Untitled Document
[an error occurred while processing this directive]
www.expresshealthcare.in INSIGHT INTO THE BUSINESS OF HEALTHCARE
August 2011  
Untitled Document
Sections

Market
Criticare
Knowledge
Hospital Infra
Radiology
Healthcare Life

Specials

In Imaging

Services
Subscribe/Renew
Archives
Search
Contact Us
Media Kit
Network Sites
Express Computer
Exp. Channel Business
Express Hospitality
Express TravelWorld
Express Pharma
Group Sites
ExpressIndia
Indian Express
Financial Express

[an error occurred while processing this directive]

[an error occurred while processing this directive]

[an error occurred while processing this directive]

Home - Hospital Infra - Article

SERVQUAL Model

Service quality is a focused evaluation that reflects the customers’ perception of elements of service. The SERVQUAL model has been used across various service industries including hospitals to assess and improve service quality. In this paper Dr Sarika Chaturvedi, Research Associate, Foundation for Research in Community Health, documents its use in the Indian hospital environment

In today’s world of fierce competition, rendering quality service has become management’s top-most competitive priorities and a key determinant of return on investment as well as cost reduction. In healthcare organisations, the role that patients play in defining what quality means is now crucial. Owing to information asymmetry that characterises patient-provider interactions, although the technical aspects which form the ‘what’ of a medical service are difficult for patients to evaluate, the functional aspects about ‘how’ services are delivered form important ‘soft’ components of service delivery. Evidence suggests that functional quality is usually the primary determinant of patients’ quality perceptions and is the single most important variable influencing consumers’ value perceptions, which in turn, affect their intentions to purchase products or services.


Dr Sarika Chaturvedi

Research Associate,
Foundation for Research in Community Health

Service quality is a focused evaluation that reflects the customers’ perception of elements of service. Parasuraman, Zeithaml, and Berry (1988) developed a tool to measure service quality – the SERVQUAL. The SERVQUAL has been tested across a number of service industries and its applicability to the hospital environment has also been assessed in the Western settings. However, such evidence from Indian hospital sector, and specifically medium sized hospitals, is sparse. The present paper attempts to fill this gap by reporting its use to assess the service quality at a Pune based tertiary care hospital.

Materials and Methods

Study Context: This study has been conducted at a renowned private multi-speciality hospital in Pune, Western Maharashtra, India. The hospital is functional since over three decades and an established corporate group manages the business after its take over since recent three years. The hospital has a bed strength of 110 beds, employee strength of about 400 and average bed occupancy of over 70 per cent.

Questionnaire, Development and Structure

The developers of SERVQUAL have suggested that ‘it can be adapted or supplemented to fit the characteristics or specific research needs of a particular organisation’. Hence, we subjected the scale to a preliminary evaluation. Inputs were received from senior management personnel and an academician. The decisions to modify the scale were based on relevancy of the questions to hospital services and ability of the patients to respond to those without undue frustration or confusion.

The Gap model based on the SERVQUAL depicted in figure one defines service quality as the difference between perceptions and expectations. It advocates that as service providers’ perceptions are important in design and delivery of services while those of patients are important in the evaluation of services, the views of both parties are important if a thorough understanding of service quality is to be gained. Though the SERVQUAL model considers management in the provider side, keeping in view the Indian hospital sector, which is characterised by mosty not very large hospitals with very small management teams, this study has also used staff members instead of only managers. (Henceforth, management and staff are together referred to as staffs).

The questionnaire included a section on expectations and another on perceptions. Each section consisted 20 items. These were derived from the Yousseef et al modified version of the SERVQUAL. The instrument is added with a section three on demographics (gender, age, education and income) and a final question on overall service quality of the hospital to be rated on a five point scale. All questions were close ended.

The scale used is a five point Likert scale with ends anchored strongly disagree to strongly agree. Though the original SERVQUAL scale uses a seven point Likert scale, and 22 items this study has used a five point scale with 20 items as literature shows no association between the number of items, method of administration and sample size and the reliability of the instrument.

The presentability of the questionnaire was given due attention. Considering that the scale has 20 statements related to expectations from excellent hospitals and another 20 about perceptions about the study hospital, common terms were used for statements. These terms were used instead of repeating the term for each of those statements as has been done in previous research. In the expectation scale the term ‘excellent hospitals will have’ and ‘personnel at excellent hospitals will’ was printed as a common term for the 11 statements and nine statements following these respectively. Similarly, in the perception scale the common terms were ‘this hospital has’…’ and ‘personnel at this hospital’.

Another questionnaire was developed for staffs. It included the same statements as those in the questionnaire for patients, except that the respondents were asked to mark patients’ expectations and perceptions, as understood by them. The common terms, as described above, hence in the staffs’ questionnaire were ‘patients expect excellent hospitals to’ and ‘patients expect personnel at excellent hospitals to…’

The questionnaires were made available in English and Marathi languages after pilot testing. A constant sum scale to determine relative importance of quality dimensions was put as a separate section as in the originally designed questionnaire. However, during the pilot testing it was realised that almost all respondents, in spite of explanation, marked the importance in percentages instead of from a total of 100 units as was desired. It was then decided to omit this section to avoid difficulty in response and also to reduce the length of the questionnaire and rather use regression analysis to reach the objective.

Sample and data collection: A total of 100 patients who had a stay of at least two days were voluntarily enrolled in the study on the day of/evening prior to discharge from the hospital. Patients were requested to fill the responses on the bedside after ensuring that they were comfortable. Each patient took about 20 minutes to complete. Of the 100 forms filled, five were found to be incomplete, and were excluded from analysis. All participants were approached with respect and researcher followed ethical principles in research. Informed consent was obtained from each participant.

The questionnaire for staff members was administered during their duty hours. Staff members included in the study were nurses, doctors- generalists as well as specialists, front office staff, patient assistants and those from accounts, marketing, human resource and billing departments. Staffs who have worked for a minimum of three months at the study hospital were invited to participate in the study. Each staff member took about 12 minutes to complete the form. The data collection for staff and patients was carried out simultaneously during the first quarter of 2009.

Results and Discussion

Patients: Male respondents represented about 57 per cent of the patients surveyed. The study had 52 per cent patients aged below 40 years. The largest group (25 per cent) being in the 21-30 years age group, the smallest group (five per cent) was aged below 20 years while the elderly formed about 12 per cent of respondents. Majority of the patients were educated up to secondary school. Of all the survey questionnaires completed, 39 patients (41 per cent) did not state their income and were labelled 'Not Stated'. Excluding these, majority earned below Rs. 20,000 per month. The average length of stay of the patients as on the day of the study was four days

Staffs: The staff members interviewed included 26 nurses, nine generalist and 14 specialist doctors, and 11others who were staff from other departments as mentioned above. The staff members surveyed included 65 per cent females and 35 per cent males. The higher number of female participants is representative of the hospital industry. Of the staffs interviewed most (81 per cent) were less than 40 years of age. Their average work experience in the hospital industry was 10.6 years while that at the study hospital was 7.8 years.

Validity and Reliability of SERVQUAL Instrument

Considering the objectives of the present study and the recognised instability of the dimensionality of SERVQUAL, it was considered necessary to address the construct validity of the scale. It is noteworthy that in the literature about SERVQUAL, there is no agreement as to which scores (expectation, perception or quality gap scores) should be factor analysed and indeed, all three types of scores have been used in previous research. In the present study the researcher has adopted Vogels et al(1989) view which suggests that the expectation scores should be factor analysed to determine the items that should be included in the service quality dimensions because “these scores are not influenced by possible flaws in the service rendered by various firms in the industry.” Thus, in the present study, SERVQUAL scale was factor-analysed by principal component analysis in the patients ’expectation scores. The Statistical package for Social Sciences (SPSS) was used for data analysis. A rotation procedure was applied to maximise the correlations of item on a factor. Assuming factors were uncorrelated, Varimax rotation was utilised and four factors with Eigen values above one were extracted. To measure the adequacy of the sample for extraction of the four factors the Kaiser-Mayer-Olkin (KMO) measure was computed. The KMO value (.890) indicates that the examined data set is highly adequate for factor analysis. Moreover, the data set was found to be multivariate normal and acceptable for factor analysis according to Bartlett’s test of sphericity (p = 0.000) The Bartlett’s test of Sphericity compared the correlation matrix to the identity matrix and showed clearly a significant relationship between the variables, approximately Chi-Square 990.33, df = 190, p < 0.0001.

Total variance explained (63.039) by these four components exceeds the 60 per cent threshold usually accepted in social sciences to support the solution. The first factor, which explained 24.37 per cent of the total variance, was labelled - The human aspect of the service quality. Factor one contains nine items similar in nature to assurance and empathy and hence could be regarded as the ‘soft’ dimension of quality. The second factor includes four items and explained 13.82 per cent of the total variation. It was labelled – Responsiveness dimension of service quality. Factor three that includes five items, explained 12.73 per cent of the total variance and was named 'Reliability dimension of service quality'. The fourth factor comprises three items and explained 12.1 per cent of the variance, it was named 'Tangible dimension of service quality'. The extracted factors with factor loadings are presented in table one.

The current research results highlighted that the structure proposed by Parasuraman et al., (1988) for the SERVQUAL scale was not confirmed. This finding is in line with previous relevant studies. Many of the items loaded heavily into different factors from the prior dimensions proposed by Parasuraman et.al. (1988). It was decided to keep these dimensions and analyse the data accordingly. The validity of the dimensionality of these groups supports the suggestions made by Babakus, Cronin and others that the dimensions of SERVQUAL may depend on the type of industry being studied.

An internal consistency analysis was performed to assess the reliability aspect of the derived four dimensions. The value of the alpha coefficient ranged from .74 to .89 ( alpha > .70 (Table two) indicating that the four dimensions are reliable measures of service quality. Reliability analysis was similarly conducted for the expectation scale and for the perception scale. Both scales were found to be reliable with Cronbach’s alpha value of .92 and .93 respectively.

Descriptive Statistics

Patients’ expectations (PE): In terms of patients’ expectation, the mean ranged between 3.73 and 4.60. The lowest 'expectation score' was for the statement stating ‘Excellent hospitals will have pamphlets and other communication material visually appealing’, while the highest score was for that stating ‘Excellent hospitals will have the patients best interest at heart’.

This suggests that patients are highly concerned about trust in the hospital. This could be explained by the mystified nature of medical services or simply that these are high in credence attributes and hence it is highly difficult for customers to evaluate them. Another reason for the high expectation could possibly also be news reports of growing incidences of unethical conduct and irrational practices in Indian hospitals.

The fact that all the top five expectations are in the human aspect factor indicate that the management must ensure that the patients realise that the hospital has patients’ best interest at heart. It is important that this is emphasised in communications to patients and also through staff behaviour.

Amongst the five items that received the lowest expectation scores, three are from the tangibles dimension while two are from the responsiveness dimension.

Tangible dimension includes items stating about modern looking equipment, visually appealing physical facilities and visually appealing communication material. All the tangible dimensions receiving lowest scores indicate that patients do not go very much by the look of the hospital as is usually assumed.

It is surprising that patients have one of the lowest expectations to staff never being too busy to respond to patients needs. Possibly patients perceive a hospital to have a large client base and hence likely to be offering good quality by noticing staff to be busy.

Patients’ Perceptions

Patients’ mean scores for 'perception of actual service' ranged between 3.65 and 4.32. The lowest ‘perception score' was for ‘This hospital has pamphlets and other communication material visually appealing’. The highest 'perception score' was for the two statements stating ‘The personnel in this hospital give prompt service to patients’ and ‘The personnel in this hospital are always willing to help patients’. The findings of high perceptions in the human factor dimension imply that the personnel are perceived to be serving well.

The lowest perception is for ‘the hospital has visually appealing communication material’, and about meals being served hot and of good flavour indicating patients unhappiness about catering services. Indian hospital managers need to particularly consider this in view of the varied food habits in the country probably indicating need to give choice of food items to patients.

One item from the human factor that has scored low perceptions is about personnel telling patients exactly when services will be performed.

Patients’ perceptions are low about two items from the tangible dimensions. As these items are also among the low expectation items, the implication is to include these items in areas of improvement but not in the highest priority category.

A comparison of patients’ expectations and perceptions for the four factors is presented in figure one. Statistical analysis shows that the mean patient expectations for two of the factors- Factors one and three are significantly different (p<0.05) from the respective mean patient perceptions.

Staffs Understanding of Patients’ Expectations and Perceptions

The mean value for staffs understanding of patients’ expectations ranged from 3.87 to 4.70. The lowest score is for statement stating ‘patients expect excellent hospitals to have visually appealing pamphlets and other communication material’, while highest score was for the statement ‘patients expect excellent hospitals to always be willing to help patients’.

Mean scores for staff understanding of patient perceptions ranged between 3.52 and 4.28, the lowest being for statement five which stated that – ‘patients’ perceive this hospital has pamphlets and other communication material visually appealing’ while highest was for statement which stated that ‘ patients’ perceive the personnel in this hospital understand the specific needs of their patients’. The mean patient expectation and perception scores as perceived by staff for each factor are presented in figure two.

Gaps In Service Quality

Gap five: The Customer Gap: This study finds differences between patients’ expectations from an excellent hospital and their perceptions of the service quality delivered at the study hospital. The SERVQUAL model labels this as gap five- the customer gap. This study finds that there exists gap five in the hospital analysis reveals that these gaps are significant (p<0.05) in the human factor and the reliability dimension.

Gap one: The Knowledge Gap: The SERVQUAL model defines gap one labelled ‘Knowledge gap’ as the gap between the management/ staff understanding of patients’ expectations and perceptions and the actual expectations and perceptions of patients about service quality at the hospital. It is the first step by which hospitals can proceed to reaching patients expectations. This entails identifying areas where patients expectations and perceptions of service quality mismatch with the staff and management understanding of these. The figure three shows that staff members have largely overestimated patient expectations. As regards understanding of patient perceptions of service quality, the reverse is found- staff has underestimated the hospitals performance.

Which Dimensions Matter Most?

In order to examine the effect of the quality gaps - in the four dimensions - on the patients’ overall evaluation of the quality of the service provided by the hospitals (general question in the questionnaire), regression analysis was performed. The four quality gaps were used as the predictors of overall quality of the services provided. Considering the independent variables with statistically significant coefficients, it is evident that patients’ perceptions of service quality are attributed to the responsiveness gap as presented in Table three, which is in fact the predictor of overall service quality. The above research finding is worth reporting since it indicates that the quality of the service provided to patients in the study hospital depends heavily on improving the responsiveness (Factor two).

Conclusion

This study leads to the following conclusions which are particularly important for further use of the SERVQUAL model in Indian hospital settings.

The SERVQUAL questionnaire can be modified to specific needs as recommended by Parasuraman et al. However, this raises concern about loss of the power of standardisation. Although the scale is tested for reliability and validity, the process of evolution of the scale being subjective, the possibility of negligence of important items can not be ruled out. The length of the questionnaire is another important consideration in using the SERVQUAL model. In view of the middle socio-economic class patients’ and their cultural contexts, this study has attempted to improve the presentability and the readability of the questionnaire which was found useful in keeping participants interest in it.

Involvement of staffs directly interacting with patients instead of management alone as recommended by the original SERVQUAL model is a unique feature of this study. This has been found helpful in not only better identifying understanding of patients’ expectations and perceptions, but also in creating acceptance for subsequent service quality improvement strategies.

This study concludes that the dimensional structure of SERVQUAL is unstable within the hospital industry and this finding is similar to that reported by Carman (1990) and Babakus and Boller (1992). While the original study by Parsuraman et al 1988 proposed five (universal) dimensions which were supposed to measure the service quality in any sector, this study reports four dimensions for the hospital industry rather than five. This result supports the work of quality gurus who found that quality is a relative notion with respect to a given client segment.

The regression analysis found the service quality gap in the responsiveness dimension to be the most strong predictor of overall service quality followed by reliability. However the model points that there are predictors of service quality other than the gaps in the four dimensions that this study finds.

References:
Anderson, E.A. Measuring service quality at a university health clinic. Int J of Health Care Quality Assurance 1995, 8(2), 32-37.
Donnabedian et al. “The Quality of Care: How can it be Assessed” , JAMA 1988, 260(12), 1143-48.
Parasuraman, A., Zeithaml, V.A., & Berry, L.L. SERVQUAL: A multiple-item scale for measuring consumer perceptions of service quality. Journal of Retailing 1988, 64(1), 12-40.
Youssef, F., Nel, D., & Bovaird, T. Service quality in NHS hospital. Journal of Management in Medicine 1995, 9(1), 66-74.
Nyeck, S., Morales, M., Ladhari, R., & Pons, F. Ten years of service quality measurement: reviewing the use of the SERVQUAL instrument. Cuadernos de Difusion 2002, 7(13), 101-7.
Vandamme, R., & Leunis, J. Development of a multiple-item scale for measuring hospital service quality. International Journal of Service Industry Management 1993, 4(3), 30-49.
Vogels, R., Lemmink, J. and Kasper, H. "Some Methodological Remarks on the SERVQUAL Model", in Avloniti, G.L. Papavasiliou, NK. And Kouremenos, A.G. (Eds), Marketing Thought and Practice in the 1990s, Conference Proceedings of the XVIII Annual Conference of the European Marketing Academy.1989.
Comrey, A. and Lee, H. A First Course in Factor Analysis, Lawrence Erlbaum Associates 1991, Hillsdale. NJ.
Hair, J., Anderson, R., Tathain, R. and Black, W. Multivariate Data Analysis, 4th ed. 1995, Prentice - Hall, Engelwood Cliffs, NJ.
Carman, J.M. Consumer perceptions of service quality: An assessment of the SERVQUAL dimensions. Journal of Retailing,1990, 66(1), 33-55
Licata, J., Mowen, J. and Chakraboty, G. Diagnosing perceived quality in the medical service channel, Journal of Health Care Marketing,1995, 15(4), 42-9.
Lim, P.C. and Tag, N.K. A study of patients' expectations and satisfaction in Singapore hospital, International Journal of Health Care Quality Assurance,2000 . 13 (7), 290-99.
Babakus, E., & Boller, G.W. An empirical assessment of the SERVQUAL scale. Journal of Business Research,1992, 24, 253-68.,
Cronin, J.J., & Taylor, S.A. Measuring service quality: A re-examination and extension, Journal of Marketing,1992, 56(3), 56-68.
Nunnaly, J. Psychometric Theory, McGrow-Hill Book Co.1978, New York.

 

Untitled Document

FEEDBACK: We would love to hear from you -- what you like about our content, what you dont, and even how you think we can improve. Please send your feedback to: healthcare@expressindia.com


Copyright 2001: The Indian Express Limited. All rights reserved throughout the world. This entire site is compiled in Mumbai by the Business Publications Division (BPD) of The Indian Express Limited. Site managed by BPD.