Introduction

Case studies
There can be little doubt that quality is,
Measuring and nowadays, among the most critical aspects for
managing service the strategic management of service firms.
Customer satisfaction and loyalty ± secured
quality: integrating through high-quality products and services
customer expectations providing value for money for the consumer ±
are essential for long-term survival, let alone
Marco Antonio Robledo long-term success (Cina, 1990; Daniel, 1992;
Gale, 1994; Reichheld and Sasser, 1990;
Shycon, 1992; Zeithaml et al., 1985, 1990;
Robledo, 1996, 1998).
Because of its importance, researchers
have devoted a great deal of attention
to service quality, but there are still
The author some hot debates going on in the area
Marco Antonio Robledo is Titular Professor in the that need to be addressed. One of the
Business and Economics Department of the University of more controversial ones refers to the
the Balearic Islands, Palma, Spain. conceptualisation and measurement of
service quality.
The underlying premiss of our article is
Keywords
that, if service quality assessment and
Airlines, Service quality, Customer care, Measurement management depends on customer
expectations, as other research has suggested,
Abstract then a strategic effort should be made at
The pressures driving successful organisations toward top company level to monitor and manage those
quality services make the measurement of service quality expectations.
and its subsequent management of overall importance. With that in mind, the objectives of this
The study compares four different methods for measuring article are four-fold:
service quality within an airline setting. Six instruments (1) review and integrate recent research in
are used to measure the service quality of three service quality;
international airline companies. The validity and reliability (2) test alternative perceived quality
of the six different models is examined: it is concluded measurement models;
that unweighted SERVPEX methods are superior to the (3) study if the service expectation concept is
others. The dimensionality of quality in airlines is explored justified as a component of a perceived
and three factors appear as determinants: tangibility, quality measure; and
reliability, and customer care. The author states that (4) determine whether a new area of
understanding customer expectations is a prerequisite for quality management, called
delivering superior service, since customers evaluate expectations management, needs to be
service quality by comparing their perceptions of the defined.
service with their expectations. As a result, a model to
In addressing these objectives, I provide a
manage expectations is proposed. A set of research
concise synthesis of the extant literature on
directions is offered.
the subject and extend the literature in three
significant ways.
Electronic access
First, my study involves a comprehensive
The research register for this journal is available at multicompany examination and comparison
http://www.mcbup.com/research_registers of the different service quality assessment
The current issue and full text archive of this journal is models. Second, in addition to examining the
available at general relationship between service quality
http://www.emerald-library.com/ft and expectations, I explore the different
sources that can influence customer
expectations. Third, I incorporate a model for
Managing Service Quality
Volume 11 . Number 1 . 2001 . pp. 22±31 managing customer expectations in a
# MCB University Press . ISSN 0960-4529 comprehensive way.
22
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Conceptual background (1) Disconfirmation models are based on the

disconfirmation paradigm (Bitner, 1990;
Perceived quality reflects the opinion of the Bolton and Drew, 1991a). Quality is
consumer regarding the superiority or global therefore defined as the gap between
excellence of a product or service (Zeithaml, customers' expectations and perceptions
1988). However, finding the components of
(C = P ± E), and a customer will perceive
perceived quality has proven to be more
quality positively only when the service
elusive. Perception, obviously, is one of those
provider meets or exceeds his
components. Many authors pretend
expectations. The most important model
expectations to be a second component[1].
Generally speaking, most researchers developed upon this theory is
acknowledge that customers have SERVQUAL (Parasuraman et al., 1985).
expectations and that they play a certain role (2) Perception models: this second model,
as standards or reference points used by mainly resulting from examinations and
consumers to evaluate the performance of a assessments of Parasuraman et al.'s gap
company (Oliver, 1980; Brown and Swartz, theory, is based only on perceptions of
1989; Bolton and Drew, 1991a, 1991b; performance. Frameworks for service
Zeithaml et al., 1990, p. 143; Parasuraman et quality measurement following that
al., 1985, 1988, 1994). However, the approach are, for example, SERVPERF
inclusion or not of expectations as a (Cronin and Taylor, 1992), and EP
determinant of perceived service quality have (Teas, 1993).
resulted in two conflicting paradigms:
(1) The disconfirmation paradigm: according Weighted vs. unweighted models
to this approach customers evaluate a The discussion is about the convenience of
service by comparing their perceptions of weighting or not weighting the specific criteria
the service received with their by which customers evaluate service quality
expectations. according to the importance they assign to
(2) The perception paradigm: supporters of each of those criteria. Hence, we find
this paradigm maintain that expectations
weighted models and unweighted models.
are irrelevant and even misleading
information for a model intended to
Dimensions of service quality
evaluate perceived service quality. They
Consumers evaluate providers of services
maintain that the perception of the
along dimensions, which are groupings of
customer is the only measure required.
criteria. The question is one of determining
As a result of the lack of agreement in that area, those dimensions, and if they are general for
customer expectations of service quality have any service company or, rather, are context
not been successfully understood and specific.
examined. In particular, sources of expectations
remain largely unexplored and expectations
management is an area of study yet to be
The study
developed.
However, the debate that has originated the The findings discussed in this article are a
greatest amount of attention among service result of a study in which customer
quality experts is the one related to the
assessments of service quality were measured
measurement of service quality. Basically
for three airline companies. Different
there are three areas of debate when it comes
conceptualisations and measures of service
to determining the ideal method of measuring
quality, derived from the debate areas
service quality, the first one of them directly
related to the problem of expectations. previously explained, were assessed within the
airline business setting. By comparing the
Disconfirmation models vs. perception different measures of service quality, recent
models research on service quality is extended and we
Here the topic of debate is whether customer hope to shed some light over the debate areas
expectations[2] have to be taken into account previously highlighted and to assist airline
in the measurement of perceived service practitioners with the assessment of
quality. The two different approaches are: service quality.
23
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Design of the questionnaires the SERVQUAL model. The SERVQUAL

Weighted and unweighted versions of three model identifies specific criteria by which
different approaches were conceived, resulting customers evaluate service quality. These
in six questionnaires (see Appendix 1): criteria are classed in five major dimensions
(1) Original SERVQUAL P±E Model (Parasuraman et al., 1984, 1988):
(disconfirmation model): the (1) Tangibles. The appearance of physical
questionnaire consists of two sections: a facilities, equipment, personnel, and
section to measure customers' service communications materials.
expectations of companies within a (2) Reliability. The ability to perform the
specific sector (e.g. airlines) and a promised service dependably and
corresponding section to measure accurately.
customers' perceptions of a particular (3) Responsiveness. The willingness to help
company in that sector. customers and provide prompt service.
(2) SERVQUAL measurement weighted by (4) Assurance. The competence of the system
importance. and its credibility in providing a
(3) A measurement of the disconfirmation in courteous and secure service.
a single questionnaire (SERVPEX), (5) Empathy. The approachability, ease of
which includes expectations and access and effort taken to understand
perceptions in a singles scale (from customers' needs.
``Much worse than expected'' to ``Much
However, the factor structure obtained differs
better than expected'').
from the one hypothesised in two ways. First,
(4) Weighted SERVPEX.
two items belong to different factors from the
(5) A measurement of performance only
ones hypothesised. Second, and more
(SERVPERF), based on the perception
important, the responsiveness, empathy and
models.
assurance dimensions show considerable
(6) Performance measure weighted by
overlap and load on the same factor.
importance (weighted SERVPERF).
The resulting three-factor structure
explains 60.32 per cent of the variation for the
The sample
SERVQUAL model, 65.04 per cent for
The research was done in collaboration with
SERVPERF and 61.16 per cent for
three commercial airlines. Scheduled and
SERVPEX. It includes the following
charter flights were included, with
dimensions:
destinations: mainland Spain, Majorca,
. Tangibles. The appearance of airplanes,
Canary Islands, Germany and the UK. The
physical facilities, equipment, personnel,
aggregate sample contained 1,152 customers
and communications materials, and the
and 17 flights (see Table I) for a combined
appearance and taste of food and
response rate of 35.4 per cent. A balance
beverages served on board.
between the different profiles of the airport
. Reliability. The ability to perform the
population during summer and winter
promised service dependably and
months, days of the week and hours of the day
accurately.
was considered, in order to improve the
. Customer care (a dimension consisting
representativeness of the sample.
basically of responsiveness, empathy and
Dimensionality of service quality assurance). Attitude and ability of
To verify the dimensionality of the 26 items, employees to provide personalised service
the scores for these items were factor in a courteous, efficient and secure manner.
analysed. The scales were hypothesised to
have the five-dimensional structure used in Reliability
The reliability of each of the six scales far
Table I Samples for each type of questionnaire and company exceeded Nunnally's (1967) guideline for
Company A Company B Company C Total basic research (Cronbach's alpha coefficient
5 0.70). In particular the results for every
SERVPERF 100 95 281 476
model were the following:
SERVPEX 65 311 0 366 . SERVPERF 0.9635;
SERVQUAL 100 95 115 310 . SERVPEX 0.9593;
265 501 396 1,152 . SERVQUAL 0.9548.
24
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Such high results demonstrate a high showed a high correlation between

reliability of all the instruments. Although each other (see Table II), a result
SERVPERF obtained the best results, the which means that all of them measure
differences are not significant. the same construct.
A second analysis studied the changes in the . Discriminant-validity analysis shows
alpha coefficients when any of the items were the extent to which a scale is new and
eliminated. In all cases the reliability not just a reflex of other variables.
diminished (see Appendix II for results).
The six questionnaires have this kind
of validity since they have a higher
Validity
correlation between each other than
Several different forms of validity can serve as
criteria for assessing the psychometric with other related variables (i.e. the
soundness of a scale (Grapentine, 1995): face three contrast questions included at
validity, convergent validity, discriminant the end of the questionnaires).
validity, and predictive or concurrent validity. (3) Predictive or concurrent validity. This facet
Sufficient information is available from the of validity refers to the extent to which
current study to perform a comparative scale scores are associated as
evaluation of the six scales on each of these hypothesised with other conceptually
forms of validity: related measures. Predictive/concurrent
(1) Face validity. Face validity, a subjective validity was examined in the current
criterion reflecting the extent to which study, correlating each scale separately
scale items are meaningful and appear to with the three contrast questions
represent the construct being measured,
(questions 27, 28 and 29 of the
was explicitly assessed a priori in the
questionnaires and a fourth variable
current study. Preliminary discussions
which is the mean average of the three
with airline customers and extensive
interviews with company executives and questions). A regression analysis was
technical personnel resulted in undertaken using as independent variable
modifications in wording and content of each one of the four contrast questions
the questionnaires. Final feedback from and as dependent variables the 26 items
executives (in each of the three of the questionnaires. The resulting R2
participating companies) confirmed that coefficients (see Table III) are high for
the resulting scales had face validity. the six scales, but SERVPEX performs
(2) Construct validity. The procedure to clearly better than the others. SERVPEX
examine construct validity was the is, therefore, the scale which explains a
following: higher proportion of the variation of the
. The questionnaires were pre-tested service quality variable. It is interesting to
with university students. In
note that SERVPERF explained better
particular, they were asked if they had
the variation of the variable, ``overall
any difficulties with the instructions
satisfaction'', than of ``service quality''
and the items in the questionnaires.
. Convergent-validity analysis pertains which indicates that SERVPERF can be
to the extent to which scale items more appropriate to measure customer
assumed to represent a construct do satisfaction than service quality. It was
in fact ``converge'' on the same also revealed that, for all formulations,
construct. The reliability of a scale as the unweighted versions outperformed
measured by coefficient alpha reflects their weighted counterparts on predictive
the degree of cohesiveness among the validity.
scale items and is therefore an
Drawing from the overall validity analysis and
indirect indicator of convergent
validity. As already mentioned, citing evidence from the results obtained, it is
coefficient-alpha values for the six concluded that the unweighted SERVPEX
scales are fairly high. Another has better validity than its rivals. It appears
method of examining convergent that this measure provides a more valid
validity is to correlate the six scales explanation of service quality because of its
between each other. All of them predictive validity.
25
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Table II Correlation coefficients

SERVQUAL SERVQUALP SERVPERF SERVPERFP SERVPEX SERVPEXP
SERVQUAL 1,0000 0,9599 0,8050 0,6568 0,7721 0,6760
SERVQUALP 0,6968 1,0000 0,7859 0,6598 0,6859
SERVPERF 1,000 0,9116 0,7450 0,6984
SERVPERFP 1,0000 0,6581 0,6659
SERVPEX 1,0000 0,6472
SERVPEXP 1,0000

Table III R2 by type of questionnaire and contrast variable On the other hand, including importance
assessments does not improve the predictive
Quality Satisfac. Expect. Ideal
power of the SERVPEX model. This does not
SERVPERF 0.6974 0.7880 0.4851 0.5406 mean that the relative importance of the
SERVPEX 0.7597 0.6968 0.7438 0.6983 different dimensions should not be taken into
SERVQUAL 0.5709 0.5445 0.4401 0.7438 account but, rather, that this calculation
Weighted SERVQUAL 0.5171 should not be made by the respondent but
Weighted SERVPERF 0.6580 through statistical analysis of the scores.
Weighted SERVPEX 0.7035

Diagnosis power
As Parasuraman et al. (1991) noted, however, Implications
the measurement of expectations can serve a
If, as has been demonstrated in this study,
diagnostic function for managers, and
expectations are an integral part of service
therefore SERVQUAL offers more
quality evaluations, it is necessary to obtain a
information than the other questionnaires.
better understanding of them. Another
Nonetheless, I observed that in this airline
important implication is that the company has
setting, passengers have uniformly high
to manage appropriately those expectations,
expectations across all dimensions, a finding
so it can match them with its performance.
which brings into question the diagnostic For this purpose, I offer a model for managing
utility of the expectation measurement. expectations (see Figure 1) that is divided into
two different parts.

Conclusions of the empirical study Expectations research

The research part of the model investigates all
My purpose was to examine alternative models the relevant aspects necessary to acquire a
of consumers' perceptions of service quality. sufficient understanding of the nature, origins
The examination of the alternative and level of customer expectations, in
perceived quality models suggests that models particular:
that measure service quality considering . perceived service quality;
expectation (the disconfirmation paradigm) . level of expectations; and
are superior to models that measure service . sources of expectations.
quality as a function of performance only.
However, the SERVQUAL approach is Perceived service quality
inferior to the one I called SERVPEX. On the The use of the SERVPEX questionnaire is, as
basis of correlation coefficients reflecting I have demonstrated in this article, the basic
predictive validity, I conclude that the most tool for a perceived service quality
adequate scale with the highest predictive assessment. It provides the basic information
power is an unweighted disconfirmation scale about the customer's opinion of the quality
of single administration, here called the company is offering, relative to his/her
expectations.
SERVPEX. This formulation has 26 items
and a three-factor structure that seem to Level of expectations
define quality in airline service (tangibles, If a company does not know at what level
reliability, and customer care). It is quite customer expectations are, it will be
versatile because it is compatible with the five- impossible to find out the exact reason why
gap model devised by Parasuraman et al. they are not matching them. On the one
26
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Figure 1 Expectations management model

hand, it could be that the customer . Past experience of the customer, not only
expectations are adequate but performance is with the service provider but with
not. On the other hand, it could well be that competitors and companies of other
performance is correct but customer sectors (Cadotte et al., 1987). In service
expectations are simply too high for the quality measures it is clear that this is the
company to be able to match them. The only most important source, but all the others
way to find this out is using an expectations have an effect on the customer's
questionnaire (like the one I used in this study expectations as well, especially in the
for the first part of SERVQUAL). It is not absence of past experience.
necessary to administer this questionnaire . Informal recommendations, i.e. word-of-
every time we do a perceived quality mouth communication. According to
assessment, since expectations are relatively GroÈnroos (1990, pp. 158-9) and Teboul
stable over short periods of time. Only when (1991, p. 173), this is one of the most
there are indications that customer influential sources of expectations.
expectations have changed will we need to . Formal recommendations, e.g.
update our information with a new battery of recommendations of travel agents or
expectations questionnaires. quality assessments found in certain
Sources of expectations publications, such as consumer
If a company is dealing with customers with magazines.
too high expectations, it will have to act upon . The price to be paid for a service
those expectations in order to take them to an determines in the customer's mind the
appropriate level. Doing that requires an level of quality to be demanded (Teboul,
understanding of the sources of those 1991, p. 173).
expectations. The expectations questionnaire . All elements of the promotional mix
will have to include a section devoted to convey a message to the customer that
unveiling the sources of expectations of the has an influence on expectations (Teboul,
customers. My study tried to shed some light 1991, p. 172).
on the sources of expectations for a customer. . Personal needs determine what is
Zeithaml et al. (1990, pp. 22-3) only mention important for the customer and what is
five sources of expectations: word-of-mouth not (Zeithaml et al., 1990, p. 22).
communications, promotion, price, personal . A good corporate image is a very
needs, and past experience. My research important asset because it shapes
shows that there are many others. The main positively the expectations of the
sources of expectations are: customers (GroÈnroos, 1982, 1984).
27
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

If the company properly manages all the Since the measurement of service quality
controllable sources of expectations could be dependent upon the type of service
mentioned above, it could even compensate examined, future research should also focus
for negative past experiences of the customer, on other settings.
who could decide to give another chance to This piece of research has tried to extend
the company. Therefore, they should not be recent research on service quality and to assist
forgotten. companies with the assessment of service
quality within their practices. Because many
Expectations management of the issues raised are still unsettled, I hope
The company needs to influence the that the research agenda will encourage all
customers' expectations to ensure that they interested researchers to address those issues
are realistic and that the company can fulfil and add to the service quality literature in a
them. Some tools that can be used are constructive manner.
promotional campaigns with suitable
positioning statements, mission statements,
corporate communication campaigns, service Notes
guarantees, consumer education
1 In the services marketing literature, perceptions (P)
programmes, pricing strategies and consistent
are defined as consumers' beliefs concerning the
and excellent service delivery. service received or experienced (Parasuraman et al.,
1985; Brown and Swartz, 1989) or experienced
service (Brown and Swartz, 1989). Expectations (E)
are defined by Parasuraman et al. (1988, p. 17) as
Limitations ``desires or wants of consumers, i.e. what they feel
a service provider should offer rather than would
This piece of research, taken within an airline offer''. Parasuraman et al. (1988) emphasize that
setting, adds to the growing literature which the term ``expectations'' is used differently in the
calls for a re-examination of how to manage service quality literature from in the consumer
service quality. The results of this study satisfaction literature, in that service expectations
cannot be accepted as being completely (E) do not represent predictions about what service
providers ``would'' offer, but rather what they
relevant and applicable to all services, because ``should'' offer.
of the limited sample size, the sampling 2 In the services marketing literature, perceptions (P)
procedure, and, particularly, its focus on are defined as consumers' beliefs concerning the
airline services. Therefore, caution is service received (Parasuraman et al., 1985) or
experienced service (Brown and Swartz, 1989).
necessary in generalising the findings for the
Expectations (E) are defined by Parasuraman et al.
entire airline industry, let alone other service (1988, p. 17) as ``desires or wants of consumers, i.e.
industries. what they feel a service provider should offer rather
Therefore, the conclusions reached here than would offer.'' Parasuraman et al. (1988)
have to be compared and re-examined in emphasize that the term ``expectations'' is used
future studies. differently in the service quality literature from in
the consumer satisfaction literature in that service
expectations (E) do not represent predictions about
what service providers ``would'' offer, but rather
Future research what they ``should'' offer.

My findings support the inclusion of

expectations in the assessment of service References
quality. Future research is needed to
determine the diagnostic or other uses for Bitner, M.J. (1990), ``Evaluating service encounters: the
expectation measurements and to delve more effects of physical surroundings and employee
responses'', Journal of Marketing, Vol. 54, April,
into the sources of expectations. For example, pp. 69-82.
it is conceivable that different market Bolton, R.N. and Drew, J.H. (1991a), ``A longitudinal
segments may have different service analysis of the impact of service changes on
expectations and different sources of customer attitudes'', Journal of Marketing, Vol. 55,
pp. 1-9.
expectations.
Bolton, R.N. and Drew, J.H. (1991b), ``A multistage model
Furthermore, the expectation management of customers' assessments of service quality and
model can be considered only a blueprint, value'', Journal of Consumer Research, Vol. 17,
and needs considerable refinement. March, pp. 375-84.
28
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Brown, S. and Swartz, T. (1989), ``A gap analysis of Teas, K.R. (1993), ``Expectations, performance, evaluation,
professional service quality'', Journal of Marketing, and consumers' perceptions of quality'', Journal of
April, pp. 92-8. Marketing, Vol. 57, October, pp. 18-34.
Cadotte, E.R., Woodruff, R.B. and Jenkins, R.L. (1987), Teboul, J. (1991), Managing Quality Dynamics, Prentice
``Expectations and norms in models of consumer Hall, London.
satisfaction'', Journal of Marketing Research, Tunstall, J. (1991), ``Improving your service capacity'',
Vol. XXIV, pp. 305-14. Management Review, Vol. 19 No. 11, pp. 1-7.
Cina, C. (1990), ``Five steps to service excellence'', Zeithaml, V.A. (1988), ``Consumer perceptions of price,
The Journal of Services Marketing, Vol. 4, Spring, quality and value: a means-end model and synthesis
pp. 39-47. of evidence'', Journal of Marketing, pp. 2-22.
Cronin, J.J. and Taylor, S.A. (1992), ``Measuring service Zeithaml, V.A., Berry, L.L. and Parasuraman, A. (1993),
quality: a reexamination and extension'', Journal of ``The nature and determinants of customer
Marketing, Vol. 56, July, pp. 55-68. expectations of service'', Journal of the Academy of
Daniel, A.L. (1992), ``Overcome the barriers to superior Marketing Science, Vol. 21 No. 1, pp. 1-12.
customer service'', Business and Society Review, Zeithaml, V.A., Parasuraman, A. and Berry, L.L. (1990),
Spring, pp. 53-5. Delivering Quality Service: Balancing Customer
Gale, B.T. (1994), Managing Customer Value, Free Press, Perceptions and Expectations, The Free Press,
New York, NY. New York, NY.
Grapentine, T. (1995), ``Dimensions of an attribute'',
Marketing Research, Vol. 7, No. 3, pp. 19-27.
GroÈnroos, C. (1982), ``A service quality model and its
marketing implications'', European Journal of Further reading
Marketing, Vol. 18 No. 4, pp. 36-44.
GroÈnroos, C. (1984), ``An applied service marketing Berry, L.L., Zeithaml, V.A. and Parasuraman, A. (1985),
theory'', European Journal of Marketing, Vol. 16 ``Quality counts in services, too'', Business Horizons,
No. 17, pp. 30-41. Vol. 28, May-June, pp. 44-52.
GroÈnroos, C. (1990), Service Management and Marketing: Berry, L.L., Zeithaml, V.A. and Parasuraman, A. (1990),
Managing the Moments of Truth in Service ``Five imperatives for improving service quality'',
Competition, Lexington Books, Lexington, MA. Sloan Management Review, Summer, pp. 29-38.
Nunnally, J.C. (1967), Psychometric Theory, McGraw-Hill, Brown, T., Churchill, G. Jr and Peter, J. (1993), ``Improving
New York, NY. the measurement of service quality'', Journal of
Oliver, R.L. (1980), ``A cognitive model of the antecedents Retailing, Vol. 69, Spring, pp. 127-39.
and consequences of satisfaction decisions'', Journal Buzzell, R.D. and Gale, B.T. (1987), The PIMS Principles,
of Marketing Research, Vol. 17, pp. 460-9. Free Press, New York, NY.
Parasuraman, A., Berry, L.L. and Zeithaml, V.A. (1991), Cadotte, E.R., Woodruff, R.B. and Jenkins, R.L. (1983),
``Refinement and reassessment of the SERVQUAL ``Modeling consumer satisfaction processes using
scale'', Journal of Retailing, Vol. 67 No. 4,
experienced based norms'', Journal of Marketing,
pp. 420-50.
October, p. 59.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1985),
Carman, J.M. (1990), ``Consumer perceptions of service
``A conceptual model of service quality and its
quality: an assessment of the SERVQUAL
implications for future research'', Journal of
dimensions'', Journal of Retailing, Vol. 66 No. 1,
Marketing, Vol. 49, Autumn, pp. 41-50.
Spring, pp. 33-55.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1988),
Carmines, E.G., Zeller, R.A. (1979), Reliability and Validity
``SERVQUAL: a multiple-item scale for measuring
Assessments, Sage, Beverly Hills, CA.
consumer perceptions of quality'', Journal of
Cronbach, L.J. (1951), ``Coefficient alpha and the internal
Retailing, Vol. 64, Spring, pp. 12-40.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1994), structure of tests'', Psychometrika, Vol. 16 No. 3,
``Alternative scales for measuring service quality: September.
a comparative assessment based on psychometric Cronbach, L.J. and Furby, L. (1970), ``How we should
and diagnostic criteria'', Journal of Retailing, Vol. 70 measure change or shouldn't we?'', Psychological
No. 3, pp. 201-30. Bulletin, Vol. 74, July, pp. 68-80.
Reichheld, F.F. and Sasser, W.E. Jr (1990), ``Zero Cronin, J.J. and Taylor, S.A. (1994), ``SERVPERF versus
defections: quality comes to services'', Harvard SERVQUAL: reconciling performance-based and
Business Review, Vol. 68, September-October, perceptions-minus-expectations measurement of
pp. 105-11. service quality'', Journal of Marketing, Vol. 58 No. 1,
Robledo, M.A. (1996), ``The importance of customer pp. 125-31.
expectations in measuring service quality'', in Dale, B. and Plunkett, J.J. (Eds) (1990), Managing Quality,
Richards, G. (Ed.), Tourism in Central and Eastern Phillip Allan, Herts.
Europe: Educating for Quality, Tilburg University Foster, K. (1989), ``Selecting the team to meet passenger
Press, Tilburg, pp. 241-6. preferences'', International Journal of Physical
Robledo, M.A. (1998), Marketing Relacional Hotelero: Distribution & Materials Management, Vol. 19
El Camino de la Fidelidad del Cliente, Ediciones No. 3, pp. 5-9.
Profesionales and Empresariales, Madrid. Morrison, S.A. and Winston, C. (1989), ``Enhancing the
Shycon, H.N. (1992), ``Improving customer service: performance of deregulated air transportation
measuring the payoff'', The Journal of Business systems'', Brookings Papers on Economic Activity,
Strategy, January/February, pp. 13-17. pp. 61-123.
29
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Nelms, D.W. (1991), ``Winning their hearts and minds'', Air Robledo, M.A. (1998), ``Una aplicacioÂn del modelo
Transport World, Vol. 28 No. 4, pp. 27-32. SERVQUAL de calidad de servicio al sector de
Ostrowski, P.L., O'Brien, T.V. and Gordon, G.L. (1993), transporte aeÂreo'', Papers de Turisme, Nueva eÂpoca,
``Service quality and customer loyalty in the NuÂmero especial sobre GestioÂn de la Calidad en
commercial airline industry'', Journal of Travel Turismo, pp. 117-35.
Research, Vol. 32 No. 2, Autumn, pp. 16-24. Teas, K.R. (1993), ``Consumer expectations and the
Oxenfeld, A.R. (1974), ``Developing a favorable price- measurement of perceived service quality'', Journal
quality image'', Journal of Retailing, Winter, of Professional Services Marketing, Vol. 8 No. 2,
pp. 8-14. pp. 33-54.
Parasuraman, A., Berry, L.L. and Zeithaml, V.A. (1990),
Teas, K.R. (1994), ``Expectations as a comparison standard
An Empirical Examination of Relationships in an
in measuring service quality: an assessment of a
Extended Service Quality Model, Marketing Science
reassessment'', Journal of Marketing, Vol. 58,
Institute, Cambridge, MA.
January, pp. 132-9.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1986),
``SERVQUAL: a multiple-item scale for measuring Zeithaml, V.A., Berry, L.L. and Parasuraman, A. (1988),
customer perceptions of service quality'', Informe ``Communication and control processes in the
nuÂm., No. 86-108, August, Marketing Science delivery of service'', Journal of Marketing, Vol. 52,
Institute, Cambridge, MA. April, pp. 35-48.
Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1994a), Zeithaml, V.A., Berry, L.L. and Parasuraman, A. (1991),
``Reassessment of expectations as a comparison ``The nature and determinants of customer
standard on measuring service quality: implications expectations of service'', Papel de Trabajo,
for further research'', Journal of Marketing, Vol. 58 No. 91-113, Marketing Science Institute,
No. 1, pp. 111-24. Cambridge, MA.

Appendix I. Questionnaire design

These questionnaires were designed taking the respondent was provided with a
the groundbreaking model developed by seven-point Likert scale. Negatively
Parasuraman et al. (1985) and its subsequent worded items were not used. Only the positive
tool for measuring service quality format was used because other studies
SERVQUAL as the starting point. (Parasuraman et al., 1991) revealed that the
The original version of SERVQUAL negatively worded items showed standard
(Parasuraman et al., 1988) consists of 22 deviations consistently higher than the
items. The revised versions for airline services positively worded ones. The wider
resulted in the expansion of the original 22 variation for the negatively worded items
items to 26 ± including minor wording implies that respondents may be confused by
changes in a few items ± plus three final those items.
contrast questions (overall satisfaction, degree Direct measures of the importance of the
of fulfilment of expectations and comparison five different dimensions of service quality
with ideal company). The resulting 29 items identified by Parasuraman et al. were included
shown below were common to all in the questionnaires. The relative importance
six questionnaires. This is something even of the five dimensions was ascertained by
recommended by Parasuraman et al. (1991), asking customers to allocate a total of 100
who say: ``context-specific items can be used points across the dimensions according to
to supplement SERVQUAL'' (see Table AI). how important they considered each to be.
Therefore, the questionnaires are This point-allocation question listed
structured in a set of statements in descriptive definitions of the five dimensions
questionnaire format. For each statement, without naming them.
30
 Measuring and managing service quality Managing Service Quality
Marco Antonio Robledo Volume 11 . Number 1 . 2001 . 22±31

Table AI Questionnaire items Appendix II. Cronbach's results from

1. Modern appeal of the aircraft eliminating an item each time (results of
2. Visual attractiveness of aircraft SERVPEX's questionnaire)
3. Appearance of the employees
4. Visual attractiveness of the materials related to the airline's services Table AII
(tickets, in-flight magazines, security instructions)
Eliminated item Alpha
5. Fulfilment of promises
6. Employees' interest in solving passenger problems 1 0.9584
7. Flight's punctuality 2 0.9585
8. Assurance that the luggage will reach the destination 3 0.9573
9. Handling of the luggage 4 0.9588
10. Speed of the service offered 5 0.9579
11. Willingness of employees to attend passengers 6 0.9576
12. Willingness of the employees to respond to the passengers' questions 7 0.9591
13. Level of communication to the passengers regarding unusual 8 0.9581
circumstances and unexpected occurrences on flight 9 0.9588
14. Degree of trust transmitted to the passengers by employees 10 0.9575
15. Kindness of employees 11 0.957
16. Level of knowledge of employees in responding to questions of the 12 0.9572
clients 13 0.9579
17. Communication of security procedures 14 0.957
18. Clear identification of the emergency equipment 15 0.9568
19. Individualized attention provided to passengers 16 0.9573
20. Convenience of flight schedules 17 0.9571
21. Cordiality and kindness of employees 18 0.9573
22. Searching for the best for the passenger 19 0.9568
23. Understanding of the specific needs of the passengers 20 0.9582
24. Utilization of an understandable terminology whenever attending the 21 0.9566
passenger 22 0.957
26. Comfort of airplanes (seats, corridors, hand luggage set) 23 0.957
27. Overall level of satisfaction 24 0.957
28. Assessment of the service received in relation to expectations 25 0.959
29. Comparison with ideal company 26 0.9587

31