Available online at www.sciencedirect.

com

Procedia Social and Behavioral Sciences 24 (2011) 1232–1242

7th International Strategic Management Conference

Measuring service quality and a comparative analysis in the

passenger carriage of airline industry
Sabri T. Erdila , O÷uz Yıldızb , a∗
a,b
Marmara University, Istanbul, 34180, Turkey

Abstract

Acquiring new customers and retaining existing customers have long been two important subjects for service
organizations. Creating difference in every section of the service is a key factor to success on realization of these
purposes. When service quality is thought as an important factor for creating difference in service industry, the
importance of estimating service quality provided to customers comes out. Estimating service quality is a hard issue
for service firms because services are intangible, heterogeneous, perishable, and inseparable. Estimating service
quality provides service firms how to manage their marketing operations appropriately. Therefore, this estimation
should be performed with right measurement scales. In this study, first, service marketing literature was reviewed and
then data were gathered via questionnaire forms. Lastly exploratory factor analysis was conducted and two scales
which estimate service quality were compared in the research. The findings reveal that the weighted Servperf scale
has explained perceived service quality more much than the weighted Servqual scale did in the research.

Keywords: Creating difference, service quality, estimating service quality, exploratory factor analysis, Servperf scale, Servqual
scale.

© 2011
© 2011Published
Publishedby
byElsevier
ElsevierLtd.
Ltd.Open access under CC BY-NC-ND license.
Selection and/or peer-review under responsibility 7th International
Selection and/or peer-review under responsibility of 7th International Strategic Management Conference
Strategic Management Conference

1. Introduction

With the effect of globalization, marketplaces have been closer to each other providing consumers to
have lots of information about all products and services. This situation has brought out high competition
level in service industry like other industries have had.

∗
Corresponding author. Tel. + 90-212-308-2226 fax. +90-212-505-9332
Email address: [email protected]

1877–0428 © 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
Selection and/or peer-review under responsibility of 7th International Strategic Management Conference
doi:10.1016/j.sbspro.2011.09.117
 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242 1233

Acquiring new customers and retaining existing customers have long been two important subjects for
service organizations. Creating difference in every section of the service is a key factor to success on
realization of these purposes. When service quality is thought as a great factor on creating difference in
service industry, the importance of estimating service quality provided to customers comes out.
Estimating service quality is a hard issue for service firms because services are intangible, heterogeneous,
perishable, and inseparable. Estimating service quality provides service firms how to manage their
marketing operations correctly. Therefore, this estimation should be done with right measurement scale.

The main purpose of this study is to provide some information about the differences between two
different service quality scales with findings from Turkish application in airline industry and to discuss
comparative results which are debated issues in the service quality literature. One of the two widely used
models has been SERVQUAL model that is based on difference between expectation and performance,
and the other model SERVPERF that is based on only service company’s performance.

First, conceptualization of service quality, SERVQUAL scale, criticisms about SERVQUAL scale, and
SERVPERF scale were examined and then the findings of an application in the passenger carriage of
airline industry were discussed in this research.

In the research, data were gathered via the questionnaire forms used in face- to- face interviews with
1100 passengers in Istanbul Ataturk Airport and Sabiha Gokcen International Airport. Exploratory factor
analysis was held in SPSS XVII program. The results implied that weighted SERVQUAL scale exhibited
five dimensions which are consistent with the theory. Also SERVPERF scale did not demonstrated
unidimensional structure which was defended by scale creators. As an outcome of the survey, estimating
service quality with weighted SERVPERF scale will be a more precious way in the passenger carriage of
airline industry.

2. Literature Review

2.1. Service Quality

There is no consensus on definition of quality. According to [9] quality is zero defect -doing it right
first time. Quality means meeting the requirements of customer. All service should be aimed at meeting
the customer’s requirements by eliminating non quality traits [16]. [14] unambiguously support the
notion that service quality, as perceived by consumers, stems from a comparison of what they feel service
firms should offer (i.e., from their expectations) with their perceptions of the performance of firms
providing the services. Service quality from the provider’s perspective means the degree to which the
service’s features conform to the organization’s specifications and requirements. From customers
perspective, service quality means how well the service meets or exceeds expectations [11].

[14] refer 10 service quality determinants which are used by customers in assessing the service quality.
These are reliability, responsiveness, competence, access, courtesy, communication, credibility, security,
understanding/knowing the customer and tangibles. Also [12] asserts that service quality has six
determinants. These are professionalism and skills, attitudes and behavior, accessibility and flexibility,
reliability and trustworthiness, recovery, reputation and credibility. After that, Bitner has added concept
of servicescape into determinants of service quality. Because services typically require direct human
contact, customers and employees interact with each other within the organization's physical facility.
Ideally, therefore, the organization's environment should support the needs and preferences of both
service employees and customers simultaneously [3].
1234 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242

2.2. Servqual Scale

Learning Servqual scale bases on gap model which proposed by Parasuraman, Zeithaml, and Berry.
The gap model maintains that satisfaction is related to the size and direction of disconfirmation of a
person’s experience vis-à-vis his/her initial expectations. Perceived service quality is further posited to
exist along a continuum ranging from ideal quality to totally unacceptable quality, with some point along
the continuum representing satisfactory quality [14]. According to [22], there have been three service
levels. First of them is desired service level which is defined as the level of service the customer hopes to
receive. Second of them is called as adequate service level which is lower level of expectation than the
other, the level of service the customer will accept. Last level is a gap. Because services are
heterogeneous, the extent to which customers recognize and are willing to accept heterogeneity we call
the zone of tolerance.

As a measurement tool, Parasuraman, Zeithaml, and Berry have identified 10 service quality
dimensions. Servqual scale has consisted of 97 items describing 10 dimensions. Each item has assessed
customer’s expectations and service firm’s performance. For each item a difference score Q (representing
perceived quality along that item) has been defined as Q = P – E, where P and E are the ratings on the
corresponding perception and expectation statements, respectively. Coefficient alpha and item-to-total
correlations for each dimension have been computed. Items whose item-to-total correlations were low and
those whose removal increased coefficient alpha have been deleted. Factor analysis has been applied to
verify the dimensionality of the overall scale. Items have been reassigned and dimensions of scale have
been restructured when necessary. Upon the analysis, the scale has been reduced to 35 items and 7
dimensions. As a result of last purification, the scale has been designed as 22 items and 5 dimensions.
These dimensions are tangibles, reliability, responsiveness, assurance and empathy. Most factors involve
communication and control process implemented in service organizations to manage employees and
consequences of these processes, such as role clarity and role conflict of contact personnel. Within this
perspective the model has been extended [21]. Later, Servqual measurement scale has been revised [13]
by replacing “should” word by “would” and in 1994 by reducing the total number of items to 21, but five
dimensional structure remaining the same [17].

2.3. Criticisms about Servqual Scale

We find useful to give some criticisms about Servqual in this section. According to [7] Servqual
conceptualization is in fact flawed because it is based on a satisfaction paradigm rather than an attitude
model. Also they alleged that expectancy-disconfirmation judgments however, are distinct from both
consumer satisfaction judgments and service quality perceptions, involve calculated [8]. In relation to the
gaps of Servqual, the expectation section is not useful. The dominant component in difference scores is
clearly the perceptions scores [1]. Also Teas has investigated the same subject. There were same scores of
different processes (1 – 1 = 0, 2 – 2 = 0, 3 – 3 = 0 etc.). He has questioned whether this scores have same
meanings or not [19]. Another criticism is that Servqual fails to capture the dynamics of changing
expectations. An expectation score in a particular year may not mean the same thing as expectation score
which is taken the year after one [5]. [10] has pointed out that Servqual includes only certain aspects of
service quality, and that it fails to capture other potentially less controllable components of service
provision. There was another criticism about dimensionality. Servqual scale has different number of
dimensions in some service industries. The Scale was appropriate for utility services [1]. Parallel with it,
[6] has suggested that Servqual dimensions are not generic. The scales should be refined by factor
analysis and reliability tests before commercial application. Servqual involves the calculation of the
difference between expectations and perceptions. Although Servqual had high reliability, its reliability
was below that of a non-difference score measure of service quality. Moreover, not only did Servqual fail
 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242 1235

to achieve discriminant validity from its components, but the perceptions component by itself performed
as well as the difference score on a number of criteria [4].

2.4. Servperf Scale

Servperf scale has come out as an opposite of Servqual scale. According to Servperf model, service
marketing literature clearly supports the performance-only (Servperf) approach. Also Servperf scale
explains more of variations in service quality than does Servqual. Furthermore literature review and the
analysis of the structural models suggest that Servqual conceptualization is in fact flawed: it is based on a
satisfaction paradigm rather that an attitude model. The performance-based developed (Servperf) is
efficient in comparison with the Servqual scale; it reduces by 50% the number of items that must be
measured. The analysis of the structural models also supports the theoretical superiority of Servperf scales
[7]. In the paper which assessed Servperf empirically it was alleged that Servperf scale does not exhibit a
five-factor structure in the research in a generalizable fashion and has unidimensional structure [18].
Servperf can provide managers with summed overall service quality score that can be plotted relative to
time and specific consumer subgroups [8].

3. Methodology

3.1. Research Goal

The present study is an attempt to make a comparative assessment of the SERVQUAL and the
SERVPERF scales in Turkish context in terms of their validity, ability to explain variance in the overall
service quality. Data for making comparisons among the unweighted and weighted versions of the two
scales were collected through a survey of the passengers of the airline companies in Turkey.

3.2. Sample and Data Collection

Using the face to face survey method, 1100 passengers of Istanbul Ataturk Airport and Sabiha Gokcen
International Airport were approached. The field work was done between 24 August 2010 and 03
September 2010. After repeated follow-ups, 1080 duly filled-in questionnaires could be collected
constituting a 98 per cent response rate. The sample was deliberately restricted to passengers of Istanbul
Ataturk Airport and Sabiha Gokcen Airport and was equally divided between these two groups.
Convenience sampling was employed for selecting respondents from these two groups. More than half of
the participants were men and under the age of 35. Participants have six different occupation and income
groups. 1080 surveys were processed by using factor analysis in SPSS XVII.

3.3. Measures

Parasuraman, Zeithaml and Berry’s (1988) 22-item SERVQUAL instrument was employed for
collecting the data regarding the respondents’ expectations, perceptions, and importance weights of
various service attributes. Wherever required, slight modifications in the wording of scale items were
made to make the questionnaire understandable to the surveyed respondents. These items were included
to assess the validity of the multi-item service quality scales used at our end.

All the scale items were obtained on a 5-point Likert scale ranging from (5) for “strongly agree” to
(1) for “strongly disagree”. Variables which provide a comparison of the basic variables of the scales are
listed in table – I.
1236 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242

Table 1 Variables of and Dimensions of Servqual

Oprerationalized SERVQUAL statements

Service quality dimensions Variables Statements

Tangibles V1 Airlines have modern-looking aircraft.

V2 The office, terminal and gate facilities of airlines are visually appealing.

V3 Airline staff uniform appearance.

V4 Airline’s materials associated with its service are visually appealing.

Reliability V5 When airlines promise to do something by a certain time, they do.

V6 When a customer has a problem, an airline shows a sincere interest in solving it.

V7 Airlines perform the service right the first time.

V8 Airlines provide their services at the time they promise to do so.

V9 Airline companies keep accurate records

Responsiveness 10 Airlines tell customers exactly when services will be performed.

V11 Employees in airlines give prompt service to customers.

V12 Employees of airlines are willing to help customers.

V13 The airline employees are never too busy to not answer customers' requests.

Assurance V14 Behavior of airline employees creates a sense of confidence in customers.

V15 Customers of airlines feel safe in their transactions.

V16 The airline employees are always respectful to customers.

V17 Employees of airlines have the knowledge to answer customers’ questions.

Empathy V18 Airlines give customers individual attention.

V19 Airlines have operating hours convenient to all their customers.

V20 Airline companies have employees who give customers personal attention.

V21 Airlines have the customer’s best interests at heart.

V22 The employees of airlines understand the specific needs of their customers.
 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242 1237

Table 2. Results of Exploratory Factor Analysis

Unweighted SERVPERF Unweighted SERVQUAL

Factor Factor
Variables Dimensions Variables Dimensions
Loadings Loadings
Responsiveness/Assurance (Į=0.838; VE=%17.658) Reliability/Responsiveness/Assurance (Į=0.887;
VE=%22.735)
V10 0.530 V6 0.583
V11 0.590 V7 0.620
V12 0.512 V8 0.700
V13 0.515 V9 0.631
V14 0.568 V10 0.654
V15 0.664 V11 0.611
V16 0.668 V12 0.543
V17 0.566 V13 0.514
V19 0.547 V15 0.593
Reliability (Į=0.790; VE=%12.708) V16 0.525
V5 Reliability 1 0.645
V6 Reliability 2 0.615
V7 Reliability 3 0.750
V8 Reliability 4 0.765
V9 Reliability 5 0.524
Empathy (Į=0.804; VE=%17.409)
Empathy (Į=0.781; VE=%11.825) V17 Empathy 1 0.526
V18 Empathy 1 0.538 V18 Empathy 2 0.563
V20 Empathy 2 0.599 V20 Empathy 3 0.669
V21 Empathy 3 0.768 V21 Empathy 4 0.738
V22 Empathy 4 0.749 V22 Empathy 5 0.727
Tangibles (Į=0.614; VE=%8.908) Tangibles (Į=0.682; VE=%11.937)
V1 Tangibles 1 0.659 V1 Tangibles 1 0.645
V2 Tangibles 2 0.672 V2 Tangibles 2 0.703
V3 Tangibles 3 0.602 V3 Tangibles 3 0.602
V4 Tangibles 4 0.554 V4 Tangibles 4 0.560
N = 1080, KMO = 0.938 N = 1080, KMO = 0.959
Bartlett’s Sph X²= 7741.276; p = 0.000 Bartlett’s Sph X²= 9972.155; p = 0.000
Total Explained Variance = %51.099 Total Explained Variance = %52.081

4. Findings

In this section, findings will be given about a comparative study of SERVQUAL and SERVPERF
scales which have weighted and unweighted forms. Basic variables have been showed in table – I.
Dimensions of scales have been evaluated in terms of Servqual’s five basic dimensions. Dimensions,
items of dimensions, factor loadings of items, reliability of dimensions and variances of dimensions have
been located in table – II and table – III.

The report Kaiser-Meyer-Olkin (KMO) values which show the accordance of exploratory factor
analysis, have appeared as 0.959 for the unweighted Servqual, as 0.907 for the weighted Servqual on the
1238 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242

other hand as 0.938 for the unweighted Servperf, as 0.908 for the weighted Servperf in table – II and III.
When KMO takes value of 1, this indicates that these variables could predict each other perfectly [20].
Bartlett’s sphericity test and p values of scales are X² = 9972,155 and p < 0,05 for unweighted Servqual;
X² = 10331,295 and p < 0,05 for weighted Servqual; X² = 7741,276 and p < 0,05 for unweighted Servperf;
X² = 32185,991 and p < 0,05 for weighted Servperf. If the p value is below 0,05, the relationship between
the variables have proven to be sufficient to factor analysis [20].

It can be seen from table II that; the unweighted Servperf which bases only service firm’s performance
has four dimensions. All tangibles and reliability variables of Servqual have emerged in the unweighted
Servperf. The dimensions of responsiveness and assurance have merged under a single dimension in the
unweighted Servperf. The dimension of responsiveness and assurance has taken place together with V19.
The unweighted Servperf has had the rates of explained variance of 17.658% for
responsiveness/assurance, 12.708% for reliability, 11.825% for empathy, 8.908% for tangibles.

According to the exploratory factor analysis, the unweighted Servqual has emerged in three
dimensions which are tangibles, empathy, reliability/responsiveness/assurance (Table – II). As a result of
factor analysis; V5, V14 and V19 were excluded from the analysis because they have factor loadings
below 0.50. There have been all tangibles items in the unweighted Servqual. V17 was accompanied by
the dimension of empathy. The dimensions of reliability, responsiveness and assurance have merged
under a single dimension in the unweighted Servqual. The unweighted Servqual has had the rates of
explained variance of 22.735% for reliability/responsiveness/assurance, 17.409% for empathy, 11.937%
for tangibles. It is possible to say that dimensions which have more variables than the others, have
explained more much variance both in the unweighted Servqual and in the unweighted Servperf.

While the unweighted Servperf has explained 51.099 per cent of perceived service quality, unweighted
Servqual has explained 52.081 per cent of perceived service quality.

All dimensions of the unweighted Servperf and Servqual have enough reliability values. If Cronbach’s
alpha (Į) value is equal to and above 0.70, the scale is considered to be reliable. On the other hand, when
the scale has less question in the dimension, the acceptable limit is 0.60 and above [20]. In this regard,
sub-dimensions of both scales have reliability and it is possible to say that scales measure the desired
property.

Exploratory factor analysis results can be seen in table – III. Both weighted Servperf and weighted
Servqual scales have an equal number of dimensions and dimensions consisted of the same variables.
These are tangibles, reliability, responsiveness, assurance and empathy. The weighted Servperf has had
rates of explained variance of 15.937% for tangibles, 20.719% for reliability, 16.626% for
responsiveness, 16.764% for assurance, 20.350% for empathy. On the other hand, the weighted Servqual
has had rates of explained variance of 10.860% for tangibles, 14.905% for reliability, 11.462% for
responsiveness, 11.748% for assurance, 15.479% for empathy.

Table 3. Results of Exploratory Factor Analysis

Weighted SERVPERF Weighted SERVQUAL

Variables Dimensions Factor Loadings Variables Dimensions Factor Loadings

Tangibles (Į=0.958; VE=%15.937) Tangibles (Į=0.757; VE=%10.860)

 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242 1239

V1 Tangibles1 0.903 V1 Tangibles1 0.771

V2 Tangibles2 0.903 V2 Tangibles2 0.734

V3 Tangibles3 0.894 V3 Tangibles3 0.745

V4 Tangibles4 0.893 V4 Tangibles4 0.740

Reliability (Į=0.975; VE=%20.719) Reliability (Į=0.861; VE=%14.905)

V5 Reliability1 0.924 V5 Reliability1 0.770

V6 Reliability2 0.921 V6 Reliability2 0.745

V7 Reliability3 0.937 V7 Reliability3 0.814

V8 Reliability4 0.932 V8 Reliability4 0.815

V9 Reliability5 0.911 V9 Reliability5 0.698

Responsiveness (Į=0.967; VE=%16.626) Responsiveness (Į=0.815; VE=%11.462)

V10 Responsiveness1 0.946 V10 Responsiveness1 0.765

V11 Responsiveness2 0.938 V11 Responsiveness2 0.747

V12 Responsiveness3 0.944 V12 Responsiveness3 0.722

V13 Responsiveness4 0.937 V13 Responsiveness4 0.683

Assurance (Į=0.970; VE=%16.764) Assurance (Į=0.817; VE=%11.748)

V14 Assurance1 0.937 V14 Assurance1 0.719

V15 Assurance2 0.940 V15 Assurance2 0.753

V16 Assurance3 0.933 V16 Assurance3 0.742

V17 Assurance4 0.940 V17 Assurance4 0.748

Empathy (Į=0.969; VE=%20.350) Empathy (Į=0.878; VE=%15.479)

V18 Empathy1 0.926 V18 Empathy1 0.754

V19 Empathy2 0.903 V19 Empathy2 0.762

V20 Empathy3 0.937 V20 Empathy3 0.803

V21 Empathy4 0.927 V21 Empathy4 0.833

V22 Empathy5 0.918 V22 Empathy5 0.803

N = 1080, KMO = 0.908 N = 1080, KMO = 0.907

Bartlett’s Sph X² = 32185.991; p = 0.000 Bartlett’s Sph X² = 10331.295 ; p = 0.000

Total Explained Variance = %90.395 Total Explained Variance = %64.454

1240 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242

When examining the total explained variance in table – III, the weighted Servperf has explained
90.395 per cent of perceived service quality on the other hand; the weighted Servqual has explained
64.454 per cent of perceived service quality.

Table 4. Alternate Service Quality Scales – Correlation Coefficients

SERVQUAL (P SERVPERF (P) WEIGHTED SERVQUAL WEIGHTED

– E) I(P – E) SERVPERF I(P)
SERVQUAL (P – E) 1

SERVPERF (P) 0.760 1

WEIGHTED SERVQUAL 0.967 0.751 1

I(P – E)
WEIGHTED SERVPERF 0.744 0.963 0.776 1
I(P)

At this stage, convergent and discriminant validity of four measurement scales was assessed by
computing correlations coefficients for different pairs of scales. In this regard, it is useful to give some
information about the validity of the convergent and discriminant. Convergent validity [2] expresses
uniform and a high level of correlation among variables or sub-dimensions which form a structure.
Otherwise, discriminant validity expresses an acceptable limit of correlation level among these variables.
The results are summarized in table – IV. The presence of a high correlation between alternate measures
of service quality is a pointer to the convergent validity of all the four scales. The unweighted Servqual
scale is, however, found to have a stronger correlation with other similar measures, Servqual and
importance weighted service quality measures. In conclusion, it is the unweighted Servqual scale which is
found possessing the highest discriminant validity.

5. Conclusion

According to exploratory factor analysis and correlations coefficients, the validity of and the reliability
of the unweighted and weighted Servperf/Servqual scales have been verified in Turkish culture once more
in this study.

The dimension which has the largest percentage of explanation is responsiveness/assurance for the
unweighted Servperf in the research. Also reliability/responsiveness/assurance has been the dimension
which has the largest percentage of explanation for the unweighted Servqual in the research. When the
comparison is made on the basis of dimensions, the dimension of empathy has the explanation rate of
11.825 per cent in the unweighted Servperf scale and the explanation rate of 17.409 per cent in the
unweighted Servqual. The dimension of tangibles has the explanation rate of 8.908 per cent for in the
unweighted Servperf scale and the explanation rate of 11.937 per cent in the unweighted Servqual. The
unweighted Servqual has explained more much variance both in the empathy dimension and in the
tangibles dimension.

When examining the total explained variance, the unweighted Servqual has explained perceived
service quality more than the unweighted Servperf. The difference is close to 1 per cent. The unweighted
Servqual has been found to be superior to the unweighted Servperf.

The weighted Servperf has consisted of five dimensions. The dimension which has the largest
percentage of explanation is reliability which is closely followed by empathy, for the weighted Servperf
 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242 1241

in the research. The weighted Servqual has consisted of five dimensions too. The dimension which has
the largest percentage of explanation is empathy which is closely followed by reliability, for the weighted
Servqual in the research. In this regard, it is possible to say that the dimensions’ variables of empathy and
reliability are important both in the weighted Servperf and in the weighted Servqual, in Turkish air
transportation. When the comparison is made on the basis of dimensions, the dimension of tangibles has
the explanation rate of 15.937 per cent in the weighted Servperf scale and the explanation rate of 10.860
per cent in the weighted Servqual. The dimension of reliability has the explanation rate of 20.719 per cent
in the weighted Servperf scale and the explanation rate of 14.905 per cent in the weighted Servqual. The
dimension of responsiveness has the explanation rate of 16.626 per cent in the weighted Servperf scale
and the explanation rate of 11.462 per cent in the weighted Servqual. The dimension of assurance has the
explanation rate of 16.764 per cent in the weighted Servperf scale and the explanation rate of 11.748 per
cent in the weighted Servqual. Lastly, the dimension of empathy has the explanation rate of 20.350 per
cent in the weighted Servperf scale and the rate of 15.479 per cent in the weighted Servqual.
In terms of weighted scales, Servperf has explained more much variance in all dimensions. The
weighted Servperf scale’s variables also have higher factor loadings. Moreover Servperf has taken the
lead within all reliability rates.

When examining the total explained variance, the weighted Servperf has explained perceived service
quality much more than the weighted Servqual. The difference is close to 26 percent. The weighted
Servperf has been found superior to the weighted Servqual.

The results demonstrated that the weighted SERVQUAL scale exhibited five dimensions which are
consistent with theory. Also SERVPERF scale did not demonstrate unidimensional structure which was
defended by scale creators. Moreover, the unweighted estimating models have explained service quality
in lower rate. As an outcome of the research, estimating service quality with the weighted SERVPERF
scale will be a more precious way in the passenger carriage of airline industry.

References

[1] Babakus, E. and G.W. Boller. (1992). An Empirical Assessment of the SERVQUAL Scale. Journal of Business
Research. 24, 253-268.
[2] Bagozzi, R.P. (1981). Evaluating Structural Equation Models with Unobservable Variables and Measurement Error. A
Comment. Journal of Marketing Research 18.3, 375-381.
[3] Bitner, M.J. (1992). Servicescapes: The Impact of Physical Surroundings on Customers and Employees. The Journal of
Marketing, Vol. 56.2, 57-71.
[4] Brown, T. J., G. A. Churchill and J. P. Peter. (1993). Research Note: Improving the Measurement of Service Quality.
69.1, 127-139.
[5] Buttle, F. (1996). SERVQUAL: review, critique, research agenda. European Journal of Marketing. 30.1, 8-32.
[6] Carman, M.C. (1990). Consumer Perceptions of Service Quality: An Assessment of the SERVQUAL Dimensions.
Journal of Retailing. 66.1, 33-55.
[7] Cronin, J. J. and S. A. Taylor. (1992). Measuring Service Quality: A Reexamination and Extension. Journal of
Marketing. 56.Jully, 55-68.
[8] Cronin, J. J. and S. A. Taylor. (1994). SERVPERF Versus SERVQUAL: Reconciling Performance-Based and
Perceptions-Minus-Expectations Measurement of Service Quality. Journal of Marketing. 58.January, 125-131.
[9] Crosby, P.B. (1979). Quality is Free. New York: McGraw-Hill.
[10] Coulthard, L. J. M. (2004). Measuring Service Quality; A review and critique of research using SERVQUAL.
International Journal of Market Research. 64.Quarter 4, 479-497.
[11] Fisk, R.P., S.J.Grove and J.John. (2004). Interactive Services Marketing. Second Edition. United States of America:
Houghton Mifflin Company.
Gronroos, C. (1988). Service Quality: The Six Criteria Of Good Perceived Service. Review of Business. Winter, 10-13.
1242 Sabri T. Erdil and Oğuz Yıldız / Procedia Social and Behavioral Sciences 24 (2011) 1232–1242

[12] Parasuraman, A., L.L.Berry and V.A.Zeithaml. (1991). Refinement and Reassessment of the SERVQUAL scale. Journal
of Retailing. 67.4, 420-450.
[13] Parasuraman, A., V.A.Zeithaml and L.L.Berry. (1985). A Conceptual Model of Service Quality and Its Implications for
Future Research. Journal of Marketing. 49.Fall, 41-50.
[14] Parasuraman, A., V.A.Zeithaml and L.L.Berry. (1988). SERVQUAL: A Multiple-Item Scale for Measuring Consumer
Perceptions of Service Quality. Journal of Retailing. 64.1, 12-40.
[15] Rosander, A.C. (1991). Deming’s 14 Points Applied to Services. New York: Marcel Dekker, Inc.
[16] Seth, N., S.G.Deshmukh and P.Vrat. (2005). Service quality models: a review. International Journal of Quality and
Reliability Management. 22.9, 913-949.
[17] Taylor, S.A and J.J. Cronin, (1994). An Empirical Assessment of the Servperf Scale. Journal of Marketing. Fall, 52-68.
[18] Teas, K.T. (1993). Expectations, Performance Evaluation, and Consumers’ Perceptions of Quality. 57.4, 18-34.
[19] Sipahi, B, E.S. Yurtkoru and M. Çinko. (2008). SOSYAL BøLøMLERDE SPSS’LE VERø ANALøZø. østanbul: Beta
Basım Yayım Da÷ıtım A.ù.
[20] Zeithaml, V.A., L.L. Berry and A. Parasuraman. (1988). Communications and Control Processes in the Delivery of
Service Quality. Journal of Marketing. 52.April, 35-48.
[21] Zeithaml, V.A., L.L. Berry and A. Parasuraman. (1993). The Nature and Determinants of Customer Expectations of
Service. Journal of the Academy of Marketing Science Volume 2.1,1-12.