variable and multiple independent variables. By conducting the
tests and interpreting the results, we can determine the adjusted R
2
value which tells us how good
the regression model fits to the data. If the value is high, then the model fits well to the data and
that there is a high correlation between the variables. On the other hand, if the value is low, then
the model does not fit very well to the data and there is no significant correlation between the
variables.
ANOVA: Analysis of variance, better known as ANOVA, helps us to group the data into various
population samples and then check their relationship with an independent variable, which we
consider to be significant depending on the responses from the questionnaire. The null hypothesis
for this is also created at a 95%significant variable and then depending on the significant value from
the results, the hypothesis is accepted or not accepted.
CLUSTER ANALYSIS: This technique is used for segmentation of consumers on the basis of
similarities between them. The similarities could be of demographics, buying habits, or
psychographics. Hierarchical clustering is used to find the initial cluster solution, and K Means is later
used to determine cluster membership of respondents, cluster labelling is also done.
DISCRIMINANT ANALYSIS WITH CLUSTERS: A combination of Discriminant Analysis with clusters to
create a model which helps in predicting cluster membership of a consumer on the basis of the input
factors.
Online Buying Behavior Page 7
FACTOR ANALYSIS: This is a technique to reduce data complexity by reducing the number of
variables being studies. It helps identify latent or underlying factors from an array of seemingly
important variables. This procedure helps gaining insight into psychographic variables.
Findings
CROSS TABULATIONS
a) Credit Card- Frequency of Purchase
Null Hypothesis: At 95%significance level, owning a credit card does not have any impact on
the frequency of purchase.
Alternate Hypothesis: At 95%significance level, owning a credit card has an impact on the
frequency of purchase.
OwnCreditCard * FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once
a
Month
2 -3
Times
a
Month
Once in
3
Months
Once in
6
Months
Never
Tried
Once
a
Month
OwnCreditCard
Yes 23 14 28 11 1 77
No 3 2 11 10 5 31
Total 26 16 39 21 6 108
Online Buying Behavior Page 8
As the p-value from the table is lesser than 0.05, which is our assumed level of significance, we do
not accept the null hypothesis, that is, for the sample population, owninga credit card has an
impact on the frequency of purchase.
b) E-banking-Frequency of Purchase
Chi-Square Tests
Value df
Asymp. Sig.
(2-sided)
Pearson Chi-
Square
18.222(a) 4 .001
Likelihood Ratio 17.962 4 .001
Linear-by-Linear
Association
15.313 1 .000
N of Valid Cases 108
a 3 cells (30.0%) have expected count less than
5. The minimum expected count is 1.72.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N Percent
OwnCreditCard
*
FreqofPurchase
108 100.0% 0 .0% 108 100.0%
Online Buying Behavior Page 9
Null Hypothesis: At 95%significance level, e-banking does not have any impact on the
frequency of purchase.
Alternate Hypothesis: At 95%significance level, e-banking has an impact on the frequency
of purchase.
E_banking* FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once
a
month
2-3
times
a
month
Once in
3
months
Once in
6
months
Never
tried
Once
a
month
E_banking
Yes 22 15 32 11 0 80
No 3 1 6 10 7 27
Total 25 16 38 21 7 107
Chi-Square Tests
Value df
Asymp.
Sig. (2-
sided)
Pearson Chi-
Square
33.492(a) 4 .000
Likelihood Ratio 32.845 4 .000
Linear-by-Linear
Association
20.737 1 .000
N of Valid Cases 107
a 2 cells (20.0%) have expected count less
than 5. The minimum expected count is
1.77.
Case ProcessingSummary
Online Buying Behavior Page 10
As the p-value from the table is lesser than
0.05, which is our assumed level of
significance, we do not accept the null
hypothesis, that is, for the sample
population, E-bankinghas an impact on the
frequency of purchase.
c) Gender-Amount Spent
Null Hypothesis: At 95%significance level, gender does not have any impact on the average
amount spent per purchase made online.
Alternate Hypothesis: At 95%significance level, e-banking has an impact on the average
amount spent per purchase made online.
Cases
Valid Missing Total
N Percent N Percent N Percent
E_banking*
FreqofPurchase
107 100.0% 0 .0% 107 100.0%
Chi-Square Tests
Value df
Asymp. Sig.
(2-sided)
Pearson Chi-Square 2.789(a) 4 .594
Likelihood Ratio 2.811 4 .590
Linear-by-Linear
Association
.003 1 .955
N of Valid Cases 106
a 1 cells (10.0%) have expected count less than 5.
The minimum expected count is 4.81.
Gender * AmountSpent Crosstabulation
Count
AmountSpent Total
Less
than
500
500
-
1000
1000
-
2000
2000
-
5000
Greater
than
5000
Less
than
500
Gender
Male 11 13 9 27 12 72
Female 7 4 6 9 8 34
Total 18 17 15 36 20 106
Online Buying Behavior Page 11
As the p-value fromthe table is
greater than 0.05, which is our
assumed level of significance, we
accept the null hypothesis, that is,
for the sample population; gender
does not have any impact on the
average amount spent per purchase
made online.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N
Perc
ent
Gender *
AmountSpent
106 100.0% 0 .0% 106
100.
0%
Online Buying Behavior Page 12
d) Gender-Frequency of Purchase
Null Hypothesis: At 95%significance level, gender does not have any impact on the
frequency of purchase of online products and services.
Alternate Hypothesis: At 95%significance level, gender has an impact on the frequency of
purchase of online products and services.
As the p-value from the table is lesser than 0.05, which is our assumed level of significance, we do
not accept the null hypothesis, that is, for the sample population; gender has an impact on the
frequency of purchase of online products and services.
Gender * FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once
a
Month
2-3
Times
a
Month
Once in
3
Months
Once in
6
Months
Never
Tried
Once
a
Month
Gender
Male 20 14 27 9 3 73
Female 4 2 13 11 3 33
Total 24 16 40 20 6 106
Chi-Square Tests
Value df
Asymp.
Sig. (2-
sided)
Pearson Chi-Square 11.278(a) 4 .024
Likelihood Ratio 11.499 4 .021
Linear-by-Linear
Association
9.084 1 .003
N of Valid Cases 106
a 3 cells (30.0%) have expected count less
than 5. The minimum expected count is 1.87.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N Percent
Gender *
FreqofPurchase
106 100.0% 0 .0% 106 100.0%
Online Buying Behavior Page 13
e) Income-Frequency of Purchase
Null Hypothesis: At 95%significance level, income of respondents does not have any impact
on the frequency of purchase of online products and services.
Alternate Hypothesis: At 95%significance level, income of respondents has an impact on
the frequency of purchase of online products and services.
Income * FreqofPurchase Crosstabulation
Count
FreqofPurchase Total
Once a
Month
2-3
Times a
Month
Once in 3
Months
Once in 6
Months
Never
Tried
Once a
Month
Income
Less than
10000
2 0 1 0 0 3
10000-
20000
1 0 2 5 1 9
20000-
30000
2 2 11 2 0 17
30000-
50000
5 0 3 1 0 9
50000-
100000
2 1 0 1 0 4
Greater
than
100000
1 1 2 0 0 4
Total 13 4 19 9 1 46
Online Buying Behavior Page 14
As the p-value from the table is greater than 0.05, which is our assumed level of significance, we
do not accept the null hypothesis, that is, for the sample population; income does not have an
impact on the frequency of purchase of online products and services.
Chi-Square Tests
Value df
Asymp. Sig. (2-
sided)
Pearson Chi-Square 28.966(a) 20 .088
Likelihood Ratio 29.758 20 .074
Linear-by-Linear
Association
2.806 1 .094
N of Valid Cases 46
a 29 cells (96.7%) have expected count less than 5. The
minimum expected count is .07.
Case ProcessingSummary
Cases
Valid Missing Total
N Percent N Percent N Percent
Income *
FreqofPurchase
46 100.0% 0 .0% 46 100.0%
Online Buying Behavior Page 15
REGRESSION ANALYSIS
The Regression Analysis between the dependent variable “Average Amount spent per purchase
made online” and the independent variables of Frequency of Purchase of products and services
online, owning a Credit Card, Marital Status, Education and Age, was done using SPSS. The details
are as below:
Variables Entered/Removed(b)
Model Variables Entered
Variables
Removed
Method
1
MaritalStatus, FreqofPurchase,
Education, CreditCard, Age(a)
. Enter
2
Age
Backward (criterion: Probability of
F-to-remove >=.100).
3 . Education
Backward (criterion: Probability of
F-to-remove >=.100).
4 . MaritalStatus
Backward (criterion: Probability of
F-to-remove >=.100).
a All requested variables entered.
b Dependent Variable: AmtSpent
Online Buying Behavior Page 16
Coefficients(a)
Model
Unstandardized Coefficients Standardized Coefficients t Sig.
B Std. Error Beta B Std. Error
1
(Constant) 1.696 1.954
.868 .388
FreqofPurchase .402 .122 .330 3.305 .001
Age .054 .078 .083 .696 .489
CreditCard -.695 .318 -.234 -2.186 .032
Education -.152 .202 -.076 -.753 .454
MaritalStatus .384 .464 .096 .828 .410
2
(Constant) 2.897 .912
3.178 .002
FreqofPurchase .403 .121 .331 3.323 .001
CreditCard -.755 .305 -.254 -2.477 .015
Education -.134 .199 -.067 -.671 .504
MaritalStatus .534 .409 .134 1.304 .196
3
(Constant) 2.564 .762
3.364 .001
FreqofPurchase .415 .120 .341 3.467 .001
CreditCard -.772 .303 -.259 -2.547 .013
MaritalStatus .561 .406 .140 1.380 .171
4
(Constant) 3.366 .496
6.781 .000
FreqofPurchase .408 .120 .335 3.393 .001
CreditCard -.882 .294 -.297 -3.005 .003
a Dependent Variable: AmtSpent
Online Buying Behavior Page 17
As can be seen from the above table, the independent variables can be gradually removed in the
regression model as they don’t have any significant impact on the value of R
2
. The value of R
2
is quite
low and so it can be said that the regression model does not fit into the data very well. Also, the sum
of squares of regression is lesser than the sum of squares of residuals and this reiterates the findings
of R
2
. This is because if the sum of squares of regression is lesser than the sum of squares of
residuals, then the independent variables do not explain the variation in the dependent variable
well. While cross tabs suggest a positive relationship between multiple pairs of factors, the linear
correlation model, with all factors together, does not fit in with the outcomes.
Excluded Variables(d)
Model
Beta In t Sig. Partial Correlation Collinearity Statistics
Tolerance Tolerance Tolerance Tolerance Tolerance
2 Age .083(a) .696 .489 .077 .682
3
Age .071(b) .606 .546 .067 .694
Education -.067(b) -.671 .504 -.074 .962
4
Age .122(c) 1.164 .248 .127 .874
Education -.080(c) -.798 .427 -.087 .971
MaritalStatus .140(c) 1.380 .171 .150 .926
a Predictors in the Model: (Constant), MaritalStatus, FreqofPurchase, Education, CreditCard
b Predictors in the Model: (Constant), MaritalStatus, FreqofPurchase, CreditCard
c Predictors in the Model: (Constant), FreqofPurchase, CreditCard
d Dependent Variable: AmtSpent
Online Buying Behavior Page 18
ANOVA
Null hypothesis: At 95%confidence interval for the population taken, income does not have any
impact on the frequency of purchase of online products and services.
Alternate Hypothesis: At 95%confidence interval for the population taken, income has an impact on
the frequency of purchase of online products and services.
ANOVA
Frequency
Sum of
Squares df
Mean
Square F Sig.
Between
Groups
5.317 6 .886 .605 .726
Within Groups 149.417 102 1.465
Total 154.734 108
N Mean
Std.
Deviation
Std.
Error
95%Confidence
Interval for Mean
Minim
um
Maxim
um
Lower
Bound
Upper
Bound
.00 64 2.3125 1.29560 .16195 1.9889 2.6361 .00 4.00
Less than
10000
3 2.3333 .57735 .33333 .8991 3.7676 2.00 3.00
10000-20000 7 2.8571 1.46385 .55328 1.5033 4.2110 .00 4.00
20000-30000 16 2.7500 .85635 .21409 2.2937 3.2063 1.00 4.00
30000-50000 7 2.7143 .75593 .28571 2.0152 3.4134 2.00 4.00
50000-100000 5 2.0000 1.22474 .54772 .4793 3.5207 1.00 4.00
Greater than
100000
7 2.4286 1.27242 .48093 1.2518 3.6054 .00 4.00
Total 109 2.4312 1.19696 .11465 2.2039 2.6584 .00 4.00
Online Buying Behavior Page 19
Means Plots
Income
Greater than
100000
50000-
100000
30000-50000 20000-30000 10000-20000 Less than
10000
.00
M
e
a
n
o
f
F
r
e
q
u
e
n
c
y
3.00
2.80
2.60
2.40
2.20
2.00
The p-value from the ANOVA table is greater than the significance value of 0.05 assumed by us.
Thus, at this significance level we accept the null hypothesis. So we can conclude that income does
not have an impact on the frequency of purchase of online products and services for these
respondents. Do remember that the same conclusion was arrived at when Cross tabulation of
location and usage rate was performed earlier.
CLUSTER ANALYSIS
The cluster analysis was run where people were surveyed about their attitudes towards internet
shopping. The preferences indicated by respondents were used to find out the consumer segments
that react differently to different parameters related to online shopping. The segments obtained
would give an understanding as to how the consumers are placed in terms of their attitudes.
Hierarchical clustering was done to determine the initial cluster solution. While the initial cluster
solution by SPSS gives us 2 clusters, we take a difference of coefficients greater than or equal to 2.72
form another cluster. Now we execute the K-Mean cluster to get the final cluster solution and
through ANOVA table we get that all the variables bear significance at 95%confidence level.
Online Buying Behavior Page 20
ANOVA
Clust er Error F Sig.
Mean Square df Mean Square Df
Mean
Square df
InternetvsMall
27.190 2 .515 103 52.814 .000
LatestInfo
1.744 2 .341 103 5.116 .008
Accesibility
6.364 2 .536 103 11.874 .000
Convenience
27.439 2 .367 103 74.819 .000
Savetime
7.485 2 .608 103 12.312 .000
AnywhereAnytime
2.935 2 .559 103 5.255 .007
CreditCardSafe
3.441 2 .619 103 5.562 .005
SpecificDateTime
6.393 2 .553 103 11.554 .000
GuaranteedQuality
4.378 2 .469 103 9.334 .000
Discounts
9.581 2 .710 103 13.500 .000
Hasslefree
8.649 2 .691 103 12.518 .000
CashonDelivery
1.011 2 .791 103 1.278 .283
EasyFind
5.585 2 .838 103 6.668 .002
FacedProblems
.866 2 .730 103 1.186 .309
Continue
6.682 2 .823 103 8.121 .001
TouchandFeel
5.330 2 .728 103 7.320 .001
DeliveryProcess
8.284 2 .499 103 16.612 .000
NoCreditCard
12.382 2 .950 103 13.035 .000
The F tests should be used only for descriptive purposes because the clusters have been chosen to maximize
the differences among cases in different clusters. The observed significance levels are not corrected for this and
thus cannot be interpreted as tests of the hypothesis that the cluster means are equal.
Final Clust er Cent ers
Clust er
1 2 3
InternetvsMall
3.38 2.09 4.00
LatestInfo
1.58 1.78 2.05
Accesibility
1.37 1.94 2.18
Convenience
2.62 1.81 3.86
Savetime
2.33 1.59 2.55
AnywhereAnytime
1.88 2.09 2.50
CreditCardSafe
2.52 2.78 3.18
SpecificDateTime
2.10 2.28 3.00
GuaranteedQuality
2.98 3.56 3.55
Discounts
2.15 2.72 3.23
Hasslefree
2.54 2.03 3.18
CashonDelivery
2.15 2.34 2.50
EasyFind
2.00 2.63 2.68
FacedProblems
2.52 2.81 2.59
Continue
2.40 2.94 3.27
TouchandFeel
1.81 2.53 1.95
DeliveryProcess
2.42 2.66 3.45
NoCreditCard
4.08 3.81 2.82
Online Buying Behavior Page 21
Variable Description Cluster 1 (52) Cluster 2(32) Cluster 3(22)
I prefer making a purchase from internet than using local
malls or stores
Disagree Strongly Agree Strongly
Disagree
I can get the latest information from the Internet regarding
different products/ services that is not available in the
market.
Strongly Agree NAND Strongly
Disagree
I have sufficient internet accessibility to shop online. Strongly Agree Mildly Disagree Strongly
Disagree
Online shopping is more convenient than in-store
shopping.
Mildly Agree Strongly Agree Strongly
Disagree
Online shopping saves time over in-store shopping. Disagree Strongly Agree Strongly
Disagree
Online shopping allows me to shop anywhere and at
anytime.
Strongly Agree Mildly Agree Strongly
Disagree
It is safe to use a credit card while shopping on the
Internet.
Strongly Agree NAND Strongly
Disagree
Online shopping provides me with the opportunity to get
the products delivered on specific date and time anywhere
as required.
Strongly Agree Moderately Agree Strongly
Disagree
Products purchased through the Internet are with
guaranteed quality.
Strongly Agree Strongly Disagree Strongly
Disagree
Internet provides regular discounts and promotional
offers to me.
Strongly Agree NAND Strongly
Disagree
Internet helps me avoid hassles of shopping in stores. NAND Strongly Agree Strongly
Disagree
Cash on Delivery is a better way to pay while shopping on
the Internet.
Strongly Agree NAND Strongly
Disagree
Sometimes, I can find products online which I may not find
in-stores.
Strongly Agree Strongly Disagree Strongly
Disagree
I have faced problems while shopping online. Strongly Agree Strongly Disagree Agree
I continue shopping online despite facing problems on
some occasions.
Strongly Agree Mildly Disagree Strongly
Disagree
It is important for me to touch and feel certain products
before I purchase them. So I cannot buy them online.
Strongly Agree Strongly Disagree Agree
I trust the delivery process of the shopping websites. Strongly Agree Agree Strongly
Disagree
I do not shop online only because I do not own a credit
card.
Strongly Disagree Disagree Strongly Agree
Online Buying Behavior Page 22
On the basis of the above scales, obtained by rating the relative results, we can name our clusters as:
? Cluster 1 : Confident Online Buyer
? Cluster 2 : Unsure surfer
? Cluster 3 : Mall Shopper
DISCRIMINANT WITH CLUSTER ANALYSIS
By combining cluster analysis with Discriminant analysis, we could derive a model which could
predict cluster membership of the respondents on the basis of the variables analysed.
The cluster solution is changed with K Means cluster>Save option selected for Cluster Membership.
This gives a new column in the Data Sheet, showing the cluster membership of the responses. Using
this data sheet, Discriminant analysis is done over the cluster membership column as the dependent
variables.
Eigenvalues
a First 2 canonical discriminant functions were used in the
analysis.
Wilks' Lambda
Test of
Function(s)
Wilks'
Lambda
Chi-
square df Sig.
1 through 2 .072 248.627 36 .000
2
.289 117.402 17 .000
Canonical Discriminant Function Coeff icients
Unstandardized coefficients
The Eigenvalues are greater than 1, and the Wilk’s Lamba
is below 0.5, this indicates that the Discriminant Model is
able to explain the data fairly well.
The Canonical Discriminant Function Coefficients table contains Group-1 number of functions; the
equations as derived from above are:
F
1
=.901(InternetvsMall) - .090 LatestInfo - .122 Accesibility -.989 Convenience +.370 Savetime - .228
AnywhereAnytime – 0.080 CreditCardSafe +.028 SpecificDateTime - .621 GuarenteeQuality +.045
Discounts - .097 Hasslefree +.193 CoD - .038 EasyFind +0.029 FacedProblems - .198 Continue - .447
TouchandFeel +.308 DeliveryProcess +.008 NoCreditCard – 3.372
Functio
n
Eigenvalu
e
% of
Variance
Cumulative
%
Canonical
Correlation
1 3.009(a) 55.0 55.0 .866
2 2.464(a) 45.0 100.0 .843
Function
1 2
InternetvsMall
.901 -.372
LatestInfo
-.090 .216
Accesibility
-.122 .346
Convenience
.989 .750
Savetime
.370 -.747
AnywhereAnytime
-.228 .185
CreditCardSafe
-.080 .241
SpecificDateTime
.028 .510
GuaranteedQuality
-.621 .543
Discounts
.045 .232
Hasslefree
.097 .180
CashonDelivery
.193 .262
EasyFind
-.038 .162
FacedProblems
.029 .272
Continue
-.198 .179
TouchandFeel
-.447 .623
DeliveryProcess
.308 .262
NoCreditCard
.008 -.609
(Constant)
-3.372 -7.122
Online Buying Behavior Page 23
F
2
=-.372 InternetvsMall +.216 LatestInfo +.346 Accesibility +.750 Convenience - .747 Savetime +
.185 AnywhereAnytime +.241 CreditCardSafe +.510 SpecificDateTime +.543 GuarenteeQuality -
.232 Discounts +.180 HassleFree +.262 CoD +.162 EasyFind +.272 FacedProblems +.179 Continue +
.623 TouchnFeel +.262 DeliveryProcess - .609 NoCreditCard – 7.122
Classif ication Function Coeff icients
Cluster Number of Case
1 2 3
InternetvsMall
6.099 2.479 5.961
LatestInfo
10.113 10.871 10.813
Accesibility
-4.203 -3.056 -3.047
Convenience
5.202 3.798 9.499
Savetime
.077 -2.731 -2.261
AnywhereAnytime
1.323 2.440 1.705
CreditCardSafe
5.904 6.687 6.715
SpecificDateTime
4.090 5.135 6.087
GuaranteedQuality
4.190 7.322 5.385
Discounts
1.330 1.707 2.285
Hasslefree
2.116 2.215 2.944
CashonDelivery
8.338 8.321 9.619
EasyFind
1.519 1.999 2.087
FacedProblems
5.910 6.424 6.994
Continue
-.681 .331 -.279
TouchandFeel
6.076 8.846 7.825
DeliveryProcess
2.462 2.087 3.909
NoCreditCard
3.878 2.501 1.551
(Constant)
-79.869 -87.731 -116.575
Fisher's linear discriminant functions
The above table provides the linear Discriminant function which can be used to judge the membership of each
data set by putting the value in each of them, and choosing the one which provides the highest value.
Classif ication Result s(a)
Cluster Number of Case
Predicted Group Membership Total
1 2 3 1
Original Count 1
51 1 0 52
2
0 32 0 32
3
1 0 21 22
% 1
98.1 1.9 .0 100.0
2
.0 100.0 .0 100.0
3
4.5 .0 95.5 100.0
a 98.1% of original grouped cases correctly classified.
Our Discriminant Model is able to explain 98.1%of the data set correctly, which is significantly high,
and indicates that this is a dependable Discriminant Model for prediction.
Online Buying Behavior Page 24
FACTOR ANALYSIS
The responses are put into SPSS for data reduction through Factor Analysis. The details of the above
are provided below:
Total Variance Explained
Component
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings
Total %of Variance Cumulative % Total %of Variance Cumulative % Total %of Variance Cumulative %
1 3.854 21.410 21.410 3.854 21.410 21.410 2.559 14.219 14.219
2 2.175 12.082 33.491 2.175 12.082 33.491 2.318 12.879 27.098
3 1.652 9.175 42.666 1.652 9.175 42.666 2.160 11.997 39.095
4 1.514 8.413 51.080 1.514 8.413 51.080 1.727 9.593 48.688
5 1.277 7.096 58.176 1.277 7.096 58.176 1.422 7.901 56.589
6 1.098 6.101 64.276 1.098 6.101 64.276 1.243 6.903 63.492
7 1.066 5.924 70.200 1.066 5.924 70.200 1.207 6.708 70.200
8 .875 4.859 75.059
9 .771 4.283 79.342
10 .737 4.095 83.438
11 .545 3.030 86.467
12 .529 2.939 89.407
13 .384 2.134 91.541
14 .377 2.092 93.633
15 .358 1.988 95.621
16 .293 1.626 97.247
17 .278 1.546 98.793
18 .217 1.207 100.000
Extraction Method: Principal Component Analysis.
Online Buying Behavior Page 25
We see from the Cumulative Percentage column that there have been seven components or factors
extracted which explain 70.2%of the total variance (information contained in the original 18
variables). This is an acceptable solution as generally, 70%of the total variance should be explained
by the factors for the solution to be accepted.
Rotated Component Matrix(a)
Component
1 2 3 4 5 6 7
InternetvsMall .714 -.279 .149 .056 .046 -.197 -.241
LatestInfo .099 .322 -.146 .804 -.072 -.125 -.105
Accesibility -.043 .157 .158 .859 .105 .084 .020
Convenience .796 .045 .202 .206 .030 -.078 -.109
Savetime .726 .270 -.040 -.015 .017 -.120 .065
AnywhereAnytime .314 .571 .153 .109 .251 .025 -.034
CreditCardSafe .023 -.046 .657 -.047 .090 -.271 -.416
SpecificDateTime .272 -.198 .466 .404 -.197 .321 .107
GuaranteedQuality .023 .412 .647 .016 .071 -.124 .309
Discounts .069 .714 .208 .233 .095 -.095 -.158
Hasslefree .694 .225 -.022 -.153 -.129 .320 -.017
CashonDelivery -.121 -.024 -.022 .008 .123 .882 -.126
EasyFind .013 .869 .018 .143 -.043 .055 .012
FacedProblems -.203 .020 -.091 .084 .788 -.025 .049
Continue .071 .379 .518 -.040 .555 .015 -.077
TouchandFeel -.351 -.060 -.006 .079 -.546 -.193 -.065
DeliveryProcess .112 .135 .796 .065 -.101 .126 -.059
NoCreditCard -.142 -.130 -.047 -.053 .084 -.147 .885
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 8 iterations.
Online Buying Behavior Page 26
Looking at the rotated component matrix, we see that the loadings of variables-Internet preferred
over mall, Convenience and Saves Time, on factor 1 are high (greater than 0.7) and thus factor 1 is
made up of these variables. Similarly, we can interpret for the other factors as well and a table for
the same has been provided at the end of this section.
Factor Variables Label for the Factor
1 Internet over Mall
Convenience
Saves Time
Time-bound and comfort seeking
2 Discounts
Easy Find
Value for Money
3 Delivery Process
Credit Card Safe to Use
Guaranteed Quality
Trust
4 Latest Information
Accessibility
Connected and Up to date
5 Faced Problems Problems Faced
6 Cash on Delivery
Don’t Own a credit card
Traditionalism
7 Don’t Own a credit card N/A
We have combined the sixth and the seventh factor as they go hand-in-hand. A person who does not
own a credit card but shops online would invariably prefer to pay by cash on delivery.
PERCEPTUAL MAPS
Online Buying Behavior Page 27
Online Buying Behavior Page 28
Others:
Conclusions
We found a strong inter-dependence between a few variables affecting online buying behaviour. For
example, we found that owning a credit card has a significant impact on the frequency of online
purchases as credit card is the most popular mode of payment on the Internet. Apart from the credit
card, E-Banking is also slowly becoming a popular mode of payment and we found a relationship
between people who use E-Banking and their frequency of online purchases too.
Interestingly, we found that gender does not have any major impact on the average amount spent
over the Internet in a month, but it does have a relationship with the frequency of purchases. Also,
the income of an individual does not have show any significant relationship with the frequency of
purchases. These findings are starkly similar to the findings of Changing Consumer Perceptions
towards Online shopping in India – IJMT, which was a part of our secondary data.
Based on the responses, we could divide the respondents in three clearly distinct groups. We named
them: Confident Online Buyer, Unsure surfer and Mall Shopper. We were also able to successfully
able to create a discriminant model which could predict cluster membership of users.
We could also arrive at six factors which can explain the data with 70%significance, these factors
could be categorised into Time-bound and comfort seeking Value for Money, Trust, Connected and
Up to date, Problems Faced, and Traditionalism.
We also found that the most popular product category sold online is Air/Rail Tickets. This forms a
major chunk of the average amount spent by our respondents on the internet, Books come a close
second. It must be noted that both the above products have a relatively low touch-and-feel need.
Online Buying Behavior Page 29
These findings depict almost the same ranking as found by ACNielson on popular services on the
Internet.
The most popular websites for these were found to be Makemytrip.com and Yatra.com. Apart from
Air Tickets, Books, Gifts and Electronic Products are also very popular with the Online Shoppers and
they are spending, on an average, Rs2000-Rs5000 per month on online purchases.
Online Buying Behavior Page 30
Annexures
Online Buying Behavior Page 31
Annexure 1a: Agglomeration Schedule for Cluster Analysis
Agglomeration Schedule
Stage
Cluster
Combined
Coefficients
Stage Cluster
First Appears
Next
Stage
Cluster
1
Cluster
2
Cluster
2
Cluster
1
1
105 106 0.000 0 0 2
2 16 105 0.000 0 1 4
3
103 104 0.000 0 0 4
4
16 103 0.000 2 3 6
5
101 102 0.000 0 0 6
6
16 101 0.000 4 5 8
7 99 100 0.000 0 0 8
8
16 99 0.000 6 7 10
9
97 98 0.000 0 0 10
10
16 97 0.000 8 9 72
11
19 85 3.000 0 0 31
12 1 91 4.000 0 0 53
13 84 86 5.000 0 0 32
14
31 49 5.000 0 0 44
15
92 94 6.000 0 0 20
16
6 93 6.000 0 0 26
17
25 89 6.000 0 0 36
18 43 81 6.000 0 0 45
19
21 65 6.000 0 0 32
20
24 92 7.000 0 15 55
21
52 90 7.000 0 0 48
22
45 46 7.000 0 0 44
23 27 44 7.000 0 0 49
24
70 87 8.000 0 0 35
25
39 82 8.000 0 0 46
26
6 69 8.000 16 0 37
27
42 64 8.000 0 0 70
28 26 58 8.000 0 0 57
29 2 37 8.000 0 0 49
30
15 33 8.000 0 0 61
31
19 47 8.500 11 0 36
32
21 84 9.000 19 13 54
33
30 83 9.000 0 0 64
34 73 74 9.000 0 0 63
35
34 70 9.000 0 24 54
36
19 25 9.333 31 17 55
37
6 88 9.667 26 0 50
38
4 78 10.000 0 0 53
39 54 55 10.000 0 0 67
40
12 53 10.000 0 0 68
41
5 41 10.000 0 0 45
42
7 20 10.000 0 0 65
43
76 79 11.000 0 0 83
44 31 45 11.000 14 22 52
45 5 43 11.000 41 18 73
Online Buying Behavior Page 32
46
22 39 11.000 0 25 58
47 3 23 11.000 0 0 79
48
52 60 11.500 21 0 63
49
2 27 11.500 29 23 62
50
6 32 11.750 37 0 58
51
18 71 12.000 0 0 94
52 31 61 12.000 44 0 62
53 1 4 12.000 12 38 67
54
21 34 12.250 32 35 61
55
19 24 12.333 36 20 68
56
17 67 13.000 0 0 87
57 26 56 13.000 28 0 64
58 6 22 13.933 50 46 73
59
50 63 14.000 0 0 66
60
35 48 14.000 0 0 70
61
15 21 14.286 30 54 69
62
2 31 14.450 49 52 71
63 52 73 14.500 48 34 72
64
26 30 14.500 57 33 74
65
7 59 15.000 42 0 81
66
36 50 15.000 0 59 86
67
1 54 15.250 53 39 78
68 12 19 15.250 40 55 71
69 15 38 15.889 61 0 77
70
35 42 16.000 60 27 85
71
2 12 16.522 62 68 74
72
16 52 17.200 10 63 84
73 5 6 17.375 45 58 77
74 2 26 17.653 71 64 78
75
28 75 18.000 0 0 95
76
29 40 18.000 0 0 88
77
5 15 18.017 73 69 82
78
1 2 18.458 67 74 82
79 3 72 18.500 47 0 90
80
11 57 19.000 0 0 100
81
7 62 19.333 65 0 83
82
1 5 20.361 78 77 85
83
7 76 20.750 81 43 87
84 16 51 21.750 72 0 89
85 1 35 22.029 82 70 88
86
9 36 22.333 0 66 92
87
7 17 22.667 83 56 93
88
1 29 23.786 85 76 89
89 1 16 24.603 88 84 92
90 3 80 24.667 79 0 93
91
14 96 26.000 0 0 94
92
1 9 27.223 89 86 95
93
3 7 27.250 90 87 97
94
14 18 28.000 91 51 96
95 1 28 29.506 92 75 96
96
1 14 30.753 95 94 97
97
1 3 32.106 96 93 98
98
1 10 34.825 97 0 100
99
8 68 36.000 0 0 103
100 1 11 37.082 98 80 101
Online Buying Behavior Page 33
101
1 95 38.550 100 0 102
102 1 13 43.089 101 0 103
103
1 8 43.637 102 99 104
104
1 77 48.423 103 0 105
105
1 66 66.914 104 0 0
Annexure 1b: Correlation Matrix for Factor Analysis
Correlation Matrix
Inter
netvs
Mall
Lat
estI
nfo
Acc
esib
ility
Con
veni
ence
Sav
eti
me
Anywh
ereAn
ytime
Credi
tCard
Safe
Specif
icDate
Time
Guara
nteed
Qualit
y
Dis
cou
nts
Has
slef
ree
Cash
onDe
livery
Ea
syF
ind
Face
dPro
blem
s
Co
nti
nu
e
Touc
hand
Feel
Deliv
eryPr
ocess
NoCr
edit
Card
Cor
rela
tion
Intern
etvsM
all
1.000
.05
3
-
.049
.569
.28
5
.134 .189 .192 -.020
-
.00
1
.34
1
-.176
-
.14
9
-.111
.02
3
-.174 .208 -.205
LatestI
nfo
.053
1.0
00
.602 .212
.13
2
.164 -.046 .125 .078
.37
4
.02
6
-.089
.37
9
-.047
.03
0
-.025 .016 -.151
Accesi
bility
-.049
.60
2
1.00
0
.159
.05
1
.234 .113 .255 .233
.27
7
-
.06
3
.116
.23
6
.101
.12
4
-.003 .161 -.057
Conve
nience
.569
.21
2
.159
1.00
0
.54
1
.260 .182 .264 .133
.25
2
.43
3
-.091
.09
3
-.115
.18
9
-.200 .297 -.190
Saveti
me
.285
.13
2
.051 .541
1.0
00
.325 .098 .089 .130
.16
8
.44
3
-.118
.18
1
-.103
.17
3
-.158 .023 -.080
Anywh
ereAny
time
.134
.16
4
.234 .260
.32
5
1.000 .136 .170 .263
.41
3
.21
2
-.031
.47
5
.086
.40
6
-.210 .142 -.135
Credit
CardSa
fe
.189
-
.04
6
.113 .182
.09
8
.136 1.000 .093 .310
.10
3
-
.03
1
-.092
-
.03
3
-.062
.33
5
-.005 .373 -.241
Specifi
cDateT
ime
.192
.12
5
.255 .264
.08
9
.170 .093 1.000 .113
.02
9
.17
0
.063
-
.04
7
-.142
.13
5
.051 .342 -.103
Guara
nteed
Qualit
y
-.020
.07
8
.233 .133
.13
0
.263 .310 .113 1.000
.33
3
.09
8
-.119
.31
4
.014
.44
6
-.109 .425 .104
Online Buying Behavior Page 34
Discou
nts
-.001
.37
4
.277 .252
.16
8
.413 .103 .029 .333
1.0
00
.14
3
-.073
.53
6
.129
.39
7
-.048 .341 -.185
Hasslef
ree
.341
.02
6
-
.063
.433
.44
3
.212 -.031 .170 .098
.14
3
1.0
00
.121
.14
9
-.112
.06
5
-.139 .104 -.187
Casho
nDeliv
ery
-.176
-
.08
9
.116
-
.091
-
.11
8
-.031 -.092 .063 -.119
-
.07
3
.12
1
1.000
.01
2
.066
.02
9
-.115 .050 -.119
EasyFi
nd
-.149
.37
9
.236 .093
.18
1
.475 -.033 -.047 .314
.53
6
.14
9
.012
1.0
00
.014
.28
5
-.076 .187 -.098
FacedP
roblem
s
-.111
-
.04
7
.101
-
.115
-
.10
3
.086 -.062 -.142 .014
.12
9
-
.11
2
.066
.01
4
1.000
.30
9
-.071 -.107 .118
Contin
ue
.023
.03
0
.124 .189
.17
3
.406 .335 .135 .446
.39
7
.06
5
.029
.28
5
.309
1.0
00
-.267 .344 -.120
Toucha
ndFeel
-.174
-
.02
5
-
.003
-
.200
-
.15
8
-.210 -.005 .051 -.109
-
.04
8
-
.13
9
-.115
-
.07
6
-.071
-
.26
7
1.00
0
-.069 .015
Deliver
yProce
ss
.208
.01
6
.161 .297
.02
3
.142 .373 .342 .425
.34
1
.10
4
.050
.18
7
-.107
.34
4
-.069 1.000 -.123
NoCre
ditCar
d
-.205
-
.15
1
-
.057
-
.190
-
.08
0
-.135 -.241 -.103 .104
-
.18
5
-
.18
7
-.119
-
.09
8
.118
-
.12
0
.015 -.123
1.00
0
Sig.
(1-
tail
ed)
Intern
etvsM
all
.29
4
.308 .000
.00
2
.086 .026 .025 .420
.49
5
.00
0
.035
.06
3
.129
.40
9
.037 .016 .017
LatestI
nfo
.294 .000 .014
.08
9
.046 .320 .100 .213
.00
0
.39
5
.181
.00
0
.316
.37
9
.401 .436 .061
Accesi
bility
.308
.00
0
.052
.30
3
.008 .125 .004 .008
.00
2
.25
9
.117
.00
7
.151
.10
3
.487 .050 .281
Conve
nience
.000
.01
4
.052
.00
0
.004 .031 .003 .087
.00
5
.00
0
.178
.17
0
.120
.02
6
.020 .001 .025
Saveti
me
.002
.08
9
.303 .000 .000 .160 .181 .093
.04
2
.00
0
.113
.03
2
.147
.03
8
.053 .407 .208
Anywh
ereAny
.086
.04
.008 .004
.00
.082 .041 .003
.00 .01
.376
.00
.190
.00
.015 .073 .084
Online Buying Behavior Page 35
time 6 0 0 5 0 0
Credit
CardSa
fe
.026
.32
0
.125 .031
.16
0
.082 .172 .001
.14
6
.37
5
.174
.36
8
.264
.00
0
.478 .000 .006
Specifi
cDateT
ime
.025
.10
0
.004 .003
.18
1
.041 .172 .124
.38
5
.04
1
.260
.31
6
.073
.08
4
.301 .000 .148
Guara
nteed
Qualit
y
.420
.21
3
.008 .087
.09
3
.003 .001 .124
.00
0
.15
9
.112
.00
1
.442
.00
0
.132 .000 .145
Discou
nts
.495
.00
0
.002 .005
.04
2
.000 .146 .385 .000
.07
2
.228
.00
0
.093
.00
0
.312 .000 .029
Hasslef
ree
.000
.39
5
.259 .000
.00
0
.015 .375 .041 .159
.07
2
.108
.06
4
.126
.25
6
.077 .144 .027
Casho
nDeliv
ery
.035
.18
1
.117 .178
.11
3
.376 .174 .260 .112
.22
8
.10
8
.45
1
.249
.38
3
.121 .305 .112
EasyFi
nd
.063
.00
0
.007 .170
.03
2
.000 .368 .316 .001
.00
0
.06
4
.451 .444
.00
2
.218 .027 .158
FacedP
roblem
s
.129
.31
6
.151 .120
.14
7
.190 .264 .073 .442
.09
3
.12
6
.249
.44
4
.00
1
.236 .139 .115
Contin
ue
.409
.37
9
.103 .026
.03
8
.000 .000 .084 .000
.00
0
.25
6
.383
.00
2
.001 .003 .000 .111
Toucha
ndFeel
.037
.40
1
.487 .020
.05
3
.015 .478 .301 .132
.31
2
.07
7
.121
.21
8
.236
.00
3
.242 .437
Deliver
yProce
ss
.016
.43
6
.050 .001
.40
7
.073 .000 .000 .000
.00
0
.14
4
.305
.02
7
.139
.00
0
.242 .105
NoCre
ditCar
d
.017
.06
1
.281 .025
.20
8
.084 .006 .148 .145
.02
9
.02
7
.112
.15
8
.115
.11
1
.437 .105
Online Buying Behavior Page 36
Annexure 1c: ANOVA Table for Regression Analysis
ANOVA(e)
Model
Sum of Squares df Mean Square F Sig.
1
Regression 35.202 5 7.040 4.396 .001(a)
Residual 129.717 81 1.601
Total 164.920 86
2
Regression 34.427 4 8.607 5.408 .001(b)
Residual 130.493 82 1.591
Total 164.920 86
3
Regression 33.711 3 11.237 7.108 .000(c)
Residual 131.209 83 1.581
Total 164.920 86
4
Regression 30.700 2 15.350 9.607 .000(d)
Residual 134.219 84 1.598
Total 164.920 86
a Predictors: (Constant), MaritalStatus, FreqofPurchase, Education, CreditCard, Age
b Predictors: (Constant), MaritalStatus, FreqofPurchase, Education, CreditCard
c Predictors: (Constant), MaritalStatus, FreqofPurchase, CreditCard
d Predictors: (Constant), FreqofPurchase, CreditCard
e Dependent Variable: AmtSpent
Online Buying Behavior Page 37
Annexure 2a: Questionnaire for Exploratory Research
Hi,
We are conducting a research on consumer behavior on the internet. As a part of our exploratory
research we request you to spare a few minutes in answering the questions below. Do feel free to
ask us for any clarifications or doubts.
Your responses would serve as a base to our research; and would guide us in discovering important
facts about the buying process.
Thanks.
1. What are the online services that you generally use while surfing the net (answer as many)?
2. Which are the websites that you visit frequently for the following purposes:
a. E-mail:
b. Chat:
c. Shopping:
d. Job Search:
e. Social Networking:
f. Banking and other Financial Services(Stock Trading):
g. Education:
h. General Browsing:
2. How frequently do you shop online?
3. If you do not shop online, then what are the reasons behind not shopping online?
4. What are the occasions when you buy online?
5. While purchasing online, what are the goods and services that you generally buy?
6. What are the payment methods you generally use for online purchases?
7. What are the different types of payment options that you have come across?
Online Buying Behavior Page 38
8. Do you feel it is safe to buy online?
9. What are the features that make a website more attractive than the others while buying online?
10. On an average, how much do you spend while buying online?
11. What is your general experience of buying online as compared to conventional shopping?
12. What additional features, do you feel, would enhance your buying experience on the internet?
Any additional insights:
Name: Age:
Profession: Sex: M/F
Years since you started using the Internet:
Do you own a credit card? Y/N
Do you use Internet Banking? Y/N
Annexure 2b: Final Questionnaire
QUESTIONNAIRE
We would be thankful for your cooperation if you spare a few minutes to answer the
following questions:
1. Do you like to purchase clothes via E-Shopping?
Yes No
Online Buying Behavior Page 39
2. On an average, how much time (per week) do you spend while surfing the Net?
a) 0-2 hours b) 2-6 hours
c) 6-10 hours d) 10-15 hours
e) Greater than 15 hours
3. I am more comfortable for purchasing clothes via
Offline clothes purchasing Online Clothes Purchasing
4. How much you spend money for purchasing clothes on monthly basis?
Yes No
5. I am/ would not comfortable buying clothes online because
a) have no trust b) Qaulity issue
c) fitting issue d) Fake
Any other product, please specify
6. If you already buying cloths online How frequently do you purchase cloths online?
a) Once a month b) 2-3 times a month
c) Once in 3 months d) Once in 6 months
Any other, please specify
7. What is the average amount that you spend per purchase while shopping online?
a) Rs 5000
8. Which of the following web sites do you shop at?
a)
Any other, please specify
9. Recall your earlier online buying/ shopping experience and please indicate you degree of
agreement with the following statements:
Strongly
Agree
Agree Neither
Agree nor
Disagree
Disagree Strongly
Disagree
I prefer making a purchase from
internet than using local malls or
stores
I can get the latest information
from the Internet regarding
different products/ services that is
not available in the market.
I have sufficient internet
accessibility to shop online.
Online shopping is more
Online Buying Behavior Page 40
convenient than in-store shopping.
Online shopping saves time over
in-store shopping.
Online shopping allows me to shop
anywhere and at anytime.
It is safe to use a credit card while
shopping on the Internet.
Online shopping provides me with
the opportunity to get the
products delivered on specific date
and time anywhere as required.
Products purchased through the
Internet are with guaranteed
quality.
Internet provides regular
discounts and promotional offers
to me.
Internet helps me avoid hassles of
shopping in stores.
Cash on Delivery is a better way to
pay while shopping on the
Internet.
Sometimes, I can find products
online which I may not find in-
stores.
I have faced problems while
shopping online.
I continue shopping online despite
facing problems on some
occasions.
It is important for me to touch and
feel certain products before I
purchase them. So I cannot buy
them online.
I trust the delivery process of the
shopping websites.
I do not shop online only because I
do not own a credit card.
Name: Gender: Occupation:
Online Buying Behavior Page 41
Monthly Income: Education: Marital Status:
doc_198233867.pdf