The Examining of GSM Operators

International Journal of Business and Management; Vol. 8, No. 3; 2013
ISSN 1833-3850 E-ISSN 1833-8119
Published by Canadian Center of Science and Education
1
The Examining of GSM Operators’ Customer Complaint Management
(CCM) Applications in Turkey with Discriminant Analysis
Yildiray Kizgin1
& Tahir Benli2
1
Mugla Vocational College, Mugla Sitki Kocman University, Mugla, Turkey
2
Department of Business Administration, Kastamonu University, Kastamonu, Turkey
Correspondence: Yildiray Kizgin, Mugla Vocational College, Mugla Sitki Kocman University, 48000, Mugla,
Turkey. Tel: 90-252-211-3193. E-mail: ykizgin@mu.edu.tr
Received: December 7, 2012 Accepted: December 28, 2012 Online Published: January 15, 2013
doi:10.5539/ijbm.v8n3p1 URL: http://dx.doi.org/10.5539/ijbm.v8n3p1
Abstract
In the active decision-making process, during the purchase, the most notable case is satisfaction or
dissatisfaction of the customer is caused in the purchasing decision. In the recent years, the new structures have
been composed under the name of “Customer Complaint Management” (CCM). In terms of emphasizing the
importance of research and in order to understand the effectiveness of CCM, researched that how the consumers
were evaluated the CCM performance of GSM companies in Turkey. The research was conducted on the
primarily-handed data. This study is a survey that aims to evaluate the “solving the consumer’s complaint for the
Turkcell, Vodafone, Avea companies”. And their performances related to CCM dimensions such as transparency,
accessibility, responsiveness, objectivity and consumer oriented strategies is perceived by the consumers.
Frequency distribution of survey data investigated and reliability of research variables tested by Cronbach Alpha
analysis. Two discrimination functions were evaluated. With the first discrimination function, it is seen that the
Avea has the highest value in comparison with the other companies (0,566); with the second discrimination
function, it is seen that Vodafone has the highest value (0,583). Turkcell has no highest discrimination value in
comparison with the other GSM companies.
Keywords: Customer Complaint Management (CCM), GSM firms, discriminant analysis
1. Introduction
Whatever its identity within the sector is one of the most significant problems threatening the companies in
today's market conditions is the risk of considerably reduction of the customer loyalty which was gradually
acquired in years. Intensively preferred marketing strategies of today are the adaptation of the rapid
technological developments to the productions in order to increase the customer loyalty, and differentiation and
diversification of goods and services. However, what direct the consumers to purchase are not always the needs,
but the perception of the stimulants of marketing and sales strategies surrounding him.
When the changes in marketing strategies are examined, it is noticed that consumers in the 90s recognized that
they had something to say and needed to make it accepted while the companies were trying to sell
goods/services through mass marketing strategies in the 70s and 80s. Accordingly, the companies realized that
they had to care about the consumers and understand them in order to exist in the market (Bozkurt, 2000), and
they directed to target markets. The integrity in the markets in the 2000s was replaced by more personal service,
well, and interest. Customers started to act as organized masses against the companies rather than act as single
individuals.
While the goods and services the consumers prefer are sometimes the ones they previously used, those may
sometimes be their preferences which they opt with new assessments and the influence of reference groups.
During the active decision-making process, when the consumer is at the edge of making the decision purchase a
good or a service, the most prominent state awaiting the consumer after purchasing is the satisfaction and
dissatisfaction of his decision to purchase.
When considering the fact that to keep current customers in the hands is at least four times less costly than
recruiting a new customer for a company; and that the purpose of modern marketing is to increase the number of
the satisfied customers, it will clearly be noticed how vital it is to study the post-purchase consumer behaviors

www.ccsenet.org/ijbm International Journal of Business and Management Vol. 8, No. 3; 2013
2
and dissatisfactions. In general, the companies should more properly monitor and assess the complaint concept
which can be expressed as the feedback notification related to dissatisfaction about purchasing or consumption
of a good or a service.
The complaint also offers companies the possibility to establish stronger ties with their customers (Martin, 1997).
Despite the fact that the formation of complaints is not such a desirable case, the companies should consider
these complaints as feedback providing them an opportunity to be able to notice the errors and shortcomings. It
should be believed that the complaints are a valuable source of information for business management and that
they provide a valuable opportunity to gather and assess customer information effectively. Consequently,
restructuring of the companies at organizational structure and business process level in recent years “in the name
of CCM” is a result of this requirement.
In this study were emphasized the concept of “Complaint” and the importance of the Customer Complaints
Management applications in improvement of the communication services, which has a significant share in
services sector, to offer better quality of services. A survey was conducted considering the assessment of the
CCM services by customers in order to determine the degree of the effectiveness of the CCM applications by the
GSM companies, such as Turkcell, Vodafone and Avea operating in Turkey. The objectives of this study are as
follows:
 To investigate the consumer behaviors about CCM in terms of services provided by GSM companies;
 To ensure that the consumers evaluate the services GSM companies provided in terms of the CCM;
 To determine the similarities and differences of the GSM companies, if any, in terms of CCM services.
The following statistical methods were used to analyze the data obtained from the questionnaires: 1. Frequency
distributions in order to examine the overall distribution of data; 2. Cronbach's alpha based on the principles of
CCM in order to test the reliability of research variables in the questionnaire; 3. The Chi-square independence
test, which is one of non-parametric statistical methods, in order to compare gender with the status of the
complainant, and the GSM company with matters of complaints. 4. t test for parametric methods in order to test
the difference of the research variables about CCM according to gender. 5. One-way ANOVA in order to reveal
differences, if any, in perception of the CCM applications of the GSM companies by the GSM users. 6.
Discriminant analysis method for multivariate parametric methods was applied for the discrimination of the
variables defined in terms of GSM companies.
2. Theoretical Background
2.1 Customer Complaints Management (CCM)
They are Gilly and Hansen (1985) who developed the complaint resolving strategies regarding CCM. Related to
the issue, Barlow and Moller (1998) also formed complaint matrices in order to detect the sources of customers’
complaints, and to investigate the correlation between the customers’ cultural and demographic characteristics
and their behavior of complaint.
In their study, Parasuraman and Berry (1997) proved that the customers, who complained about the company
and satisfied with their complaints, have a much higher tendency to maintain the same relations with the
company compare to those dissatisfied customers.
The purpose of CCM process is to produce the solutions related to the consumer complaints as soon as possible.
Considering the studies conducted on this issue, Gerson (1997) revealed the importance of CCM, and focused
on the problems encountered by employees, who are first addressed, when implementing the complaint policies
and strategies.
Kang et al. (2009) examined the issue within the framework of customer complaints and relationship between
customer satisfaction and loyalty and studied the mobile phone industry in China. In their study was indicated
that a direct complaint has a positive influence on customer satisfaction.
Considering the e-knowledge-based complaint management, Mjahed and Trike (2009) stated that they put
forward a pioneering study in this field. They identified the factors companies need to be successful in the
complaint management.
In spite of the fact that there are numerous researches done about the behavior of complain, in Turkey, it is
particularly seen most frequently in service sectors. In their performance assessment about CCM, Sarıdaldı and
Sevim (2009) conducted an application study ranging from the service sector to transportation. Evaluating the
performance of CCM from the customer perspective, they identified the reasons why it does not work
influentially, and they detected the factors reducing the performance of CCM within this process; and

3
consequently, rather than complaint management, they investigated the issue within the framework of
customers’ complaint tendency and the influence of customer satisfaction over the future purchasing tendency of
the customers.
Stauss (2002) in brief defined “complaint satisfaction or dissatisfaction refers to the customer’s assessment of
the company’s response to a complaint” (p.174). While the concept is defined as “secondary satisfaction” by
Etzel and Silverman (1981) and Oliver (1997); and as “complaint response satisfaction” by Lewis (1983); and as
“satisfaction with organization response complaint” by Blodgett and Granbois (1992), “service recovery
satisfaction” definition was first used by Boshoff (1999) and de Ruyter and Wetzels (2000). CCM mainly
“aiming to retain an existing customer rather than to gain a new one” (Odabaşı, 2001) is “a recovery strategy”
(McDougall & Levesque, 1999, p.7).
CCM and complaints in changing market conditions enable companies to understand the changing customer
needs (Barlow & Moller, 1998). A well established Complaint Management System is essential for successful
companies to manage the customers' needs, to protect their brand, and to achieve their brand value, and to
ensure customer loyalty (Doğan & Erdoğan, 2010). The companies will not only find solutions to the complaints
with the help this system they establish, but also they will reach the problem factors, make the necessary
adjustments, and will have the opportunity to practice the idea of customer-focused continuous improvement
(Naylor, 2003; Odabaşı, 2000).
On the other hand, as the customers are aware how powerful they are in today's competitive environment, they
now request more than the price paid for goods and services. They prefer the companies which generate value
for them, that is, which listen to their requests and complaints (Çoban, 2005). In one, survey, it was stated that
the complaining and dissatisfied customers were more likely to come again to the company compare to those
uncomplaining and dissatisfied customers (Barlow & Moller, 1998). According to another survey, there were
strong links between the satisfaction after post-resolved complaints and re-purchase while there was no link
between the post-purchase satisfaction and re-purchase (Barış, 2006). Identifying the reasons of dissatisfactions
of the dissatisfied customers and making the necessary adjustments, or ignoring the dissatisfactions of the
customers instead are the behaviors which will affect the success of the companies.
Companies can turn the complaints of the customers into an opportunity only through CCM. The main
opportunities are as following;
 The complaint just like advertising and sales increasing efforts can be used as a marketing variable creating
a feed back (Barış, 2006).
 Holding the light on the factors that reduce the quality of service, the complaint shows what to be done to
improve the quality (Bell et al., 2004).
 The complaining customer ensures that the company realizes the operational deficiencies related to the
goods or services (Odabaşı, 2000).
 The complaint sends signals to the customers to ensure that they can continue to come to the company.
According Kelleher (1999); evaluation of the customer complaints is extremely significant for three reasons in
terms of company: 1. It illuminates the management about the employee performance; 2. It draws attention to
areas in the company for improvement; 3. Complaints are the best indicators of performance in a company.
Within the scope of a study conducted, Stauss and Seidel (1998) suggested nine satisfaction forms related to
"complaint satisfaction". And these are adequacy of the problem solution; access: ease of finding a competent, a
contact person; friendliness: politeness, courtesy communication style; empathy: willingness to take the
customer's perspective, understanding the customer's annoyance, individual complaint handling; individual
handling: non-standardized response that is customized to the problem and the wishes of the complainant; effort:
visible effort to solve the customer's problem; active feedback: activity to find out the best solution for the
customer; notification about delays, feedback about procedures and decisions; reliability: keeping of promises;
speed of response: speed of reaction to the complaint, speed at which complaints are resolved.
According to ISO10002: 2004, while handling the customer complaints, the main principles are as follows:
transparency; accessibility; responsiveness; objectivity; wages; confidentiality; customer-oriented approach;
accountability and continuous improvement. When considered that the concept of satisfaction is a judgment of
the customer, it is possible to state that the use of customer-based methods is much more acceptable in
evaluating the performance of CCM (Sarıdaldı & Sevim, 2009).

4
2.2 The Significance of the CCM for GSM Companies in Turkey
Telephone, which emerged in 19th century, but has still been widely used due to its rapid changes, has become
one of the most indispensable tools of the communication sector. In parallel with the rapid development of
technology, the communication sector made up pace with this development. Moreover, the rapid development in
especially the mobile communication sector and the GSM (Global System for Mobile Communications)
operators sectors have led to the growth of the companies operating in the sector and have caused companies to
make various innovation.
Beside the growth in the GSM sector in Turkey, the competition between companies has started to increase
rapidly. As a result of the competition in the Turkish GSM sector, the target market reached a significant
numbers of GSM users/subscribers. According to the data by Information and Communication Technologies
Authority (BTK, 2012) compiled by the report of the last quarter of the year 2012, the number of the mobile
phone subscribers in Turkey increased to 67.17 million. Consequently, Turkcell has 35.15 million subscribers
while Vodafone has 18.81 million, and Avea has 13,21 million subscribers.
The competition has been accelerated with growth aimed infrastructure works diversifying the target market
carried out by Avea especially over the last two years; on the other hand, with the diversity of much commercial
new service pack released by Vodafone; and marketing strategies by Turkcell to increase the numbers of current
users. The consumer behaviors have been strongly influenced by this process (Kızgın, 2008).
Studies on customer satisfaction; customer complaints management practices; And works to improve the quality
of service, the customers have had difficulty to decide which GSM company to subscribe, due to the expansion
of the good/service ranges offered in the GSM sector; Especially wealth of options and prices offered by the
operators; . This sometimes makes the mobile phone subscriber change his GSM operator by moving his phone
number from one operator to another constantly or sometimes forces the customer to be a subscriber of more
than one operator simultaneously.
In such a competition, it is possible to determine which goods and services of the GSM operators are preferred.
It is possible for GSM operator to develop them further, and to be more successful in the sector and to increase
their market share by converting these preferences into goods. Therefore, it is essential for the GSM operators to
conduct researches considering the reasons why the consumers prefer which goods and services (Kızgın, 2008).
In today's world constantly changing with its values, a company is required to deliver error-free service for the
consumer to be successful (Gerson, 1997). It is essential in relations with the customer to listen to them; to
speak the same language with customers; to understand their demands and needs; to provide the necessary
flexibility in order to reach an agreement; to measure the results and to make the necessary arrangements so as
to achieve zero defect level; in other words, to achieve the total customer satisfaction (Ertürk & Kıyak, 2011).
When considered the fact “complaints have a direct impact on the customer satisfaction” (Kang et al., 2009,
p.24), as a requirement to compete and maintain the existence, the concept of “customer satisfaction” should be
considered necessary; and with the thought that monitoring the dissatisfactions as well as satisfaction is equally
a significant means of ensuring customer satisfaction, the GSM companies should focus on the concept of the
complaint and should improve the processes in which the complaints are dealt with.
Only some of the objectives of this study are to supply literature scanning on gathering information on the
complaints about the GSM goods and the services, investigating the complaint behavior of the GSM customers,
and detecting common or/and different complaint behaviors of these companies.
3. Methodology
3.1 Hypotheses
It is known that in terms of consumer behaviors and of it is impact on purchasing decision, the gender’s reaction
on purchasing behavior, its level of satisfaction, its post-purchase expectations are rather different respectively.
This approach is also a pre-indication of the fact that as consumers the individuals with the different gender state
different behaviors regarding the goods and services they purchase to feedback their dissatisfaction. Özdemir
and Tokol (2008) indicate that the references and the loyalty the female consumers provided for the companies
are influential on sales and the decisions of those companies, due to their biological and sensory differences and
their differences during the process of purchasing; and that one of the most decisive factors while establishing
the marketing strategies is to know the properties of female nature. The hypotheses can be established as
following, taking into account this information.
H1: Consumers' persistence to complain due to their dissatisfaction about goods and services of the GSM
companies varies according to gender.

5
H2: Perception of CCM performances of the GSM companies varies according to gender.
It is observed that the GSM companies often offer new goods and services to the interests of their consumers in
order to increase their market share, and that they also apply a large number of strategies and tactics to promote
them due to intense competition environment that both threatens and ensures today's companies improve. Even
though, they offer different goods and services, and the goods and services provided should be similar that the
technologies they applied in the CCM, and the resolution processes are analogous indicates that the complaints
received from the customers are also similar. However, these companies are expected to be different in terms of
the complaints about their goods and services owing to the different establishment ages, the stable vision and
mission, the number of customers, coverage, adequacy, quality understanding of these companies,. Here are the
hypotheses established for that purpose:
H3: The goods and services of the GSM companies complained are different from the other GSM companies.
H4: The complaint management performance of the GSM companies is different from the other GSM
companies.
3.2 Samples and Measures
Research method consists of the first-hand data, and is based on a survey developed to assess the complaints
after the dissatisfaction of GSM subscribers regarding goods and services provided by the GSM companies.
The survey mainly consists of three parts. In the first part, the aim is to compile the demographic and
socio-economic conditions of GSM subscribers while it is aimed to gather the complaining experiences of the
GSM users. In the second part, the third and last part of the questionnaire aims to measure how the GSM users
perceive the CCM principles such as transparency, accessibility, responsiveness, objectivity and customer
orientation considered within the framework of ISO10002: 2004. Considering the fact that the information
related to implementations of GSM company regarding accountability, privacy, continuous improvement, and
wages can only be obtained from the GSM companies; and that there are bureaucratic barriers to obtain this
information; and that customers should be the only inspectors of concept of complaints satisfaction, these
principles were not taken into consideration.
The survey was administered via the Internet, and invitation links to participate the survey were sent to the
participants’ e-mail addresses at random. The sampling Formula size of the questionnaires included 384 persons,
with Populations with an unknown status (Özdamar, 2003), with p=0.5, q=0.5, d=0.05 and 5% margin of error.
The surveys were expected to return for nearly 3 months, and answers were eliminated of the participants, who
deliberately did not answer an critical survey question, and/or left a large portion of the survey questions empty,
and/or sent more than one questionnaire. A total of 432 questionnaires were assessed.
In terms of CCM principles considered within the scope of ISO 1002: 2004, thirteen (13) propositional
questions were directed in order to determine to what extend the GSM users perceive the services provided by
the GSM companies. GSM users were directed four (4) propositional questions to determine how their
“complaints” behavior would be in case they encountered customer dissatisfaction as GSM subscribers. A
5-point Likert-type scale was used as the scale of the survey as follows: 1. I totally disagree; 2. I disagree; 3. I
less agree; 4. I agree, and 5. I totally agree. However, since two questions related to general perception of the
complaints were thought not to be understood well, agreed upon the value of Cronbach's alpha, they were taken
off the study.
4. Findings
4.1 Participants’Findings
Table 1 indicates the demographic and social status of the participants. Accordingly, 47,2 (%) of the participants
were worker-civil servants; 43.1 (%) were college graduates; 56.9 (%) were females while the income level of
27.8 (%) was in the range between 1500 and 3000 TL, and 45.8 (%) were between the ages of 25 and 35.
Table 1. Indicates the demographic and social status of the participants
Education Frequency Percent(%) Age Frequency Percent(%)
Primary-intermediate school 0 0 18 – 24 120 27.8
High school 126 29.2 25 – 35 198 45.8
Vocational College 126 29.2 36 – 45 78 18.1

6
Bachelor 186 43.1 46 – 59 36 8.3
Graduate(MA, Doctoral) 102 23.6 > 60 0 0
Total 432 100 Total 432 100
Occupation Income
Student 96 22.2 < 630
TL
102 23.6
Self-employed-Tradesman 96 22.2 631 -
1500
TL
114 26.4
Worker - Civil Servants 204 47.2 1501 -
3000
TL
120 27.8
Retired 6 1.4 3001 -
5000
TL
72 16.7
Housewife 18 4.2 > 5001
TL
24 5.6
No reponse 12 2.8 Total 432 100
Total 432 100
Gender
Female 246 56.9
Male 186 43.1
Total 432 100
Table 2 shows information about the complaint experiences by the GSM users about the GSM Products and
Services. It was observed that Avea Users complained their company with the highest percentage of 47.2 as of
the complaints directed within last six months; that the participant, with a percentage of 69.4, opted for the
option “I have complained and sought for my right” which indicates the persistence”; and that 30.6% of the
complaints were “tariff-related issues”.
Table 2. The findings of complaints
Complaints directed to the
GSM companies within last
six months
Frequency Percent(%)
Issues of the
complaints
directed
Frequency Percent(%)
Avea 204 47. 2 Problems related
to tariffs
132 30. 6
Turkcell 126 29. 2 Problems related
to bill
84 19. 4
Vodafone 102 23. 6 Problems related
to coverage
66 15. 3
Total 432 100. 0 Slow processing
of the call center
18 4. 2
The situation of make Disposal of 24 5. 6

7
complainant unnecessary short
messages (SMS)
I have complained and
sought for my right
300 69. 4 Other 24 5. 6
I have complained and be
displeased at, at last
changed my GSM company
126 29. 2 Unanswered 84 19. 4
Unanswered 6 1. 4 Total 432 100. 0
Total 432 100. 0
4.2 Non-Parametric Test Findings
The complaints of the GSM users concerning their dissatisfaction about the purchased products and services
were given by gender (for H1 hypothesis testing) in table 3. It was understood that the complaints of participants
or their persistence in complaining within last 6 months did not change in terms of gender factor (P=
0.146>0.05); and that H1 hypothesis was not supported on this issue.
Table 3. The findings the complaints of participants by gender
Gender
Case of the complaints directed within last six months
Total
Pearson
Chi-square
Df. P-value
I have complained and
sought for my right
I have complained and be
displeased at, at last changed my
GSM company
Female 180 66 246 2,111 1 0,146
Male 120 60 180
Total 300 126 426
For H3 hypothesis testing, in table 4 were given the findings compiled in order to search for the correlation
between the subjects of the complaints directed to the GSM companies. Accordingly, it was seen that when the
complaint issues were compared with the complained GSM companies, the Chi-square test was significant (P=
0,000 (<) 0,001); and that H3 hypothesis was supported on this issue,
Table 4. The findings compiled in order to search for the correlation between the subjects of the complaints
directed to the GSM companies
Complaint Subjects
Complaints directed to the GSM companies Pearson
Chi-square
Df.
P-valu
eTurkcell Vodafone Avea Total
Problems related to tariffs 60 42 30 132
61.728 10 0.000
Problems related to bill 24 24 36 84
Problems related to coverage 48 12 6 66
Slow processing of the call
center
12 6 0 18
Disposal of unnecessary
short messages (SMS)
6 6 12 24
Other 12 12 0 24
Total 162 102 84 348

8
The reliability values of the questions asked in accordance with the principles of CCM were given in table 5.
Cronbach’s alpha value was noticed above 0.7 which is considered adequate for a social science research.
Additionally, the sampling adequacy can be considered good as the Kaiser-Meyer-Olkin Test for sampling
adequacy was 0.838 and P value was 0.000<0.05.
Table 5. The findings of the research variables
CCM Variables Frequency Mean Df
CCM
Principles
Mean
Cronbach
Alpha
To return to the customer regarding
the complaints
420 2.857 1.087 Responsiveness 3.087 0.732
To follow up the complaint records
with SMS, telephone and e-mail
432 2.889 1.186
To analyze the complaint by
applying to an authorized personnel
at the first stage
426 2.775 1.117
To return to the customer as soon as
possible about the complaints
432 3.819 1.086
To reach an authorized personnel
about the complaints as soon as
possible
432 2.458 1.054 Accessibility 2,944 0.713
The ease to access to the complaints
system
432 2.653 1.134
To know where and how to make the
complaint
432 3.722 1.058
To have a clear and well defined
process in order to phase in the
complaint
420 3.014 1.166 Visibility 3.030 0.822
To indicate the way/ways to submit a
complaint
414 3.072 1.109
the know-how adequacy of the
authorized personnel who was
reported the complaint
426 3.028 1.035 Objectivity 2.920 0.786
The accuracy of official knowledge
of the authorized personnel who is to
evaluate the complaint
420 2.800 1.024
Satisfaction with the efforts provided
during the complaint resolution
process
420 2.686 1.037 Customer-focused
approach
3.193 0.708
Positive aspects of the established
and well-performing complaint
process in terms of service quality
432 3.708 1.149
4.3 Parametric Test Findings
The Skewness and Kurtosis values can be applied for questioning the validity of the assumption of normal

9
distribution, which is one of the significant assumptions, in order to apply t Test, ANOVA and Discriminant
Analysis of Parametric Methods. These values are required to be in between -2<X<+2 (Şencan, 2002). In the
case of a violation, in this matter, according to the central limit theory, the data are assumed to meet the normal
distribution condition as the number of the samples is over 40 (Stat trek, 2012; Statsoft, 2012; Lumley et al.,
2002). In the study, it was seen that there was no variable violating the normal distribution assumption.
While seen that the perceptional differences about the CCM variables given in table 6 had a correlation with the
gender, three other variables did not have any perceptional differences related to gender.
Table 6. The findings to be compared CCM variables by gender
◊ Variances is not homogeneous; *p<0.05 **p<0.001.
Note: P-values of the other three complaint-variables were not stated in table 6 as they were not significant.
H2 hypothesis formulated in correlation with this is acceptable. When the variables were grouped in accordance
with the CCM principles, significant differences related to gender were found out in the complaint perceptions
in all the variables constituting the principles of accessibility, customer-orientedness, and objectivity.
Within the scope of one-way ANOVA test conducted, related to the CCM variables involved in the research,
CCM variables
Female
Mean
(df)
Male
Mean
(df)
t-value
To have a clear and well defined
process in order to phase in the
complaint
3.20
(1.032)
2.90
(1.188)
2.771*◊
To reach an authorized personnel
about the complaints as soon as
possible
2.61
(1.011)
2.26
(1.080)
3.476*
the know-how adequacy of the
authorized personnel who was
reported the complaint
3.25
(0.970)
2.74
(1.049)
5.173**
The accuracy of official knowledge
of the authorized personnel who is to
evaluate the complaint
2.97
(1.002)
2.58
(1.012)
3.979**◊
To return to the customer as soon as
possible about the complaints
3,93
(1.070)
3,68
(1.092)
2.377*
To follow up the complaint records
with SMS, telephone and e-mail
3.07
(1.137)
2.65
(1.209)
3.769**
The ease to access to the complaints
system
2.78
(1.073)
2.48
(1.191)
3.737**
Satisfaction with the efforts provided
during the complaint resolution
process
2.85
(0.939)
2.47
(1.121)
3.715**◊
To know where and how to make the
complaint
3.54
(1.064)
3.97
(1.002)
-4.276**◊
Positive aspects of the established
and well-performing complaint
process in terms of service quality
3.56
(1.063)
3.90
(1.231)
-3.095*

10
while no significant difference was reached between the mean scores of variables of “Reaching an authorized
person concerned with the complaint as soon as possible” and “Satisfaction with the efforts supplied during the
complaint resolution process”, the mean scores of all other research variables, considering three GSM
companies, had a significant difference.
H4 Hypothesis formulated for this issue was accepted. A multiple comparison analysis was not applied on the
results of ANOVA (Post-Hoc) as Discriminant Analysis was applied in order to determine the distinguishing
hallmarks of the 11 variables, which have significant differences, according to the GSM companies.
Table 7. The findings to be compared CCM variables by complaints directed to the GSM companies
CCM Variables
Complaint Directed to the GSM Operators
Levene ANOVA
Statistic
P
value
F
value
P
value
To return to the customer regarding the complaints 8,159 0,000◊
9,689 0,000**
To follow up the complaint records with SMS, telephone and
e-mail
6,940 0,001◊
4,531 0,011*
To analyze the complaint by applying to an authorized
personnel at the first stage
8,420 0,000◊
9,394 0,000**
To return to the customer as soon as possible about the
complaints
18,104 0,000◊
13,895 0,000**
To reach an authorized personnel about the complaints as
soon as possible
2,328 0,099 0,071 0,932
The ease to access to the complaints system 8,319 0,000◊
3,668 0,026*
To know where and how to make the complaint 32,937 0,000◊
6,389 0,002*
To have a clear and well defined process in order to phase in
the complaint
8,736 0,000◊
14,615 0,000**
To indicate the way/ways to submit a complaint 10,633 0,000◊
17,012 0,000**
The know-how adequacy of the authorized personnel who
was reported the complaint
8,887 0,000◊
8,166 0,000**
The accuracy of official knowledge of the authorized
personnel who is to evaluate the complaint
5,708 0,004◊
4,599 0,011**
Satisfaction with the efforts provided during the complaint
resolution process
7,225 0,001◊
1,199 0,302
Positive aspects of the established and well-performing
complaint process in terms of service quality
16,282 0,000◊
8,962 0,000**
◊ Variances is not homogeneous; *p<0.05 **p<0.001.
4.4 Discriminant Analysis Findings
Assumptions of Discriminant Analysis, Normal distribution hypotheses;
 Normal distribution condition; Normal distribution condition suitability required in order to apply both t
test and ANOVA was tested and discriminant analysis findings and there is no such indication violating this
condition.
 Another assumption required for application of linear discriminant Analysis is that group covariance
matrices are equal to each other. Quadratic Discriminant Analysis must be applied when this condition is not
achieved (Özdamar, 1999). Multiple quadratic Discriminant Analysis is possible when there are at least three

11
clusters, in other words to say when there are three clusters and two separation functions (Nakip, 2003).
 Another major assumption of the discriminant analysis is the lack of multicollinearity link between the
variables. It is concluded that there is no multicollinearity, due to the lack of a correlation greater than 0.70 in
the matrix given in table 9 (Akgül & Çevik, 2003).
 A multiple Discriminant Analysis in terms of complained GSM companies was applied to discriminate the
11 significant variables resulted from ANOVA analysis of CCM variables. Box's M testing the equality of
covariance matrices was 330.177 and the probability value was P (0.158)>0.05, the covariance matrices was
concluded to be equal at a 5% error level. Accordingly, the linear discriminant Analysis can be used.
 There should not exist an outlier value in data. In order to determine and eliminate the outlier values during
the Discriminant Analysis, critical Mahalanobis D2 value was at 16.81 with 99% reliability while s.d. was 6. As
there did not exist any data which has D2 greater than this critical value during the discrimination analysis, there
was no need to eliminate this value.
 There must be 30 or 4-5 times the amount of data for each independent variable: while average 330 data
are sufficient for 11 variables included in the survey, 432 survey data meets this criterion.
Discriminant Analysis
In this study, there were a total of 13 variables considered as CCM variables. However, under the terms of the
ANOVA test, two variables with insignificant difference in terms of the GSM companies were taken out of the
Discriminant Analysis. In total, 11 variables were considered during the Discriminant Analysis. Stepwise
Discriminant Analysis was used in order to best estimate which of these variables shows the characteristics of
the complained GSM companies.
Stepwise Discriminant Analysis must be applied when the number of the variables exceeds the required number
during the discriminant analysis, or when it is required to separate the wanted variables within the set of data
(Özdamar, 1999).
In table 8 were given the Discriminant Analysis results obtained from three complained GSM companies.
According to these data, two discriminant functions were obtained. As the probability value of these two
discriminant functions is (0.00001<0.05), It was seen that discriminant function is a statistically significant
function; and that it can be applied in discrimination process.
Table 8. Canonical discriminants functions
Functions ()
Variance
(%)
Canonical
Correlation
Wilks'
Lambda()
Chi-square Df.
P
value
1 0,234 61,3 0,435 0,706 127,357 12 0,000*
2 0,147 38,7 0,358 0,872 50,376 5 0,000*
* p<0.001.
While testing the significance of the discriminant functions and coefficients used in the Discriminant Analysis,
Wilk's lambda value is used. Wilk's lambda value is a value between zero and one. The smaller this value is, the
greater the distinguishing power of the model will be (Akgül & Çevik, 2003). In addition, the variance indicated
in the first function was 61.3%, and the variance of the second function was 38.7%.
Table 9. Correlation matrix of discriminant analysis variables
CCM variables 1 2 3 4 5 6 7 8 9 10 11
To have a clear
and well
defined process
in order to
phase in the
1.00 0.631 0.527 0.534 0.497 0.237 0.469 0.440 0.516 0.201 0.285

12
complaint
To indicate the
way/ways to
submit a
complaint
0.631 1.00 0.617 0.528 0.511 0.265 0.611 0.635 0.696 0.158 0.421
the know-how
adequacy of the
authorized
personnel who
was reported
the complaint
0.527 0.617 1.00 0.661 0.538 0.382 0.653 0.646 0.581 -0.034 0.328
The accuracy of
official
knowledge of
the authorized
personnel who
is to evaluate
the complaint
0.534 0.528 0.661 1.00 0.440 0.360 0.499 0.485 0.431 -0.004 0.389
To return to the
customer
regarding the
complaints
0.497 0.511 0.538 0.440 1.00 0.414 0.516 0.672 0.652 0.156 0.424
To return to the
customer as
soon as possible
about the
complaints
0.237 0.265 0.382 0.360 0.414 1.00 0.292 0.282 0.289 0.023 0.480
To follow up
the complaint
records with
SMS. telephone
and e-mail
0.469 0.611 0.653 0.499 0.516 0.292 1.00 0.515 0.538 0.060 0.339
The ease to
access to the
complaints
system
0.440 0.635 0.646 0.485 0.672 0.282 0.515 1.00 0.688 0.162 0.355
To analyze the
complaint by
applying to an
authorized
personnel at the
first stage
0.516 0.696 0.581 0.431 0.652 0.289 0.538 0.688 1.00 0.164 0.405
To know where 0.201 0.158 -0.034 -0.004 0.156 0.023 0.060 0.162 0.164 1.00 0.121

13
and how to
make the
complaint
Positive aspects
of the
established and
well-performing
complaint
process in terms
of service
quality
0.285 0.421 0.328 0.389 0.424 0.480 0.339 0.355 0.405 0.121 1.00
While constituting the discriminant model. A variable should have an equal or greater value than F-input value
to participate in the model; a variable to be taken out of the model must have an equal value than F-output
value(Özdamar. 1999). In this context. while constituting the stepwise discriminant model. The F-input value
was 0.05. and F-output value was 0.10 which was twice the value when SPSS preliminary value was applied.
The number of the variables involved in the model dropped to six. Afterwards. after removing the variables
obtained from the system. the Discriminant Analysis was applied once again to see that the result did not
change.
Discriminant loads indicate the contribution of independent variables to the model (Churchill. 1999). The loads
of the variables thought to be a significant contribution to the model. both negative and positive. must be 0.3 or
above (Whitelock & Jobber. 2004). It was observed that this rule was not violated but one variable (-0.263). As a
result of implementing the second phase of Discriminant Analysis. It was decided to keep the variable in this
model. The structure matrix of these variables obtained was given in table 10.
Table 10. Structure matrix
CCM Variables
Discriminant Functions
1 2
To return to the customer as soon as possible about the complaints 0.422(*) 0.248
To follow up the complaint records with SMS. telephone and e-mail 0.334(*) 0.123
To know where and how to make the complaint -0.263(*) -0.206
To indicate the way/ways to submit a complaint 0.063 0.784(*)
To have a clear and well defined process in order to phase in the complaint -0.190 0.641(*)
To analyze the complaint by applying to an authorized personnel at the first
stage
-0.402 0.422(*)
When the table 11 of group averages was analyzed. concerning the complaints directed. it was observed that
Avea had a distinguishing difference (0.566) compare to other GSM companies; and that Vodafone had the
highest value (0.583) in the second discriminant function. It can be stated that in two respective discriminant
functions. Turkcell did not have any group average superior to the other GSM companies.

14
Table 11. Functions at group centroids
Discriminant Function
1 2
Avea 0.566 0.014
Turkcell -0.392 -0.434
Vodafone -0.432 0.583
Classification coefficients are calculated separately for each group and the classification process of the
observation. According to the independent variables, it will be determined by the score the observation will
obtain. The coefficients obtained for this purpose were given as the classification function coefficients in table
12.
Table 12. Standardized canonical discriminant function coefficients
CCM Variables
Complaint Directed to the GSM
Operators
Avea Turkcell Vodafone
To clear and defined process related to phasing in the complaint -0.072 0.219 0.478
To return to the customer as soon as possible about the complaints 2.907 2.421 2.514
To follow up the complaint records with SMS. telephone and e-mail 0.864 0.630 0.088
To analyze the complaint by applying to an authorized personal at the
first stage
-0.588 0.381 0.323
To know where and how to make the complaint 3.383 3.689 3.313
To indicate the way/ways to submit a complaint 1.245 0.213 1.179
Constant -15.366 -14.613 -15.974
Fisher's linear discriminant functions
Accurate estimate. carried out through Discriminant Analysis in accordance with the income groups. is expected
to be 25% more than the Proportional Chance Criterion (PCC) (Li et al.. 2007). Proportional Chance Criterion
by group numbers; the sum of possibilities for priority of the groups was calculated as
(0.4192+0.3392+0.2422=0.349). Accordingly, as the 0.349*(1.25)=0.433 value is less than the correct
classification rate of 0.57. The classification is considered to be correct.
Tablo 13. The probabilities of groups
Complaint Directed to the GSM Operators Prior
F value
Unweighted Weighted
Avea 0.419 156 156.000
Turkcell 0.339 126 126.000
Vodafone 0.242 90 90.000
Total 1.000 372 372.000

15
Table 14. The classify results
Group member of be estimated
Avea Turkcell Vodafone Total
Estimate F. Avea 120 30 24 174
Turkcell 60 54 12 126
Vodafone 36 24 36 96
% Avea 69.0 17.2 13.8 100.0
Turkcell 47.6 42.9 9.5 100.0
Vodafone 37.5 25.0 37.5 100.0
a) 53.0% of original grouped cases correctly classified.
b) 53.0% of cross-validated grouped cases correctly classified.
In addition. the Press's Q statistic can be used to test the power of the discrimination function(Hair. et al.; Chan.
2005). As below:
)1(/)]([ 2
 kNnKNQ (1)
Where Formula: N=432. n=correct classification number (0.53*432) =229; and k (the number of groups) =3. As
Press’s Q=75.2>9.21 (chi-square table 0.99.2) is greater than Press’s Formula of Q. the classification can be
stated to have a higher statistical significance than the Chance Criterion.
5. Discussion
By the findings compiled. in this study were analyzed the CCM implementations at GSM service. which is the
most powerful and the most common representative, and which is one of the most crucial areas of
communication service sectors in particularly Turkey. In this study was tried to explain how the customers have
perceived the processes of CCM that the GSM companies applied to their customers.
To compare other GSM companies, it is possible to indicate that related to the complaints directed to Avea. The
customer perception has a higher level of satisfaction. positive in terms of the variables of “return to the
customer regarding the complaints as soon as possible". and of "follow up the complaint records with SMS.
telephone and e-mail". But it can be indicated that the customer perception has a higher level of dissatisfaction.
negative in terms of the variable “not to know where and how to make the complaint”.
The distinctive function is the second function in terms of the variables: “indication of the way/ways to submit a
complaint”. “To have a clear and well defined process in order to phase in the complaints” and “resolution of the
complaint by applying to authorized personnel at the first stage”. Vodafone has a distinguishing superiority in
this function. Accordingly, it can be indicated that the company performance. which the customers perceived
about this company in terms of the complaints stated in these variables. is better than that of the other companies.
Although Turkcell has a market share based upon long years and has more complaints directed than that of the
other companies. it is noticed that it has no superiority over the other companies in terms of perception of its
complaint assessment performance.
6. Conclusion
It is better understood how valuable the customer satisfaction is in order to maintain the customer in especially
today’s competition conditions in which it is cheaper to keep the existing customers than to gain new ones.
Especially the indifference to the complaints has the highest share among the reasons of customers to quit a
company, the efforts provided in this respect should be perceived as a necessity to confront today's competitive
environment. In this context, it should not be ignored that the customer complaints as the feedbacks from
customers are extremely valuable for a company to notice its shortcomings and errors.
CCM processes should be defined by all the companies declared for the customers explicitly, and CCM
processes should be shortened and accelerated as possible. Most importantly the idea “the complaint is a
valuable source of information” should be institutionalized in the organization rather than understanding “the
complaining customer is the problematic customer”. It should be understood more adequately that better
management processes can be achieved with more complaint management processes.

16
7. Limitations and Future Research
In addition to the performance perception of customer complaints taken into consideration in the study. the
measuring of the internal applications of the GSM companies. such as accountability; privacy; wage; continuous
improvement, which were not taken into consideration. is essential in order to see the CCM performance as a
whole. With an integrated approach the real CCM performances will be reached by constantly monitoring the
performances of the companies related to both the perceptions on the client-side and the CCM applications
provided by these companies. Such study matters to be conducted on a larger scale and with such a continuity
were left to the researchers who want to conduct a research on this issue. In this study, attention was drawn only
to the complaint behaviors and the perception of the CCM applications by the customers.
References
Akgül, A., & Çevik, O. (2003). İstatistiksel Analiz Teknikleri SPSS’te İşletme Yönetimi Uygulamaları. Ankara:
Emek Ofset.
Barış, G. (2006). Kusursuz Müşteri Memnuniyeti İçin Şikâyet Yönetimi. MediaCat Kitapları. İstanbul.
Barlow, J., & Moller. C. (1998). Her Şikâyet Bir Armağandır (Çev: Günhan Günay). Rota Yayınları, Istanbul.
Bell, S. J., Mengüç, B., & Stefani, S. L. (2004). When Customers disappoint: A model of relational internal
marketing and customers complaints. Journal of the Academy of Marketing Science, 32(2), 112-126.
http://dx.doi.org/10.1177/0092070303261467
Blodgett, J. G., & Granbois, D. H. (1992). Toward an integrated conceptual model of consumer complaining
behavior. Journal of Consumer Satisfaction. Dissatisfaction and Complaining Behaviour, 5, 95-103.
Retrieved from http://lilt.ilstu.edu/staylor/csdcb/articles/volume5/blodgett%20et%20al%201992.pdf
Boshoff, C. (1999). An instrument to measure satisfaction with transaction-specific service recovery. Journal
of Service Research, 1(3), 236-249. http://dx.doi.org/10.1177/109467059913005
Bozkurt, İ. (2000). Bütünleşik Pazarlama İletişimi. MediaCat Yayınları, İstanbul.
BTK. (2012). Bilgi Teknolojileri ve İletişim Kurumu Üç Aylık Pazar Verileri Raporu. Retrieved from
http://btk.gov.tr/kutuphane_ve_veribankasi/pazar_verileri/ucaylik12_3.pdf
Chan, Y. H. (2005). Basic statistics for doctors. Singapore Med J., 46(2), 54-62. Retrieved from
http://www.sma.org.sg/smj/4602/4602bs1.pdf
Churchill, G. A. (1999). Marketing Research. Methodological Foundations (7th
ed.). The Dryden Press.
Çoban, S. (2005). Müşteri Sadakatının Kazanılmasında Veritabanlı Pazarlamanın Kullanımı. Erciyes
Üniversitesi S.B.E. Dergisi, 19, 295-307. Retrieved from
http://sbe.erciyes.edu.tr/dergi/sayi_19/sayi_19.htm
De Ruyter, K., & Wetzels, M. (2000). Customer equity considerations in service recovery: a cross-industry
perspective. International Journal of Service Industry Management, 11(1), 91-108.
Doğan, G., & Erdoğan, S. (2010). ISO 10002: Müşteri Memnuniyeti ve Müşteri Şikâyetleri Yönetimi. Borusan
Lojistik Örneği, Lisans Proje Ödevi. İktisadi ve İdari Bilimler Fakültesi, Sakarya Üniversitesi.
Ertürk, K. Ö., & Kıyak, C. M. (2011). Müşteri Memnuniyetini Artırma Aracı Olarak Halkla İlişkilere
Maslow’un İhtiyaçlar Hiyerarşisi Pencersinden Bakmak. İletişim Kuram ve Araştırma Dergisi, 32, 127-150.
Retrieved from http://www.iletisimdergisi.gazi.edu.tr/arsiv/32.pdf
Etzel, M. J., & Silverman, B. I. (1981). A managerial perspective on directions for retail customer satisfaction
research. Journal of Retailing, 57(3). 124-136. Retrieved from
hattp://ehis.ebscohost.com/ehost/pdfviewer/pdfviewer?sid=5ed4a35a-c510-4a26-8343-9febb5359262%4
0sessionmgr114&vid=1&hid=103
Gerson, R. (1997). Müşteri Tatmininde Süreklilik (Çev: Tülay Savaşer). İstanbul: Rota Yayınları.
Gilly, M. C., & Hansen, R. W. (1985). Consumer Complaint Handling as a Strategic Marketing Tool. Journal of
Consumer Marketing, 2(4), 5-16. http://dx.doi.org/10.1108/eb008139
Hair, Jr. J. F., Anderson, E. R., Tahtam, R. L., & Black, C. W. (1998). Multivariate Data Analysis. Prentice Hall
Inc.
Kang, J., Zhang, X., & Zheng, Zhao-Hong. (2009). The relationship of customer complaints. satisfaction and
loyalty: Evidence from China’s mobile phone industry. China-USA Business Review, 8(12), 22-36.
Retrieved from

17
http://ehis.ebscohost.com/ehost/pdfviewer/pdfviewer?sid=10153858-72b0-4f78-9342-31842622e238%40s
essionmgr110&vid=1&hid=103
Kızgın, Y. (2008). Genç gsm abonelerinin operatör seçimlerini etkileyen değişkenlerin konumlandırılması
üzerine bir alan araştırması: Muğla Üniversitesi öğrencileri örneği. Balıkesir Üniversitesi Bandırma İ.İ.B.F.
Akademik Fener Dergisi, 10, 134-152.
Lewis, R. C. (1983). Consumers complain what happens when business responds. In Day, R. L. and Hunt. H.
K. (Eds.), International Fare in Consumer Satisfaction and Complaining Behaviour. Department of
Marketing, School of Business, Indiana University, Bloomington.
Li, D., Browne, G. J., & Wether be, J. C. (2007). Online Consumers’ Switching Behavior: A Buyer-Seller
Relationship Perspective. Journal of Electronic Commerce in Organizations, 5(1), 30-42.
http://dx.doi.org/10.4018/jeco.2007010102
Lumley, T., Diehr, P., Emerson, S., & Chen, L. (2002). The Importance of the Normality Assumption in Large
Public Health Data Sets. Annu. Rev. Public Health, 23, 151-169.
http://dx.doi.org/10.1146/annurev.publhealth.23.100901.140546
Martin, B. W. (1997). Müşteri Hizmetlerinde Kalite (Çev. Ahmet Ünver). İstanbul: Rota Yayınları.
Mcdougall, H. G., & Levesque, J. T. (1999). Waiting for Service: the Effectiveness of Recovery. International
Journal of Contemporary Hospitality Management, 11(1), 6-15.
Mjahed, S., & Trike, A. (2009). Toward E-Knowledge Based Complaint Management. Retrieved from
http://www.iajet.org/iajet_files/vol.1/no.1/Toward%20E-Knowledge%20Based%20Complaint%20Manage
ment.pdf
Nakip, M. (2003). Pazarlama Araştırması Teknikler ve (SPSS Destekli) Uygulamalar.
Naylor, G. (2003). The Complaining Customer: A Service Provider’s Best Friend. Journal of Consumer
Satisfaction. Dissatisfaction and complaining Behaviour, 16, 241-248. Retrieved from
http://lilt.ilstu.edu/staylor/csdcb/articles/Volume16/Naylor%202003.pdf
Odabaşı, Y. (2000). Satış ve Pazarlamada Müşteri İlişkileri Yönetimi. Sistem Yayıncılık, İstanbul.
Odabaşı, Y. (2001). Pazarlama Planı Rehberi. Girişimciliği Geliştirme Merkezi, Ankara.
Oliver, R. L. (1997). Satisfaction: A Behavioural Perspective on the Consumer. Boston, MA: McGraw-Hill.
Özdamar, K. (1999). Paket Programlar İle İstatistiksel Veri Analizi. Kaan Kitabevi, Eskişehir.
Özdamar, K. (2003). Modern Bilimsel Araştırma Yöntemler. Kaan Kitabevi, Eskişehir.
Özdemir, E., & Tokol, T. (2008). Kadın Tüketicilere Yönelik Pazarlama Stratejileri. Anadolu Üniversitesi Sosyal
Bilimler Dergisi, 8(2), 57-80.
Parasuraman, A., & Berry, L. (1997). Listening to the Customer: The Concept of Service Quality Information
System. Sloan Management Review, 38(3), 65-76.
Sarıdaldi, E., & Sevim, Ş. (2009). Müşteri Şikayet Yönetim Performansının Değerlendirilmesi. TCDD 3. Bölge
Müdürlüğünde Bir Uygulama. Sosyal Bilimler Enstitüsü Dergisi, 12(22), 110-126.
Stat Trek. (2012). Sampling Distributions. Retrieved December 3. 2012 from
http://stattrek.com/sampling/sampling-distribution.aspx
Statsoft. (2012). Are All Test Statistics Normally Distributed? Retrieved December 3 2012 from
http://www.statsoft.com/textbook/elementary-statistics-concepts/
Stauss, B., & Seidel, W. (1998). Beschwerdemanagement Fehler vermeiden Leistung verbessern Kunden
binden (2nd
ed.). Hanser Verlag, Munchen/Wien.
Stauss, B. (2002). The dimensions of complaint satisfaction: process and outcome complaint satisfaction versus
cold fact and warm act complaint satisfaction. Managing Service Quality, 12(3), 173-183.
Şencan, H. (2002). Bilimsel Yazım. İstanbul Üniversitesi. İşletme Fak. Yayınları.
Whitelock, J., & Jobber, D. (2004). An evaluation of external factors in the decision of UK industrial firms to
enter a new non-domestic market: an exploratory study. European Journal of Marketing, 38(11),
1437-1454. http://dx.doi.org/10.1108/03090560410560182

The Examining of GSM Operators

More Related Content

What's hot (20)

Similar to The Examining of GSM Operators (20)

More from Hossein Abbasi (15)

The Examining of GSM Operators