Chapter 4- Data Analysis

4.1 Introduction
This chapter interprets the statistical results derived from the collected responses and
examines the proposed hypotheses.
4.1.1 Response Rate
Initially, in total 300 questionnaires were distributed among the desired respondents. 250
were returned by the respondents and 199 (66%) were found to be usable. According to
Babbie (1990) response rate over 60% is acceptable and is considered good in the field of
social sciences.
4.2 Demographic Profile of Respondents

Table 4.1: Age Group

Frequency

Percent

Valid Percent

Cumulative
Percent

Valid

18-30

168

84.4

84.4

84.4

31 and above

31

15.6

15.6

100.0

Total

199

100.0

100.0

Out of the 199 responses, 168 respondents (84%) were aged between 18-30 years.And 31
respondents (16%) were aged more than 31 years.
Table 4.2: Gender
Frequency

Percent

Valid Percent

Cumulative
Percent

Valid

Male

156

78.4

78.4

78.4

Female

43

21.6

21.6

100.0

Total

199

100.0

100.0

Out of the 199 responses, 156 respondents (78%) were male. And 43 respondents (22%)
were females.

35

Chapter 4- Data Analysis

Table 4.3: Occupation

Frequency

Percent

Valid Percent

Cumulative
Percent

Valid

Service

16

8.0

8.0

8.0

Student

183

92.0

92.0

100.0

Total

199

100.0

100.0

Out of the 199 responses, 16respondents (8%) were in private service. And 183
respondents (92%) were students.
Table 4.4: Income
Frequency

Percent

Valid Percent

Cumulative
Percent

Valid

Less than 3 lacs

192

96.5

96.5

96.5

3 lacs to 6 lacs

3.5

3.5

100.0

Total

199

100.0

100.0

Out of the 199 responses, 192 respondents (96.5%) were having income of less than 3
lacs. And 7 respondents (3.5%) were having income of more than 3 lacs per annum.

Table 4.5: Educational Qualification

Frequency

Percent

Valid Percent

Cumulative
Percent

Valid

UG

24

12.1

12.1

12.1

Grad

11

5.5

5.5

17.6

PG

164

82.4

82.4

100.0

Total

199

100.0

100.0

Out of the 199 responses, 24 respondents (12%) were under graduates. 11 respondents
(6%) were having degree of graduate. 164 respondents (82%) were post graduates.

4.3 Reliability Analysis

It is important to assess the reliability of instruments in order to design good scales for
measurement. Reliability can be defined as the extent to which measures are free from

36

Chapter 4- Data Analysis

error and thus yield consistent results (Zikmund, 1994). It is viewed that unwavering
reliability more than .06 is considered great. The data set used in the study has reliability
coefficient more than 0.6, thus indicating that the data collection is reliable. The data set
had reliability coefficient of 0.711. thus the reliability can be termed good (refer table
4.6).
Table 4.6: Reliability Statistics
Cronbach's Alpha

N of Items

.711

16

4.4 Factor Analysis

Initially, the factorability of the 18 items was examined. Several well-recognised criteria
for the factorability of a correlation were used. All the 16 items correlated at least .3 with
at least one other item, suggesting reasonable factorability. Secondly, the Kaiser-MeyerOlkin measure of sampling adequacy was .661, above the recommended value of .6, and
Bartletts test of sphericity was significant (df= (2, 120) = 590.692, p < .05). The
diagonals of the anti-image correlation matrix were all over .5, supporting the inclusion
of each item in the factor analysis. Finally, the communalities were all above .3 (see
Table 4.8), further confirming that each item shared some common variance with other
items. Given these overall indicators, factor analysis was conducted with all 16 items.
The data collected from the selected sample respondents are scrutinized before the
commencement of tabulation. The frequency distribution is used by the researcher to
arrive at percentage figures. So, the percentage analysis was undertaken at first. Then, the
statistical analysis section consists of testing of hypotheses so that the Annova Test, F
Test, T test and factor analysis were used as analytical tools. The SPSS software is
utilized for analyzing the data. factor analysis (CFA) was conducted to confirm the
relationships between observed variables and their underlying constructs to assess the
degree to which the data fit the proposed measurement.

4.4.1 Sample Adequacy

37

Chapter 4- Data Analysis

Table 4.7: KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy.

Bartlett's Test of Sphericity

.661

Approx. Chi-Square

590.692

df

120

Sig.

.000

4.4.2 Communalities

Table 4.8: Communalities

Initial
Landline is easy to use for old- 1.000

Extraction
.419

aged / illiterate people

Landline is the cheapest means 1.000

.485

of communication
Landline connection is easy to 1.000

.554

get
Landline

is

the

must

for 1.000

.560

Airtel provides all details to 1.000

.546

Broadband connection

customers
I

read/

see

the

Airtel 1.000

.658

Unnecessary advt./ messages 1.000

.691

advertisements carefully.

are not possible in the landline

Landline is not necessary in the 1.000

.685

cell phone era

Usage of landline is out of 1.000

.613

fashion

38

Chapter 4- Data Analysis

I continue with the landline 1.000

connection

due

to

.477

my

satisfaction towards call rates

I continue with the landline 1.000

.633

connection due to extended free

calls provided
I

continue

with

landline 1.000

.453

connection due to its rental

plan.
I am continuing with the 1.000

.522

landline connection as billing

system is transparent.
I am continuing with my 1.000
landline

connection

as

.498

the

quality of services offered are

best in class.
I continue with my landline 1.000

.584

connection as the grace period

for late bill payment is more
than of mobile operator
I continue with my landline 1.000

.598

connection I have sentiments

associated with it.
Extraction Method: Principal Component Analysis.

4.4.3 Variance Explained

Table 4.9: Total Variance Explained

Component Initial Eigenvalues

Extraction

Sums

of

Squared Rotation

Loadings
Total

of Cumulative Total

Variance

Sums

of

Squared

Loadings
%

of Cumulative Total

Variance

of Cumulative

Variance

3.282

20.513

20.513

3.282

20.513

20.513

2.114

13.210

13.210

1.804

11.275

31.788

1.804

11.275

31.788

1.989

12.434

25.644

1.544

9.652

41.440

1.544

9.652

41.440

1.912

11.952

37.596

1.234

7.714

49.155

1.234

7.714

49.155

1.642

10.264

47.860

1.112

6.949

56.104

1.112

6.949

56.104

1.319

8.244

56.104

39

Chapter 4- Data Analysis

Extraction Method: Principal Component Analysis.

4.4.4 Rotated Component Matrix

Table 4.10: Rotated Component Matrixa
Component
1
I continue with the landline .679
connection due to extended free
calls provided
I continue with the landline .664
connection
due
to
my
satisfaction towards call rates
I continue with landline .598
connection due to its rental
plan.
Landline connection is easy to .505
get
I am continuing with the .491
landline connection as billing
system is transparent.
I read/ see the Airtel
advertisements carefully.
I continue with my landline
connection I have sentiments
associated with it.
I continue with my landline
connection as the grace period
for late bill payment is more
than of mobile operator
I am continuing with my
landline connection as the
quality of services offered are
best in class.
Usage of landline is out of
fashion
Unnecessary advt./ messages
are not possible in the landline
Airtel provides all details to
customers
Landline is the must for
Broadband connection
Landline is easy to use for oldaged / illiterate people
Landline is not necessary in the
cell phone era
Landline is the cheapest means .422
of communication

.502

.756
.747

.576

.437

.518

.763
.758
.620
.419

.606
.572
.819
.534

40

Chapter 4- Data Analysis

Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
a. Rotation converged in 8 iterations.

4.4.5 Principle Component Analysis of Retention Factors among Landline Users

Principle Component Analysis (PCA) was applied on the responses filled in the
questionnaire on retention factors among landline users of Airtel ltd. PCA with Kaiser
Normalization and Varimax Rotation converged into five factors which were named in
the context of retention factors.

Table 4.11: Principle Component Analysis of Retention Factors among Landline Users
Factor
Percent Variance
Eigen Values
Item Converged
I continue with the
landline connection
due to extended free
calls provided
I continue with the
landline connection
due to my satisfaction
towards call rates
Factor satisfaction
I
continue
with
20.513
2.114
call rates
landline connection
due to its rental plan.
Landline connection
is easy to get
I am continuing with
the
landline
connection as billing
system is transparent.
I read/ see the Airtel
advertisements
carefully.
I continue with my
landline connection I
have
sentiments
associated with it.
I continue with my
landline connection as
the grace period for

Factor Loading
.679

.664

.598

.505
.491

.756

.747

.576

41

Chapter 4- Data Analysis

Factor 2

Grace period
and quality
services

Factor 3

11.275

9.652

1.989

1.912

Unnecessary
addvt.

Factor 4

Provide all
details to
customers

Factor 5

Cheapest means
of
communication

7.714

6.949

1.642

1.319

late bill payment is

more than of mobile
operator
I am continuing with
my
landline
connection as the
quality of services
offered are best in
class.
Usage of landline is
out of fashion
Unnecessary
advt./
messages are not
possible
in
the
landline
Airtel provides all
details to customers
Landline is the must
for
Broadband
connection
Landline is easy to
use for old-aged /
illiterate people
Landline
is
not
necessary in the cell
phone era
Landline
is
the
cheapest means of
communication

.518

.763
.758

.620
.606

.572

.819

.534

From the factor analysis 5 factors were identified. These five factors contributed for
56.104 % of variance.

4.5 Testing of Hypotheses

This section of the dissertation contains the testing of hypotheses. The opinion difference
between the respondents was identified and proposed hypotheses were examined using
statistical tests.
4.5.1 H1- There is difference between retention reasons of landline connection when
they belong to different educational qualification.

42

Chapter 4- Data Analysis

Table 4.12: ANOVA

Total
Sum of Squares
Between Groups

df

Mean Square

209.248

104.624

Within Groups

12606.752

197

63.994

Total

12816.000

199

F
1.635

Sig.
.198

It can be observed in table 4.12 that there was no significant effect of educational
qualification of land line user retention thought of landline at the p<.05 level for the three
conditions [F(2,197) = 1.635, p = 0.198].

4.5.2: H2- There is difference between retention reasons of landline connection when
they belong to different genders.

Table 4.13: ANOVA

Total
Sum of Squares
Between Groups

df

Mean Square

209.248

104.624

Within Groups

12606.752

197

63.994

Total

12816.000

199

F
1.235

Sig.
.198

It can be observed in table 4.13 that there was no significant effect of gender on land line
user retention thought of landline at the p<.05 level for the three conditions [F(2,197) =
1.235, p = 0.198].
4.5.3 H3- There is difference between retention reasons of landline connection when
they belong to different age group.

Table 4.14: ANOVA

Total

43

Chapter 4- Data Analysis

Sum of Squares
Between Groups

df

Mean Square

.043

.043

Within Groups

12815.957

198

64.727

Total

12816.000

199

Sig.
.001

.980

It can be observed in table 4.14 that there was no significant effect of age group land line
user retention thought of landline at the p<.05 level for the three conditions [F(2,197) =
0.001, p = 0.980].
4.5.4 H4- There is difference between retention reasons of landline connection when
they belong to different income

Table 4.15: ANOVA

Total
Sum of Squares
Between Groups

df

Mean Square

193.996

193.996

Within Groups

12622.004

198

63.747

Total

12816.000

199

Sig.

3.043

.083

It can be observed in table 4.15 that there was no significant effect of age group land line
user retention thought of landline at the p<.05 level for the three conditions [F(2,197)
=3.043, p = 0.083].
4.6 Status of Hypotheses
From the proposed hypotheses all were not supported by the statistical results. The
status has been explained in table 4.16.

Table 4.16: Status of hypotheses

H1: There is difference between retention reasons of

Not Supported

44

Chapter 4- Data Analysis

landline connection when they belong to different

educational qualification.
H2: There is difference between retention reasons of

Not Supported

landline connection when they belong to different

genders.
H3: There is difference between retention reasons of

Not Supported

landline connection when they belong to different

age group.
H4: There is difference between retention reasons of

Not Supported

landline connection when they belong to different

income

45