15
Perlanika J. Soc. Sci. & Hum. 9(2): 71 - 85 (2001) ISSN: 0128-7702 e Universili Pulra Malaysia Press Measuring Customers' Perceived Service Quality in Hotel Industry SAMSINAR MD SIDIN', MD. ZABID ABDUL RASHID' & RAJA ANIS RAHYUWATI RAJA ZAINAL' I Depmtment oj l\1anagement and A1.arketing Faculty of Economics and Management 2Graduate School oj Management JFaculty oJ Economics and Management Univemti Putra Malaysia 43400 UPM, Serdang, Selangor Keywords: Perceived service quality, hotel, SERVQUAL ABSTRAK Kajian ini cuba mengkaji kualiti perkhidmatan tertanggap dalam industri hotel. Ia bertujuan menyiasat apakah yang dikatakan kualiti perkhidmatan sebagaimana yang terdapat dalam industri hotel dan mengkaji faktor-faktor yang mempengaruhi tanggapan ini seperti perkhidmatan peribadi inovasi teknologi dan kualiti makanan yang dihidangkan. Kaedah yang digunakan untuk mengumpul sumber kajian didapati dari SERVQUAL, sam kaedah yang popular untuk mengukur kualiti perkhidmatan tenanggap. Kaedah deskriptif dan inferens juga digunakan untuk menguji dan menganalisis hipotesis. Data dianalisis menggunakan pakej SPSS. Dapatan kajian menunjukkan bahawa secara amnya pelanggan tidak berpuas hati dengan kualiti perkhidmatan yang ditawarkan oleh pengurusan hotel. Daripada kajian ini juga didapati bahawa perkhidmatan peribadi, inovasi teknologi dan kualiti makanan yang dihidangkan adalah penting untuk memperbaiki pandangan pelanggan terhadap kualiti perkhidmatan. Oleh yang demikian, pihak hotel seharusnya cuba memenuhi atau mengatasi jangkaan pelanggan llntuk memastikan pelanggan berpuas hati. Adalah penting bagi pihak hotel memahami jangkaan untuk menghasilkan perkhidmatan yang baik. Jika kualiti tertanggap yang serupa atau melangkau perkhidmatan tertanggap, pelanggan menganggap perkhidmatan terse but berkuaLiti. ABSTRACT This research attempts to study customer's perceived se",ice quality in the hotel industry. This paper aims to discover what customers think of the quality of service as can be found in the hotel industry by looking into factors influential on this perception such as personal service, technological innovations and quality of food served. The method employed to gather the research resources was adopted from SERVQUAL which is a popular method in measuring perceived service quality. The descriptive and inferential methods were also used in testing and analysing the hypotheses. Data were analysed by using the SPSS package. The research findings indicated that generally, customers were dissatisfied with the service quality provided by the hotel management. From the research, it was also discovered that personal services technology innovation and quality of food served were vital in improving customers' outlook on the service quality. Therefore, the hotelier should try to meet or exceed the customers' expectations, in order to ensure the customers are satisfied. It is very important for the hotelier to take an effort in comprehending and understanding customers' expectations in order to deliver good service, in which if the perceived service equal or exceeded the expected service, they perceived that there is a quality in the service.

Measuring Customers' Perceived Service Quality in Hotel

  • Upload
    others

  • View
    4

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Measuring Customers' Perceived Service Quality in Hotel

Perlanika J. Soc. Sci. & Hum. 9(2): 71 - 85 (2001) ISSN: 0128-7702e Universili Pulra Malaysia Press

Measuring Customers' Perceived ServiceQuality in Hotel Industry

SAMSINAR MD SIDIN', MD. ZABID ABDUL RASHID' &RAJA ANIS RAHYUWATI RAJA ZAINAL'

IDepmtment oj l\1anagement and A1.arketingFaculty of Economics and Management

2Graduate School oj ManagementJFaculty oJ Economics and Management

Univemti Putra Malaysia43400 UPM, Serdang, Selangor

Keywords: Perceived service quality, hotel, SERVQUAL

ABSTRAK

Kajian ini cuba mengkaji kualiti perkhidmatan tertanggap dalam industri hotel. Ia bertujuanmenyiasat apakah yang dikatakan kualiti perkhidmatan sebagaimana yang terdapat dalamindustri hotel dan mengkaji faktor-faktor yang mempengaruhi tanggapan ini seperti perkhidmatanperibadi inovasi teknologi dan kualiti makanan yang dihidangkan. Kaedah yang digunakanuntuk mengumpul sumber kajian didapati dari SERVQUAL, sam kaedah yang popular untukmengukur kualiti perkhidmatan tenanggap. Kaedah deskriptif dan inferens juga digunakanuntuk menguji dan menganalisis hipotesis. Data dianalisis menggunakan pakej SPSS. Dapatankajian menunjukkan bahawa secara amnya pelanggan tidak berpuas hati dengan kualitiperkhidmatan yang ditawarkan oleh pengurusan hotel. Daripada kajian ini juga didapati bahawaperkhidmatan peribadi, inovasi teknologi dan kualiti makanan yang dihidangkan adalah pentinguntuk memperbaiki pandangan pelanggan terhadap kualiti perkhidmatan. Oleh yang demikian,pihak hotel seharusnya cuba memenuhi atau mengatasi jangkaan pelanggan llntuk memastikanpelanggan berpuas hati. Adalah penting bagi pihak hotel memahami jangkaan untuk menghasilkanperkhidmatan yang baik. Jika kualiti tertanggap yang serupa atau melangkau perkhidmatantertanggap, pelanggan menganggap perkhidmatan tersebut berkuaLiti.

ABSTRACT

This research attempts to study customer's perceived se",ice quality in the hotel industry. Thispaper aims to discover what customers think of the quality of service as can be found in the hotelindustry by looking into factors influential on this perception such as personal service, technologicalinnovations and quality of food served. The method employed to gather the research resourceswas adopted from SERVQUAL which is a popular method in measuring perceived service quality.The descriptive and inferential methods were also used in testing and analysing the hypotheses.Data were analysed by using the SPSS package. The research findings indicated that generally,customers were dissatisfied with the service quality provided by the hotel management. From theresearch, it was also discovered that personal services technology innovation and quality of foodserved were vital in improving customers' outlook on the service quality. Therefore, the hoteliershould try to meet or exceed the customers' expectations, in order to ensure the customers aresatisfied. It is very important for the hotelier to take an effort in comprehending andunderstanding customers' expectations in order to deliver good service, in which if the perceivedservice equal or exceeded the expected service, they perceived that there is a quality in theservice.

Page 2: Measuring Customers' Perceived Service Quality in Hotel

S~msjnar Md. Sidin, Md. Zabid Abdul Rashid & Raja Allis Rah}'Uwatj Raja Zainal

INTRODUCTION

Although services have been increasinglyimportant for the last decades, the interest inservice quality is of recent origin. This is becauseservice quality is a matter of knowing yourcustomers, designing services to meet customers'needs, and finally managing the serviceproduction and delivery process to customers'satisfaction. Studies have shown that customersare often willing to pay higher prices for whatthey perceive to be of better quality. The StrategicPlanning Institute of Cambridge has found thatmarket share, return on investment, and assetturnover are all highly linked to the perceivedquality of the company's goods and services.The}' found that, the most important single factorafTecting a business unit's performance is thequality of its goods and services, relative to thoseof competitors.

Quality has become the 1990's byword ofbusinesspeople, government officials, andbusiness and economics scholar, consequently,consumers, employees. management, and alsoboards emphasise on qualit},. \t\'hile quality andquality control measures have long existed fortangible goods, few such measures havetraditionally existed for sel\lices. This occursbecause quality is difficult to define. describe.and measure in services. In essence. quality isdetermined by imprecise individual factors:perceptions, expectations and experiences ofcustomers and the hotelier itself. Robert E. Allen,Chairman of the Board of AT & T said inFortune magazine in 1988, "Quality does all. Itsaves all. It sells. It satisfies." \Vhile nearlyeveryone recognises the pervasive impact ofquality, at the same time. everyone seems to behaving difficulty in grasping its many dimensions.

Deueiopment oj Hotel Indust,y in Malaysia

Malaysia's hotel room supply has grown at arapid rate over the years. Since 1981, 28,432rooms were available. However. the end of 1998the number of hotel rooms exceeded the 100.000mark (107,791). Viewing the trends over theyears, the number of hotel rooms in the countryLOok roughly 10 years LO double from 28,432rooms in 1981 to 55,866 rooms in 1992. Due LOthe rapid rate of growth. it took six years for thenumber of hotel rooms to double up toapproximately 110,000 rooms in 1998. In recenttimes, during the three year period, beginningfrom 1991 to the end 1993. prior to Visil Malaysia

Year '94, a total of 15,973 rooms came in stream,giving a total of61,005 rooms as at end 1993. In1994. as the country set to reap the success ofthe Visit Malaysia Year '94 promotional campaign.the industry added 4,902 leuabJe rooms towelcome her visitors. At the end of 1994, thenumber of lenable rooms stood at 65,907 roomsor an increase of 43.8% from the previous year.

International events have been a drivingforce in the growth of rooms. domestic tourismhave played an equally important part towardsthe overall growth. There is an increasing interestin building more. and with the provision oflarger budget and mid-priced hotels in thecoming years. Domestic tourism fueled by abuoyant economy, and the completion of the1 orth-5outh Expressway. have to a large extentimproved the average hotel occupancy rates by4.0% to 65.5% in the 1995 period as comparedto the previous year. However, due to theeconomic slow down, the average occupancyrate of hotels in Malaysia decreased by 8.1 %from 58.0% in 1997 to 49.9% in 1998. (AnnualTourism Statistical Report 1998).

The Probkm

It is clear that the hotel market is highlycompetitive. and like any other industry thatmarkets a product in a highly competitive market,competing for the consumer's disposable income,it has to be aware of the quality issues. This isespecially important to the hotel situation inwhich there is less obvious competition on price,but centers more on issues of facilities, image.service and the quality of that provision. Hotelshave therefore. improved their service quality inorder to stay competitive.

The management of quality is ,a key issue inthe management of any hotel property. It hasbeen described as one of three key areasunderpinning corporate success in the hospitalityfield. The profitability of the operation is seenas being supported by the quality of hospitalityservices, management and the market. Hotelcompanies are increasingly aware of theimportance of quality and clearly use it in theiradvertisement and promotion to customers andin the standards of performance set for theiremployees. There is still, however, somemisunderstanding in perceiving the servicequality between the hotel companies andcustomers, in which research has shown thatmany service organisations develop their own

72 PenanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 3: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel InduStry

Review of Pasl Research

According to research done by Parasuraman, A.,Berry, L.L., and Zeithaml, VA (1985,1988), akey to providing superior service is understandingand responding to a customer's expectations. Inwhich, the customers expect service companiesto do what they are supposed to do. They expectfundamentals, not fanciness; performance, notempty promises. In short. cllstomers are payinggood money. and the company should providegood seIVice in exchange. At the point of servicecreation and delivery. all the resources andknowledge of the provider of the service mustcome together in exactly the right way to ensurean excellent service encounter for the customer.It is at this point that the customers comparetheir expectations with actual service deliverysystem performance and ultimately an-ive at theirperceptions of service quality. According to

Collier (1987), by managing these moments oftruth well, we can actually manage the customer'sperception of service quality.

Parasuraman, Zeithaml, and Berry (1990)suggested that customers' service expectationshave two levels: desired and adequate. Thedesired service level is the service the CUSLOmerhopes to receive. It is a blend of what thecustomer believes "can be" and "should be".While the adequate service level is that whichthe customers find acceptable. It is in part basedon the customer's assessment of ",hat the service"will be", that is, the customer's "predictedservice", both of which can change over timeand from one service encounter to the next forthe same cllstomer.

From their findings (1991), The Zone ofTolerance is developed; a zone that separatesdesired service level from the adequate servicelevel, as shown in Fig. 1.

This zone is likely to differ for the outcomedimension of reliability and the process

Zone of Tolerance

Expectations

Fig. ISeroice level expect-ations

High

Desired

1Adequate

1Low

Objectives oj Study

The general objectives of this research are tomeasure the customers' perceptions on theservice quality they received from the hotel andto verify their satisfaction with the servicerendered. The specific objectives are:1. To determine aspects of services which are

important to customers.2. To analyse whether customers evaluation of

services by comparing service received(perceptions) with service desired(expectations) .

3. To fill the gaps in the knowledge of theorganisation regarding their own perceptionsof what the customer wants and what thecustomers actually perceived on the servicequality they received or expect to receive.

perceptions of what customers want, which oftendiffer from what the customers really want.

Since these intangible aspects are not easyto measure, it is difficult to know whether thecustomer is satisfied. Current marketingtechniques provide knowledge about thecustomer, his needs and expectations, but theprofessionals, who in this study are the hoteloperators, remain relatively helpless in perceivingwhat it is that the customer wants or does notwant and what he is ready to accept. In addition,hotels quite properly proclaimed themselves asleaders in the service industry but has beenlacking in training applied to the actualtechniques of service itself. They trained inkitchen crafts, computer skills, accountingprocedures, sales training, managementdevelopment - the list is endless - but there is anobvious lack of documentation on the fusing ofguest contact skills and senrice techniques, thesocial needs of the hotel guest, his sensitivitythreshold and the use of body language andspeech skills in service interactions. The hoteloperator must, therefore, make a start on trainingprograms, which are now usually referred to asfront line staff in service and performanceimprovements.

In conclusion, an analysis pertaining thecustomer's perception on the hotel service qualityand their satisfaction is needed in order toestablish strategies and to ascertain steps to betaken in maximising consumer's satisfactionwhich, hopefully, can form a positive image ofthe hotel industry in Malaysia.

PenanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001 73

Page 4: Measuring Customers' Perceived Service Quality in Hotel

Samsinar Md. Sidin, Md. Zabid Abdul Rashid & Raja Anis Rahyuwati Raja Zainal

dimensions of tangibles, responsiveness,assurance, and empathy. Hence, Fig. 2 reinforcesgraphically this research earlier conclusion thatthe opportunity for firms to exceed customerexpectations is greater with the processdimensions than with the outcome dimension.

Outcome ~1---------1•••---i

Process ~1-----l•••••I-----1Low High

Expectations

Fig. 2Tolerance zones fOT outcome and process

dimensions of service

The customer's quality perception dependson the degree to which quality expectations areconfirmed or disconfirmed by experience. Basedon Gronroos's perceived service quality model(1982), the quality of a service, as perceived bythe customer, is the result of a comparisonbetween the expectations of the customer andhis real-life experiences. If the experiencesexceed the expectations, the perceived servicequality is positive. If the experiences do notreach the level of expectations, the perceivedquality is low (Fig. 3).

In detail, conceptually, this confirmation/disconfirmation (gap) concept has an important

impact on people's thinking about quality. Itimplies that quality is not an objectivephenomenon that can be engineeredbeforehand. But with proper preparation priorto the service encounter, good quality may beachieved, We already know that customersperceive quality in a subjective manner.Depending on the level of expectations, thesame level of quality, as measured in someobjective sense, will be perceived in a differentway. Thus, what perceived as good quality forone person may be less acceptable for another.If expectations are raised too high - owing to anadvertising campaign, for example - the qualityof a given service may be disappointing, if onlybecause the customers had unrealisticexpectations.

As stated by Gronroos (1993), regardless ofthe content of the interface, what happens inthe simultaneous part of the production andconsumption processes is always of criticalimportance to the customer's perception of theservice. If the service encounter is perceivednegatively, the customer is often less than pleased,although most of the efforts to produce theservice are beyond the line of visibility from thecustomer's point of view and may have beenproperly taken care of. Hence, how customersperceive the service encounter - or in otherwords, how customers perceive the quality of theservice delivered in the service encounter - has acrucial marketing impact. According to theirvery nature, services are processes in which the

ExpectedQuality

1Market Communication

• Image• Word-of-mouth• Customer Needs

Total Perceived Quality

Outcome/Technical Quality:What

Fig. 3The perceived seroice qualit}1 model

ExperiencedQuality

tB

tProcess/Functional QualityHow

74 PertanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 5: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel Industry

customer in most cases gets involved. If theprocesses are perceived to function well, this hasa positive impact on the perception of the qualityof the ser....ice, and vice-versa.

Swan and Combs (1976) suggested that theperceived performance of a service can be dividedinto two sub-processes; namely, instrumentalperfonnance and expressive perJonnance. Theinstrumental performance of a service is thetechnical result of a service production process.It is, so to speak, what the customer is left with,when the production process is finished. Whereas, expressive performance is related to"psychological" level of performance, in a servicecontext, it would be related to the customer­service producer interactions, i.e., to the contactsthe consumer has with various resources andactivities of the service firm during the serviceproduction process when the technical outcome,the instrumental performance, is created. As anexample, we may think of a hotel customer'scontacts with the employees of the hotel, physicaland technical resources, such as in-checkingdesks, the hotel itself, rooms, food available andthe customer's contact with other customers.The customer's interactions with such humanand non-human resources during their stay willcertainly have an effect on his or her evaluationsof the service he or she perceives he has received.

There is evidence, however, that customersdo not evaluate the quality of their stay in ahotel on the tangible physical aspecLS of theprovision. According to Nightingale (1985), thecustomers are more likely to consider such factorsas the availability of a service; the smoothness orspeed of the response to a request; ease of use;comfortable, pleasant and relaxing surroundingsand value for money.

Sasser, Olsen, and Wyckoff (1978) haddiscussed three different dimensions of serviceperformance: level of material, facilities, andpersonnel. Implied in this trichotomy is thenotion that service quality involves more thanoutcome, in which it also includes the mannerin which the service is delivered.

In a broad sense, "personalised service" refersto any behaviours occurring in the interactionintended to contribute to the individualisationof the customer (Suprenant and Solomon 1987).This definition highlights the centrality of theservice encounter to the customer's ultimatesatisfaction with a particular service. The serviceencounter is the dyadic interaction between a

customer and a service provider. The nature ofthis interaction has been recognised to be acritical determinant of satisfaction with theservice. Though the service provider or in thiscase is known as front-liner (i.e., receptionist,housekeeping staff, food and beverage staff) istypically the very last link in the chain ofproduction and is, ironically, often the least­valued member of the service organisation interms of pay and status, this person is investedwith enormous responsibility for conveying the"personality" of the service offering to thecustomer.

Michael Nightingale (1986) has looked intoa few studies on the quality of service in thevarious organisations in the United IGngdom atthe end of 1970's. The study came up with a fewdifferent perceptions benveen the consumer andsupplier or service producer, between suppliersand between the departments in theorganisations. From this survey, it has help tosee the perception of the customers, themanagement, and the emplo),ees on thecharacteristics of the services that are needed.These perceptions have been put into theirrespective ranking. In seeing the relativeimportance of different service transactions inthe hotel, it was found out that there are a fewaspects whereby the customers and themanagement's perception in the use of servicetransaction clashes. Among them are:

I. The Use of Telephone2. Enjoying their drinks3. Making reservation4. Working in their room

The services above are seen as among theimportant factors to the consumers but to themanagement, it has no importance. And it wasalso found out that the management perceptionspertaining the selvice transaction like breakfastand snacks are not important but to the customerit does.

The perception on the service transaction isimportant because, when the management andcustomer perceptions are almost the same, thecustomers are satisfied and this gives thecustomer a perception that the services of anorganisation is excellent. Table 1 shows a rankingon the importance of perception on the servicetransactions by the management and customers.Apart from that, this survey has found out about

PertanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001 75

Page 6: Measuring Customers' Perceived Service Quality in Hotel

Samsinar Md. Sidin, Md. Zabid Abdul Rashid & ~a Anis Rahytn..... ti Raja Zainal

the perception needed pertaining the charac­teristic of the services by the customer and themanagement. There are distinctive differencesin the customers' and management's perceptionswhereby, the customers assessed that theavailability, comfort, and efficiency are neededin the hotel sen'ices but the management didnot acknowledge the customers needs. Moreoverthe employee perceived that the accreditationand friendliness are not important to thecustomers. Table 2 shows the differences ofperceptions by the customers and themanagement regarding the service quality.

Theoretical Framework

The theoretical framework of this study is basedon a model developed by Parasuraman A.,Zeithaml VA. • Berry L.L. (1990). This modelidentifies five gaps or points where serviceproduction might falter. This study will onlyconcentrate on Gap 5, which is customer'sperception of service quality. Since the differencebetween customers expectation of service andthe actual performance by the organisationrepresent the service quality perceived by thecustomer, a conceptual model is developed forfurther explanations. (Fig. 4).

TABLE]The differences of perceptions on the imporlance of service

transactions by the customers and management

Customers

A Night SlayingSnack TreatsBreakfa ,tThe U"e of the TelephoneBatheDinnerMaking ReservationDoing Business Affairs in TheirRooms

Management

BatheSwimmingInformation ReceivableMoviesA Night StayingArrival in HotelDinnerBreakfastBusiness Affairs Discussions

Notes:Source:

The categories above have been arranged according to their degree of importance.Nightingale M., The Practice of HospitalilY Management 11: Profitability In A ChangingEnvironment, 1986.

TABLE 2The ditferem perceptions pertaining the service quality by the customers,

management and employees

Customers

AvailabilityEfficiencyComfortPeaceful and RelaxingValue For MoneyAdequate FacilitiesSpacious

Management

Value For MoneyEfficiencyPeaceful and RelaxingComfortGuest Controls

Employees

Recognition By The Customers

FriendlinessEfficiencyCommunications

Sources: Nightingale M., The Practice of Hospilality Management 11: Profitability In A ChangingEnvironment, 1986.

76 PertanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 7: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel Industry

CONSUMER

Vlord of MouthCommunications

Personal Needs Past Experiences

MARKETER

QUALITY ....... _

Perceived Service

i

HI

PersonalService

........• Sen-ice deliveryor Actual

Performance

H2

TechnologicalInnovations

~ ,

H3

Food &Beverages

Fig. 4Conceptual model of seruice quality

Three hypotheses are derived to test this modeL(Hypolhesis 1 (Hl):

• Customers' perception of personal seIV'icerendered is affecled by SERVQUAL'sdimensions tangible, reliability,responsiveness, assurance, and empathy.

• Hypothesis 2 (H2):The higher the technological innovationsprovided, the better the customers'evaluation of seIV'ice quality.

• Hypolhesis 3 (H3):The higher the qualily of food available, thebetter the customers' perceptions of seIV'icequality.In addition to these hypotheses, a fourth

hypothesis was developed to test the differentSERVQUAL dimensions.

• Hypothesis 4 (H4):There are differences in SERVQUALdimensions perceived by the customers inassessing overall seIV'ice quality

METHODOLOGY

This study was carried out through a sUIV'eydone in several hotels in Kuala Lumpur. Theprimary data collecterl consist of the customer'sperception of the service quality rendered bythe particular hotel that they are staying in. Theselected respondents consisted ofhmel customersin a few chosen five star hotels in Kuala Lumpur.

Tn this study, a sample of 150 hotel customerswere chosen in order to determine thecustomers' perceived service quality in hotelindustry. The sample framework was inclusive ofall the hotel customers in a few selected five star

Pertanikaj. Soc, Sci. & Hum. Vol. 9 No.2 2001 77

Page 8: Measuring Customers' Perceived Service Quality in Hotel

Samsinar Md. Sidin, Md. Zabid Abdul Rashid & Raja Anis Rahyuwati Raja Zainal

hmels in Kuala Lumpur. Malaysia. The samplingunit was individual (foreign and local). who wasgiven a set of questionnaires.SERVQUAL (Parasuraman, Berry and Zeithaml(1988» measurement was used to measureconsumer perceptions of service quality. Thesemeasurements were subjected to reliability testsand were found to be highly reliable (Cronbachalpha ranging from .78 to ,90)

DISCUSSION

Table 3 revealed that the majority of therespondents were female (57,1 %), About 77percent of the respondents were Malaysian andonly 23% of them were non-Malaysian.

About 34.1 percent of the respondents wereChristians, followed by Muslim (28,6%), Buddhist(15,9%), Hindu (14,3%) and other religions(7.1 %). In telms of their educational level, about35.7 percent of the respondents were degreeholders, 27.0 percent were diploma holders, 18.3percent had Masters and 6.3 percent had Ph. D..Most of the respondents earned about RM16,OOO-36.000 per annum. There were 72 respondentswho are not married and 54 who were married.

Sources of Information Used in Choice of Holel

Table 4 shows that, when asked to rank differentsources of information as to their importance inchoice of hotel, the majority of the respondentsindicated that members of immediate familywere the most important sources (mean := 4.103)but friends (72,3%) (mean = 3,984) and businessassociate (69,9%) (mean = 3,825) were alsoimportant to the respondents when selecting ahotel. Table 2 also indicates that travel agency(mean := 3.238) and advertisements in TV,newspaper and magazine (mean := 3.341) wereused frequently in collecting various aspects ofinformation on hotel choices.

Hotel Location Preferences

Table 5 indicated the hotel location preferencesamong respondents. Most of the respondentspreferred down town/central location (69.8%Lfollowed by recreation sites (58.7%), suburbanlocation (49.2%), just out-side town location(46,8%), and rural location (32.6%)

Number oj Days and Money Spent

Table 6 and Table 7 indicated about 76,2 percentof the respondents spent 2 - 5 days at the hotel

TABLE 3Demographic characteristics of the respondents

Demographics Number PercentageVariable (%)

GenderMale 54 42,9Female 72 57,1

Total 126 100,0

Age< 25years 40 31.726 - 35 51 40,536 - 45 33 26.246 - 55 2 1.6

Total 126 100,0

NationalityMalaysian 97 no

10n Malaysian 9 23,0

Total 126 100,0

EthnicsMuslim 36 28,6Hindu 18 14,3Buddha 20 15,9Christians 43 34,1Others 9 7,1

Total 126 100,0

Education LevelDiploma 3427.0Degree 45 35,7Masters 23 18.3Ph,D 8 6,3Others 16 12,7

Total 126 100,0

Income<RMI6,OOO 5 4,0RMI6,OOO - 24,000 33 26,2RM24,OOI - 36,000 32 25.4RM36,OOI - 48,000 20 15,9RM48,OOI - 64,000 20 15,9>RM64,OOI 16 12.7

Total 126 100,0

StatusSingle 72 57,JMarried 54 42,9

Total 126 100,0

Source: Survey

78 PerL."\nikaj. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 9: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel Industry

TABLE 4The import.'lnce of sources of information (n=126)

3Sources

Business AssociatesFriendsMember of Immediate FamilyTravel AgencyTV, Newspaper, Magazine

Sourct: SUr'o'ey1 = Not at all important3 = Neutral5 = Very Important

11.9%2.44.8

20.715.8

18.3%25.417.538.139.7

5 Mean

69.9% 3.82572.3 3.98477.8 4.10341.2 3.23844.4 3.341

Location

TABLE 5Hotel location preferences (n=126)

3 5 Mean

DownlOwn/CentralJust Outside CityRural LocationRecreation SiteSuburban Location

13.5%15.133.314.221.5

16.7%38.134.127.029.4

69.8%46.832.658.749.2

3.9053.3332.8813.6353.317

TABLE 6Number of days spent (0=126)

Souru: Survey1 = Least Preferred3 = Neutral5 = Most Preferred

Frequency Percentage

and most of them spent about RMI,OOI - 3,000(46.8%) while staying there.

Purpose oj Staying

About 65.9% of the respondents stayed for thepurpose of pleasure, while 20.6% and 12.7%came for convention and business purpose(Table 8).

2-5 days6 - 1011 - 1516 - 20> 21

Souru: Survey

9627

2I

TABLE 7Money spent (n=126)

76221.4

1.60.8

Reasons in ChooJing the Particular Hotel Comparedto Other Hotels

Deciding on the reasons in choosing theparticular hotel, most of the respondents camebecause of their past experiences with that hotel(33.3%), 30.2 percent came because the hotel\vas convenient for them, followed by suggestions

TABLE 8Purpose of staying (n=126)

<RMI,OOORMI,OOI - 3,000RM3,OOI - 5,000>RM5,OOI

Souru: Survey

Frequency

4459194

Percentage

34.946.815.1

3.2

PleasureConventionBusinessOthers

Scmrct: Survey

Frequency

832616

1

Percentage

65.920.612.70.8

PertanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001 79

Page 10: Measuring Customers' Perceived Service Quality in Hotel

Samsinar Md. Sidin, Md. Zabid Abdul Rashid & Raja Anis Rahyuwati Raja Zainal

from friends (22.2%), unintentional (9.5%) andlastly for other reasons (4.8%) such as, it wasfixed by their company. and seminar they haveattended ,....hich required them to stay at thatparticular hotel (Table 9).

percent of the respondents will stay at the samehotel in the future if they were to be in thatarea. Only 7.9 percent will not stay at the samehotel in the future, while 1.6 percent wereundecided whether to stay or not in the future.

TABLE 9Factors in choosing hotel (n=126)

Frequency Percentage

Past ExperienceConvenienceSuggestions from FriendsUnintentionalOthers

Source: Survey

42382812

6

33.330.222.2

9.54.8

Hypotheses 1

HI: Customers' perception of personal servicerendered is affected by SERVQUAL's 5Dimensions - Assurance, Empathy, Reliabiliry,Responsiveness, and Tangibility.Y = Personal service rendered (PS)XI =Assurance (SQLASS)X2 = Empathy (SQLEMP)X3 = Reliability (SQLREL)X4 = Responsiveness (SQLRESP)X5 =Tangibility (SQLTGB)

TABLE 11Come back in the future (n=) 26)

TABLE 10Rate on Overall Basis (n=126)

Stay at the Same Hotel in the Future

Earlier on, it was reported that respondentscame to the hotel because of their pastexperience and it might be because of the goodservice they received. Table 11 shows that 90.5

Rating of Hotel on Overall Basis

The respondents did not at all rate the hotel asfair or poor. Instead most of the respondentsrate the hotel on overall basis as good (71.4%).Table 10 also indicated that 14.3 percent of therespondents rate the hotel they stayed asexcellent.

Hypotheses 2

HI! = Technological innovations provided ispositively correlated with customers' evaluationof sen'ice quality.

From Table 12, the analysis of correlationshowed that all the 5 dimensions have asignificant correlation towards technologicalinnovations. This indicates that SERVQUAL hasa very significant correlation with technologicalinnovations. Thus H:! was supported.

The inclusion of the five independentvariables correlates well with personal servicerendered by the employee (R = 0.687). The R'of 0.472 indicates that about 47 percent of thevariation in personal service rendered for thissample of 126 respondents is explained by theabove independent variables. The regressionequation above indicates that personal serviceare positively related to Xl' ~, X3• and X'I andnegatively related to Xs' This equation isstatistically significant and there does appear to

be an association between the dependent variable(personal service) and the independent variable.Thus, HI was supported. T ~value shows thatonly X, (Assurance) is significant with a value of5.270 at the 0.05 level of statistical significance.

Y= 62.153 + 1.027X, - 0.024X, +0.269X, +0.478X,+0.134X,

From tlle regression analysis conducted, thefollowing regression equation is the result of theanalysis,

14.371.414.3

90.57.91.6

Percentage

Percentage

189018

114102

Frequency

Frequency

YesNo.Undecided

Source: SUl\fey

AverageGoodExcellent

Scurce: Sun'er

80 Pert:mikaj. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 11: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel IndusU)'

TABLE 12Pearson's correlation coefficients

between SERVQUAL dimensions andtechnological innovations

TABLE 14Correlation coefficients between technological

innovations and quality of food availablewith SERVQUAL score

Technological Innovations Technological Quality of FoodInnovation

Dimensions p

SOliTU: Sun'ey• Significant Ul p <0.05

TangibleReliabilityResponsivenessAssuranceEmpathY

0.60380.45730.21150.46970.5720

0.00i}'0.000'0.017'0.00i}'0.000'

Service Quality 0.6245P = 0.000'

0.5781P = 0.000'

SOUTce: Survey~Significant at p <0.05

TABLE 13Pearson's correlation coefficients between

SERVQUAL dimensions and qualityof food available

Table 14 indicates that SRVQL has thecorrelation coefficient of 0.6245. Thus, it showsthat service quality has a strong positivecorrelation with hotel's technological innova­tions. The table also indicates that SERVQUALhas the correlation coefficient ofO.S78, showingthat SERVQUAL has a strong positive correlationwith quality of food available at the hotel.

Hypotheses 3

H:l = Quality of food available is positivelycorrelated with customers' perceptions of servicequality.

From Table 13, the analysis of correlationshowed that all the 5 dimensions have a signifi·cant correlation towards the quality of foodavailable at the hotel. This indicates that SRVQLhas a very significant correlation with the qualityof food available. Thus H)l was supported.

Hypotheses 4

H4 = There are differences in SERVQUALdimensions perceived by the cuswmers inassessing overall service quality.

Table 15 indicates the ANOVA in deter­mining the SERVQUAL dimensions perceivedby the cuswmers in assessing overall servicequality. Significant ANOVA results wereinvestigated further using Duncan's multiplerange test to identify significant differences acrossthe Overall Q categories. The results of theseanalyses are summarised in Table 15 and Table16.

Table 15 indicates that the analysis ofvariance in determining the differences ofSERVQUAL dimensions perceived to beimportant by the customers in assessing overallservice quality. The F-value for all dimensions(Empathy = 20.456, Assurance = 11.678,Reliability = 8.806, Responsiveness = 12.155 andTangible = 13.609) indicated significantdifferences at level p<O.05 for all the dimensions.Thus, hypothesis 4 was supported.

While, the numbers reported in Table 15are average SERVQUAL scores within eachOverall category, measured on a -6 to +6 scaleon which the higher (less negative) the category.Most of the combined mean values are negative.implying that customers' perceptions fall shortof expectations.score, the higher is the level ofperceived selV'ice quality. In the sample, thecombined SERVQUAL score [or those in"excellent" category is significantly higher (lessnegative) than for those in the "good" category.Furthermore, respondents in the "good" categoryhave a significantly higher combined SERVQUALscore than those in the "average".

p

0.00(}'0.00i}'0.078'0.000'0.00i}'

0.62570.28510.15760.50730.4711

Qualil)' of Food Available

TangibleReliabilityResponsivenessAssuranceEmpathy

Dimensions

Sou.rce: Survey• Signific::J.nl at p <0.05

PertunikaJ Soc. Sci. & Hum. Vol. 9 No.2 2001 81

Page 12: Measuring Customers' Perceived Service Quality in Hotel

Samsinar Md. Sidin, Md. Zabid Abdul Rashid & Raja Anis Rahyuwati Raja Zainal

TABLE 15Differences in SERVQUAL dimensions perceived to be important by the customers

in assessing overall service quality

Dimensions Source D.F. F Ratio F Prob

Empathy Between Groups 2 20.456 0.000'Within Groups 123Total 125

Assurance Between Groups 2 11.678 0.000'Within Groups 123Total 125

Reliability Between Groups 2 8.806 0.003'Within Groups 123Total 125

Responsiveness Between Groups 2 12.155 0000"Within Groups 123Total 125

Tangibility Benveen Groups 2 13.609 0.000'Within Groups 123Total 125

Souru: Surveya significant at p <0.05

TABLE 16Significant differences in mean scale values for respondents

. segmented according to the variables overall quality

Individual Scale Excellent Good AverageDimensions

Empathy -1.444 -1.611 -6.333Assurance -2.656 -3.389 -7.944Reliability -<J.778 -2.611 -4.500Responsiveness 0.111 -1.689 -3.333Tangible -2.444 -3.111 8.667Combined Scale -1.442 -2.482 -6.155Sample Size 18 90 18

~ Numbers are mean values on a scale ranging from -6 to +6, on which zero implies that consumer perceptions andexpectations coincide, negative values imply thal perceptions fall shon of expectations, and positive values imply thalperceptions exceed expectations.

Table 17Customers' mean perceptions and expectations on SERVQUAL dimensions eN = 126 )

Dimensions

TangibleReliabilityResponsivenessAssuranceEmpathyTotal

Souru: Survey

Perceptions(p)

5.315.205.125.135.235.20

Expectations(E)

5.996.085.966.015.976.00

Serqual(P - E)

-<J.68-0.88-<J.84-<J.88-<J.74-0.80

82 PertanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 13: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel lndusu-y

1. Customers evaluate service quality bycomparing their perceptions of the servicewith their expectations.

2. The content of customers' expectations isbasic. but the structure is complex,characterised by both desired and adequateservice levels that seem to change in responseto a variety of factors.While in the present study, we sought to

learn more about the nature and sources ofcustomers' expectauons given the crucial role inservice quality assessments. Below are a fewsuggestions for managing customers'expectations and raising their perceptions.

employee in each contact department should besingled out for service training.

Guest perceptions must be regularlymonitored through the use of focus groups,personal interviews, guest questionnaires,external qualitative assessments, return visitrecords and any other means by which the hotelcan keep its finger on the pulse of its clientmal:k~t. Each evaluation should be assigned instaustlcal value to demonstrate improvements.One method that hotel operators can use inmonitoring customers perceptions is SERVQUAL(the method used in this present study).

As noted earlier, SERVQUAL is a concisemultiple·item scale with good reliability andvalidity that retailers can use to better understandthe service expectations and perceptions ofcustomers and, as a result. improve service. Ahotel operator, for example. would learn a greatdeal about its service quality and what needs tobe done to improve it by administering bothSERVQUAL and an employee survey, three orfour times a year, plus systematically solicitingand analysing customer suggestions andcomplaints. The employee survey should includequestions concerning perceived impedents tobetter service, e.g., what is the biggest problemyou face to deliver high-quality service to thecustomers?

The previous research done by Parasuraman,Zeithaml and Berty (PZB) has shown that:

Customers' Evaluation oj SERVQUAL

The mean of -0.80 for the SERVQUAL measureimplies that on average respondents' perceptionsfelt short of their expectations. It was very clearfrom Table 17, the perception's mean values(5.20) for every dimension is lower compared tomean values of expectation (6.00).

CONCLUSIONS

Implications

Perhaps the clearest implications from this studyfor hotelier are by meeting or exceeding thecustomers' expectations. Whatever promisesmade to the customers through promotionalcampaign must be delivered once the customerwalks in the door. Since, hotels engaged inrendering personal services deal directly withindividuals every day. Therefore, they have to behighly sensitive to customer's wishes, desires,attitudes and taste. The customer, on the otherhand, is sensitive not only to the quality of theservice rendered but, even more important, tothe employees who are rendering the service(Wright, Handbook of Tourism). Firstimpressions, personal likes and dislikes, the toneof the voice and the expression of co-operationand courtesy are vel)' important.

Contemporary trends indicate that the battlefor the hearts and minds of guests in the 1990swill be fought in the service areas of the hotel.The weapons will be the social skills of theemployees, creating a memorable experiencefor the guest, and the performance techniqueswhich infuse the staff with confident knowledgeand a happy sense ofjob satisfaction. SpecificallyPZB's found out that although reliability is themost important dimensions in meeting customerexpectations, the process dimensions (especiallyassurance, responsiveness, and empathy) are mostimportant in exceeding customer expectations.

These process dimensions clearly involvedthe interactions between employees and thecustomer, therefore, the hotel operator must,make a start on training what are now usuallyreferred to as front line staff in service andperformance improvements. The target areasand job functions for performance and servicetraining should be front office and uniformedstaff, food and beverage service personnel,reservations and financial services staff,housekeeping and valeting functions - indeed allthose having direct contact with guests. Every

• Be reliable - one way for the firm to keepcustomers' expectations from rising is toperfonn the service properly the first time(PZB ]990). It is imperative to be a reliableservice provider that can deliver consistently

Penanikaj. Soc. Sci. & Hum. Vol. 9 No.2 2001 83

Page 14: Measuring Customers' Perceived Service Quality in Hotel

Samsinar Md. Sidin, :Md. Zabid Abdul Rashid & Raja Anis Rahyuw<iti Raja Zainal

competitive performance(de Roulet 1992).Hotel may identify fail points in serviceprocesses which are most vulnerable tomishap by soliciting inputs from employees,by analysing customer complaints, and bymapping the service process itself ­blueprinting. Such points should includenot only customer contact points. but alsoplaces where one functional unit isdependent upon another functional unitfor processing (Riddle 199J).

• Manage Promises - Because hotel operatorsinfluence customers' expectation levels bythe explicit and implicit service promisesthey make, a key approach to managingexpectations is to manage the promises.That is not to overpromising or under­promising. Overpromising raises customers'expectations beyond the capacity of the firmto meet them; while underpromisingpotentially reduces the competitive appealof the offer. Thus, hotel operators have abetter chance of meeting customer expecta­tions when their promises reflect the semceactually delivered rather than an idealisedversion of the service (PZB 1991). Forexample, a brochure photograph of abeautifully appointed hotel room mat enticethe consumer to make a reservation. but itis the room itself that contributes to thecustomer's positive or negative assessmentof the hotel's service quality.

Future Direction

While the present study does add to the presentperceived service quality literature dealing withthe customers' satisfaction, it must be seen as acatalyst for further research as an end in itself.The study must be replicated with a larger samplebefore the results can be staled definitely. Largercoverage in major cities in Malaysia such as inPenang and Johor Bahru may induce differentresults.

Preferably, the methodology would beintelvie,ving respondents. so the researcher hascontrolled over the proportion of respondents.This method can be cost effective as more sets ofquestionnaires will be filled accordingly. Further,it will reduce biasness in term of literacy.

Furthermore. the study should focus on theexpected service - a critical component ofperceived service quality - in addition to beinginfluenced by the customers' personal needs.

shaped by word·of-mouth communications, andpast experience. Research focusing on the relativeimpact of these factors on consumers' selViceexpectations, within service categories, will haveuseful managerial implications.

Limitation of the Stud)'

Notably. the limitation of the study was lack ofcontrol of sample. The researcher had no controlon the respondents' background such as age,etc., in order to get a well- proportioned sample.Besides that. customers have different standardsdepending upon their occupation, standard ofliving, income. and expectations, and a host ofother factors.

The researcher also could not control thequality of the answers as it was highly dependenton the sincerity and willingness of therespondents to answer the given questionnaire.Many of the questionnaires were not answeredproperly.

CONCLUSION

We can conclude that customers' perceptionsfall short of expectations. This is because hotelsin general, however, have been slow to respondto changes in customer expectations in bothpersonal and physical needs. In addition, therehas also been a recent growth in 'consumerism'generally which has resulted in customersexpecting to receive products and services of ahigher minimum acceptable level. People havealso travelled more widely than ever before. dueto package holidays and cheaper transportation.returning with new ideas and new standardsregarding hotel provision. In fact, one factorthat may cause expected service level to rise iscustomer experience. According to researchdone by PZB (1991), the more experiencedcustomers were more likely to have higher serviceexpectations, and to be "squeaky wheels" whenthey were not satisfied.

Consumers will make a decision and takingan action based on their perceptions. Thereforeit is very important for the organisation or hotelindustry to take an effort in comprehending andunderstanding customer's expectation in orderto deliver good services. Good service hereinfers the customer's perception of a servicepresented by the employee, in which if theperceived selV'ice equal to the expected service,they perceived that there is a quality in theservice.

84 PenanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001

Page 15: Measuring Customers' Perceived Service Quality in Hotel

Measuring Customers' Perceived Service Quality in Hotel Industry

On the whole, the customer always wantsmorc and his perception of quality is differentfrom that which the hotel operator can offerhim. Only experience, listening and attentionat every instant may let one have a marc or lesstnle idea. If courtesy and care is the silentlanguage of service, it is one that is understoodacross every international boundary. The servicechallenge of the 1990s will be to capture controlof this language so that the hotel operator mayspeak directly to the guest of their concern forhis welfare.

Putting the service back into the serviceindustry will require the same dedication andskill that has been applied to the developmentof product and functional excellence in the1980s. It is difficult, if not impossible, to calculateits benefits with any precision, but the value ofoutstanding, consistent services illuminates theprofit and loss account and balance sheet alike.The investment in effort will be handsomelyrepaid.

REFERENCES

COWER, D.A. 1987. The consumer selvice andquality challenge. Service IndustriesJournal 7(1):77-99

GRONROOS, C. 1984. A selVice quality model and itsmarketing implications. European] ojMarketing:34-44

GRONROOS, C. 1993. Quality Comes to Service. InThe Selvice Quality Handbook, cd. E.E. Scheungand W.F. Christopher. New York: AMACOM.

MAlA'5!AJ\ TOURls~1 PRO~IOTION BOARD. 1998. AnnualTourism Statistical Repon, Kuala Lumpur.

NIGHTINGALE, M. 1984. Defining QualilY for a QualiryAssurance Programme. In The Practice ofHospitality Management II, ed. Robert Lewis.New York: AVI Publishing.

NIGHTINGALE, M. 1986. Profitability in a ChangingEnvironment. In The Practice of Hospitalit)lManagement 1J, ed. Roben Lewis. New York:AVI Publishing.

PARASURA.\IA:-.l, A" V. ZIETHLo\~lLand L. BERRY. 1985. Aconceptual model of service quality and itsimplications for future research.] of Mmketing(Fall) 49: 41-50.

PARASURA\IA,',', A., V., ZIf.THIAML and L. BERRY. 1988.Communication and conlrol process in thedelivery of service quality. j. of iHarkeling(April)52: 35-48.

PARASURA.\tAN, A., V. ZIETHLA~IL and L. BERRY. 1990.Five imperatives for improving sel'vice quality.Sloan Management Review Summer: 29-40.

PARASURAMAN, A., V. ZIETIIIAML and L. BERRY. 1991.Understanding customer expectation of selvice.Sloan Management Reuiew Spring: 30-46.

SASSER. 1978. Text and Cases. In lv!mlageme1l1 ofScience Operations, cd. Earl, Jr., R. Paul Olsenand D. Daryl Wyckoff. Boston: Ally & Bacon.

SWAN, J.E. and L.J. COMBS. 1976. Prodnctperformance and consumer satisfaction: a newconcepL j. oj Consumer Research 26.

SURPRENANT, C.F. and M.R. SOLO~IO>J. 1987.Predictability and personalization in the selviceenCounter. J. of Marlicling (April)SI: 86-96.

Received: 5 November /999

PertanikaJ. Soc. Sci. & Hum. Vol. 9 No.2 2001 85