STATISTICAL ANALYSIS OF
SATISFACTION LEVEL OF CRAZY
JOHNNYS M-TONE CUSTOMERS
In partial fulfilment of covering the course of Business Statistics
Submitted To:
Mr. Praveen Kumar
Faculty, Business Statistics
FORE School of Management
Submitted By:
Parth Dhingra
Student, FORE School of Management
New Delhi
FMG-25A, Roll No. - 251036
ACKNOWLEDGEMENT
The report has been done as an end term project for the course titled A Comparative Study
of Customer Perception toward E-banking Services
I am thankful to our course instructor Mr. Praveen Kumar, FORE School of Management,
New Delhi for his invaluable guidance and assistance. I also wish to thank him for providing
us with this opportunity to learn more about the various Statistical tools.
Parth Dhingra
FMG-25A
Fore School of Management
New Delhi
Table of Contents
EXECUTIVE SUMMARY.................................................................................................................3
INTRODUCTION...............................................................................................................................4
ANALYSIS...........................................................................................................................................4
1.
2.
GENERAL FEATURES...............................................................................................................4
(a)
Customer Database.........................................................................................................4
(b)
Usage patterns.................................................................................................................5
CUSTOMER SATISFACTION.........................................................................................................6
Confidence interval for dissatisfied customers.........................................................................9
Confidence interval for satisfied customers..............................................................................9
3.
CONCLUSION..................................................................................................................................13
1.
2.
CUSTOMER SATISFACTION.......................................................................................................13
3.
Executive Summary
This report has been split into three stages of statistical analysis based on key issues raised by
Crazy Johnnys M-Tone.
The general features of the customer database and their usage patterns.
The satisfaction of customers, where satisfaction is measured according to company
From the analysis of the data sample, the following conclusions were made:
1. (a) Customer database
Database is approximately comprised of equal numbers of female and male customers.
Most of the customers are from the 15-19 and 20-24 age brackets.
(b) Customer usage patterns
An average customer approximately spends $32 and buys 13 ringtones a year.
The majority of total customer expenditure comes from the 15-19, 20-24 and 25-29
age brackets.
2. Customer satisfaction
Female customers are generally more satisfied (where satisfied refers to satisfied or
very satisfied) with the company service and content than male customers.
Generally, customer satisfaction decreases as the price per ringtone increases.
Generally, customer expenditure increases as customer satisfaction increases.
Although there was sufficient evidence to suggest that customers are buying more
than 10 ringtones a year, there was insufficient evidence to suggest that customers
spent more than $30 a year. However, there was no evidence to suggest that the
companys prediction, that customers will spend $30, was not satisfied.
Introduction
This report aims to analyse Crazy Johnny M-Tones customer database, their usage patterns,
whether they are satisfied with the company service and content and whether the company is
achieving its performance goals. An analysis of the data is based on 200 observations with
variables of gender, age, expenditure (amount spent in a year) and frequency (number of
ringtones bought in a year) considered.
Using inferential statistics (point and interval estimators and hypothesis testing), information
about population parameters will be extracted using the sample data. No data omissions will
be made, even though one customer is comparatively older than the other customers. It was
decided that the customer was representative of the mature age market of Crazy Johnny MTone, and thus not an outlier.
However, there is a limitation to this report. Although the sample data can give a valid
estimate of the population parameters because it is large and randomly selected, it is still only
an estimate and is not a definite analysis.
Analysis
1. General Features
(a) Customer Database
Statistics
Female
Male
Number
Percentage
Mean
A Median
g Mode
e Standard
deviation
105
52.50%
20.56
20
17
4.1
95
47.50%
20.76
20
17
6.04
Customer
database
200
100%
20.66
20
17
5.1
The database is comprised approximately equally of both female and male customers. In
terms of marketing, there is no target gender.
Age distribution of
females and males
60
10-14
50
Age distribution
15-19
20-24
25-29
40
Frequency (no.) 30
30-34
35-39
20
64-69
10
0
The age distributions of male and female customers are similar, as indicated by Figure 2, with
the modal class (21), mean (approximately 20) and median (17) the same for both males and
females. The histograms for both genders is positively skewed and unimodal. This implies
that most of the customers are young, more specifically in the 15-19 and 20-24 age brackets.
(b) Usage patterns
Statistic
Mean
Standard deviation
Expenditure
31.33
17.19
Ringtones
12.21
7.18
Figure 3 Average expenditure and average number of ringtones bought per customer
2% 2% 2%
13%
10-14 15-19 20-24 25-29 30-34 35-39
49%
33%
34%
65-69
55%
From Figures 4 and 5 it can be seen that the target market of Crazy Johnnys M-Tone for both
females and males should be 15-29. The proportion of total expenditure at these age groups
for females is 94% and for males 96%. Although there are customers from other age groups
(10-14 and 30+), they make a small proportion of the total sample size and thus only spend a
small proportion of the total expenditure.
2. Customer satisfaction
4 (very satisfied)
28%
(dissatisfied)
14%
3 (satisfied)
17%
61%
4 (very satisfied)
44%
3 (satisfied)
38%
From Figure 6, it can be seen that the majority of female customers (69%) show some form
of satisfaction with the company service and content, with 61% satisfied and 8% very
satisfied. Levels of dissatisfaction are not very high (31%) with only 17% of customers
dissatisfied and 14% very dissatisfied.
Figure 7 shows that compared to female customers, male customers are overall less satisfied.
Only just over half of male customers (51%) show some form of satisfaction with the
customer service and content, with 44% satisfied and 7% very satisfied. High levels of
dissatisfaction (49%) are exhibited with 38 % of customers dissatisfied and 11% very
dissatisfied.
Price paid per ringtone for each customer and customer satisfaction
In Figure 8, the correlation co-efficient between customer satisfaction and price the customers
paid for each ringtone is -0.57. This indicates a strong negative relationship between the two
variables, and suggests that customers exhibit lower satisfaction levels when they have to pay
higher prices.
In Figure 9, the correlation co-efficient between customer satisfaction and expenditure was
found to be 0.65, Therefore there is a strong positive correlation between the two variables.
That is, as customer satisfaction increases the amount customers spend on ringtones also
tends to increase.
Gender
Female
Male
Customer
database
Total
105
95
200
1
(very dissatisfied)
n
s
X
15
5.47
2.26
10
5.1
2.85
25
5.32
2.46
2
(dissatisfied)
n
18
36
54
29.33
25.94
27.07
s
20.22
15.26
16.95
3
(satisfied)
n
64
42
106
35.45
39.57
37.08
4
(very satisfied)
s
13.36
11.19
12.65
n
8
7
15
Figure 10 Descriptive statistics for customer expenditure at each level of customer satisfaction,
where n is the number of customers, x is the mean expenditure ($) and s is the standard deviation
($).
From Figure 8 and 10, it can be determined that there is notable variability in the amount of
spending at the 2nd and 3rd satisfaction level compared to the 1st and 4th level. To support
the relationship, that customer expenditure increases as customer satisfaction increases, a
90% confidence interval will be constructed to find a smaller range and it will be observed
whether the mean spending made by female or male customers lies within this confidence
interval. Confidence intervals do not need to be constructed for the 1st and 4th satisfaction
level as the standard deviation for those sample data is small.
51.38
47
49.33
s
3.34
5.8
5.01
54
27.07
16.95
1.645
X
N (0,1) . Therefore, an unknown
/n
X
z / 2 , where = 0.1.
/n
s
16.95
=27.07 1.645
= ( 23.27,30.87 )
90% CI X z /2
n
54
( )
Despite the variability in customer expenditure at the dissatisfied level, this confidence
interval infers that if 100 samples were drawn from the population, 90 of them would have a
mean spending made by customers within $23.27 and $30.87. It can be seen that the mean
spending made by both females and males when they are dissatisfied lies between this
confidence interval.
106
37.08
12.65
1.645
X
N (0,1) . Therefore, an unknown
/n
X
z / 2 , where = 0.1.
/n
s
12.65
=37.08 1.645
=( 35.06, 39.1 )
90% CI X z /2
n
106
( )
10
Even though the customer expenditure at a satisfied level shows high variability, the
confidence interval implies that, if many samples were drawn out, the mean spending made
by satisfied customers would lie within $35.06 and $39.10 with 90% probability.
Additionally, the mean spending made by females when they are satisfied lies between this
interval while the mean spending made by males lies above this interval, suggesting that
males spend relatively more compared to females.
Significance level
= 0.05
Z0.05 = 1.645
Alternative hypothesis
H1: p >10
Rejection region
Z > 1.645
sample
approximately
irrespective of population distribution.
11
mean
is
normal
(s)
n
12.21
7.18
200
Figure 11 - The standard normal showing the test statistic for KPI 1
Test statistic
Z=
X
12.2110
=
=4.353
/ n 7.18/ 200
4.353 > 1.645
12
significance level. Furthermore, the pvalue of 0 is less than 0.05 and this supports the
conclusion that there is sufficient evidence to reject the null hypothesis.
Hypothesis testing for KPI 2
Crazy Johnnys M-Tone predicted that customers would spend $30 in a year.
A one-tail hypothesis test (significance level = 0.05) can be established to determine whether
Null hypothesis
H0: p = 30
Alternative hypothesis
H1: p >30
Significance level
= 0.05
Z0.05 = 1.645
Rejection region
Z > 1.645
this
goal
achieved.
has
been
The
null
true unless there is sufficient evidence to reject it in favour of the alternate hypothesis, that
customers spend more than $30.
As the sample size is large (n = 200) CLT can be employed. That is, a sampling distribution
of a sample mean is approximately normal irrespective of population distribution.
31.33
17.19
200
13
Figure 12 - The standard normal showing the test statistic for KPI 2
Test statistic
Z=
X
31.3330
=
=1.094
/ n 17.19/ 200
1.094 1.645
Conclusion
The following conclusions were obtained from the analysis of the sample data provided.
2. Customer satisfaction
14
Female customers exhibit greater satisfaction with company service and content than male
customers with the majority of female customers (69%) just over half of male customers
(51%) showing some form of satisfaction.
The correlation co-efficient between price/per ringtone and customer satisfaction is negative
(-0.57), suggesting that as price per ringtone increases, customer satisfaction decreases.
The mean expenditure for customers is:
To identify the relationship between customer satisfaction and customer expenditure, the
correlation co-efficient was found to be largely positive (0.65), suggesting that as customer
satisfaction increases, customer expenditure increases.
15