Managing Service Quality, Vol 8 Issue 2 Date 1998 ISSN 0960-4529

The effects of survey timing on perceptions of service quality

Martin A. O’Neill

School of Marketing and Tourism, Edith Cowan University, Joondalup Campus, Joondalup, Western Australia

Adrian J. Palmer

University of Ulster, School of Leisure and Tourism, Faculty of Business and Management, Londonderry, Northern Ireland

Rosalind Beggs

University of Ulster, School of Leisure and Tourism, Faculty of Business and Management, Londonderry, Northern

Ireland

Keywords: Consumer attitudes, Model, Perception, Service quality

Type of Article: Survey

Disconfirmation models of service quality have attracted a lot of discussion about how consumers’ expectations are formed, but relatively little about the nature of

their perceptions of service performance. This paper seeks to redress the absence of literature on the psychological underpinnings of perceptions in disconfirmation

models of service quality. It argues that an individual’s perceptions may not be stable over time and that suppliers should be particularly interested in consumers’

perceptions at the time that the next repurchase decision is made. A model of the time elapsed effects of service quality perception is presented and research

reported on a longitudinal survey of hotel customers’ perceptions.

Quality Indicators: Readability**, Practice Implications**, Originality**, Research Implications**

Introduction

Disconfirmation models of service quality came to dominate the literature on service quality from the early 1980s. The idea

that service quality could only be conceptualised with reference to the psychological world of the recipient of a service

distinguished quality in the domain of services from that which occurs in the domain of manufactured goods. While the

quality level of a manufactured good could to a significant extent be measured by reference to fixed tangible benchmarks,

this was not possible with services. The intangible nature of services has led to numerous attempts to conceptualise the

meaning of quality by reference to abstract criteria which vary between individuals, depending on their psychological make

up and sociological background. Disconfirmation models have sought to define quality in terms of the difference between an

individual’s expectations of a service and their perceptions of actual service delivery.

Disconfirmation models have been challenged on a number of grounds. One stream of objections holds that absolute

measures of attitudes provide a more appropriate measure of quality than explanations based disconfirmation models

(Cronin and Taylor, 1992).

There have been numerous criticisms of SERVQUAL for the inductive nature of the original research in that it failed to draw

on the theory base in the disciplines of psychology, social sciences and economics (Andersson, 1992). The failure to define

the perceptions element of SERVQUAL in terms of attitudes has also been criticised (Iacobucci et al., 1994).

This paper addresses a weakness of service quality models which has received inadequate attention, namely the time effect

of measures of respondents’ perceptions. Perceptions of service delivery are critical to quality scores developed within

disconfirmation models. It has been claimed that while any sample’s level of expectations may show a high degree of

uniformity within the sample, perceptions are more likely to show greater levels of variability (Cronin and Taylor, 1992).

Given the attention which has been given to the study of how expectations are formed (Kahneman and Miller, 1986; Zeithaml

et al., 1993), it may seem surprising that relatively little attention has been devoted to an understanding of how perceptions

are formed and sustained. Most importantly, this paper argues that disconfirmation models are flawed because of the effects

of timing on a survey and when a respondent gives a response to their perception of service delivery can be just as important

as the actual recorded score, or the level of expectations against which perceptions are compared.

This paper reviews the literature on perceptions and relates this to models of service quality. A series of hypotheses are

proposed which relate the levels of perceptions to a number of antecedent factors, including the length of time elapsed since

a service encounter, the contextual importance of a particular service attribute, an individual’s ability to learn from sensory

stimuli and their subsequent exposure to related stimuli. These are developed into a proposed model of service quality

perceptions. This paper does not presuppose the superiority of either disconfirmation or absolute value measures of service

quality. The analysis of perceptions which it presents can be incorporated within the dominant paradigms of service quality.

The role of perceptions in models of service quality

At its most basic, quality has been defined as "conforming to requirements" (Crosby, 1984). This implies that organisations

must establish requirements and specifications; once established, the quality goal of the various functions of an organisation

is to comply strictly with these specifications. However, the questions remain: whose requirements and whose

specifications? Thus a second series of definitions state that quality is all about fitness for use, a definition based primarily

on satisfying customers’ needs (Juran, 1982). These two definitions can be united in the concept of customer perceived

quality - quality can only be defined by customers and occurs where an organisation supplies goods or services to a

specification that satisfies their needs.

Numerous attempts have been made to define Service Quality and the closely related concept of Customer Satisfaction

(Cronin and Taylor, 1992; Oliver, 1980). In its broadest sense, service quality can be defined as value generation in the eyes

of consumers, and as such has become an important research topic on account of its relationship to profitability (Buzzell and

Gale, 1987; Reicheld and Sasser, 1990; Rust and Zahorik, 1993).

The research approach of Zeithaml et al. (1993) concentrated on the belief that service quality is measurable, although due

to intangibility it may be more difficult to measure than goods quality. While the SERVQUAL technique has attracted a lot of

attention for its conceptualisation of quality measurement issues, it has also attracted criticism. Some researchers have

debated whether the dimensions of SERVQUAL are consistent across industries; others have suggested better wording for

some of the scale items (Babakus and Boller, 1992). In addition, researchers have asked whether the calculated difference

scores (the difference between expectations and evaluation) are appropriate from a measurement and theoretical perspective

(Brown et al., 1993). From a measurement perspective, there are three psychometric problems associated with the use of

difference scores: reliability, discriminant validity and variance restriction problems. A study by Brown et al. (1993) found

evidence that these psychometric problems arise with the use of SERVQUAL; they recommend instead use of non- difference

score measures which display better discriminant and nomological validity.

However, Zeithaml et al. respond by arguing that the alleged psychometric deficiencies of the difference-score formulation

are less severe than those suggested by critics. Despite their argument that the difference scores offer researchers better

diagnostics than separate measurement of perceptions and expectations, from a theoretical perspective, there is little

evidence to support the relevance of the expectations-performance gap as the basis for measuring service quality. Instead,

considerable research supports a more straightforward approach of assessing quality on the basis of simple performance

based measures (Cronin and Taylor, 1992).

The nature of expectations in disconfirmation models has been subject to a greater level of critical analysis than perceptions.

Much research has been reported which seeks to understand the processes by which expectations of service quality are

formed. Zeithaml et al. (1993) have proposed that three levels of expectations can be defined against which quality is

assessed: the desired level of service, reflecting what the customer wants; the adequate service level, defined as the

standard that customers are willing to accept; and the predicted service level - which they believe is most likely to actually

occur.

There has been debate about whether it is practical to ask consumers about their expectations of a service immediately

before consumption and their perceptions of performance immediately after. Some analyses have therefore used combined

single scales to measure gaps (Babakus and Boller, 1992).

It has been suggested that expectations may not exist or be clear enough in respondents’ minds to act as a benchmark against

which perceptions are assessed (Andersson, 1992). It has also been suggested that expectations are only formed as a result

of previous service encounters, that is, perceptions feed directly into expectations (Teas, 1993).

Theoretical bases for modelling consumer perceptions

An understanding of perceptual processes has formed an important element of models of buyer behaviour (Chisnall, 1975;

Tybout et al., 1981). Williams (1992) states: "An understanding of perceptual processes is important to the marketer as a

customer’s decision to purchase a product will be influenced to a large extent by the way he perceives the product".

Marketers’ involvement with perceptual processes has focused on how individuals process information about a product

prior to purchase, and in particular the manner in which promotional messages are perceived (Myers, 1986). Attention has

also been paid to the manner in which consumption of a product influences individuals’ perceptual processes post purchase,

for example through measures to reduce cognitive dissonance (Festinger, 1957). Relatively little attention has been given to

applying theories of perception developed within the discipline of social psychology to the processes by which an

individual perceives service quality.

Individuals differ in their propensity to perceive elements which make up a service offer; their perception being a product of

their sociological background and psychological make-up. At any one time our senses are over powered with a variety of

challenging stimuli. We are able to focus our attention on only a selected number of stimuli, and only a few are clearly

perceived (Mowen, 1995). The process of discriminating between stimuli is known as "Selectivity of perception". Within

this process, there are three important characteristics in determining a consumer’s propensity to generalise a stimulus, each

referring to threshold levels: "absolute threshold"; "difference threshold" which demands that an increase of stimuli must be

significant before a customer will notice the difference (often referred to as Weber’s Law); and "just-noticeable difference",

which refers to the differential threshold which exists when comparing two stimuli. For example, if a store reduced its own

brand washing powder by five pence, would this be noticeable to the consumer, as opposed to a ten pence reduction

(Assael, 1992). This theory plays on the rationale that some consumers are much more sensitive to stimuli than others.

It is inadequate to study perceptions of one element of a service offer in isolation from the context in which it is offered.

Because consumers are simultaneously confronted with numerous stimuli, it is apparent that a form of organising is

necessary. A group of German psychologists developed the theory based on "the whole is greater than the sum of the parts"

or the theory of Gestalt.

Some researchers argued that perception involved more than the grouping together of a series of discrete stimuli and that the

phenomena should be studied as total (Chisnall, 1975). In other words, the term Gestalt, was used to describe the fact that

perception of an object involved the appreciation of its total nature. Gestalt theorists have shown that perception involves

several different organising tendencies, which rarely operate in isolation. The basic feature of organisation is that the stimuli

may be perceived as an organised whole providing an integrated meaning (Assael, 1992). The tendencies, which rarely

operate in isolation, may be categorised under four broad headings: figure-ground, grouping, closure and contour.

In the case of figure-ground analysis, Williams (1992) points out that "Our perception of the world consists of a number of

objects which appear to stand out against the general background of our experience. Trees and words are seen as figures,

while the sky and page are seen as ground". How a consumer perceives the tangible element of a restaurant meal is

influenced by the broader context in which a meal is consumed (for example, the meal may appear a relatively insignificant

part of a formal social occasion, or a major part of a lunch time meal which is driven by the need to satisfy hunger).

Abercrombie (1967) highlights the fact that perceptions are not stable over time, "with the passage of time experiences

which at first were defined and separate from each other tend to become associated and confused. It is not so much that we

actually forget things, but that we misremember them".

Assael (1992) characterises the memory process by suggesting that processed information, firstly, becomes stored in

short-term memory, then stored in long-term memory, and, third retrieved for purposes of evaluating a product for example.

Asch (1946) argues difficulties exist in the analysis of perception; messages may be distorted to fit in with existing beliefs

and attitudes or "levelling" may occur, where messages become over simplified, so that they become more acceptable.

Finally, perceived differences frequently do not depend on intrinsic qualities; consumers evaluate products against the

background of their experiences, expectations and associations, "Perception is seldom an objective, scientific assessment of

comparative values" (Chisnall, 1975).

Hypotheses to be tested

In view of the importance attached to perceptions in measures of service quality, a number of assumptions have been made

in the literature which need challenging. The need to test assumptions about perceptual processes applies particularly to

disconfirmation but has implications for alternative absolute, performance based measures of quality.

Based on the literature of perception, the following hypotheses are developed.

The literature suggests that perceptions of an event are distorted over time (Abercrombie, 1967; Asch, 1946; Vernon, 1966).

Thus a questionnaire administered during or immediately following a service encounter may give one indicator of

perception. Through a process of selective perception, these may represent only a proportion of stimuli which an individual

is exposed to during a service encounter (Katz, 1968; Osgood, 1962). Through a further process of selectivity, only a

proportion of these will be retained intact. Individuals differ in their memory processing capabilities and perceptions may

either be forgotten completely, or selectively retained (Bartlett, 1932; Bruce and Green, 1985). For marketers it is important

to understand the time effects of perceptions, because it can be argued that the most important perception is that which is

held at the time when the next buying decision is made (Howard and Sheth, 1969). It follows that the preoccupation of many

organisations with measuring perceptions during or immediately following service consumption may have little relevance to

service quality perceptions which are important in subsequently forming future re-purchase intentions. For the purposes of

hypothesis testing, it is hypothesised here that bad experiences are selectively filtered out of memory through a process of

cognitive dissonance and that therefore perception scores will rise as the time elapsed since the original service encounter

increases.

1.H1:Perceptions are not stable over time and an individual’s perception rating of an element of a service encounter will

rise with the passage of time since the initial encounter.

The process of selective perception is influenced by the level of importance which an individual attaches to specific

attributes of a service process (Lloyd and Mayers, 1986). Communication and buyer behaviour models have recognised that

selective perception is a function, among other things, of the importance that an individual attaches to an attribute. By

extension, it would seem reasonable to expect post experience ratings of perceptions to differ from those recorded at the

time that the event reported occurred and that the difference between the two ratings is related to the importance to the

respondent of the attribute being rated (Katz, 1968; Vernon, 1966). The second hypothesis takes the view from the literature

that individuals have a greater propensity to selectively forget bad experiences compared to good experiences

(Abercrombie, 1967; Chisnall, 1975). It can therefore be expected that perception ratings will increase with the passage of

time since the encounter being reported and it is further hypothesised that this increase in perceptions ratings will be greater

where the attribute being reported is of great importance to the respondent.

1.H2:Respondents’ perceptions of a component of a service encounter will increase more as the time elapsed since the

initial encounter increases, compared to those encounters which are considered to be relatively unimportant.

An individual’s perceptions of an event are modified by subsequent stimuli related to the initial encounter (Bartlett, 1932).

There is a large body of literature relating the way in which an individual uses cues to reduce cognitive dissonance, thereby

narrowing the gap between their prior expectations and post-purchase perceptions (Chisnall, 1975; Mowen, 1995; Williams,

1992). It has been argued that individuals are particularly likely to search for cues to rationalise away negative feelings

about a purchase, with the effect that subsequent perception ratings tend to increase (Bruce and Green, 1985; Woodworth

and Schlosberg, 1960). It is hypothesised in this paper that exposure to related stimuli will increase the likelihood that initial

perceptions will be preserved, that is, subsequent stimuli act to preserve favourable memories.

1.H3:An individual’s time elapsed perception of a service encounter is positively related to the level of related stimuli to

which the individual is subsequently exposed.

Finally, individuals differ in their propensity to learn from previous experiences. There is a wide body of literature (Asch,

1946; Bartlett, 1932) which has demonstrated the effect of learning on the ways in which an individual selectively perceives

and retains stimuli. In this study, it is hypothesised that prior learning affects both initial ratings for service quality and

subsequent time elapsed ratings. Prior learning is predicted to moderate the effects of an initial encounter and the speed with

which perceptions subsequently change.

1.H4:A high level of prior experience with an attribute of a service encounter is negatively related to initial ratings of

perceptions and to the rate at which perception ratings subsequently change.

A proposed model of time elapsed perception

The hypotheses to be tested by this research are summarised in Figure 1 A proposed model of time elapsed perception . The

outcome variable is the level of perceptions at a point t+i from the time of initial perceptions, t. Two variables, an

individual’s ability to learn and the importance to the respondent of the service attribute being reported are considered to

have direct effects on time elapsed perceptions as well as an indirect effect through the mediating variable of the initial level

of perception.

Methodology

The hypotheses described above shall be tested using a longitudinal survey of service quality perceptions of UK hotel

customers. A self-completion questionnaire will be administered to guests at the time of check out. A subsequent

questionnaire shall then be despatched to respondents at the address given in the initial response, at times ranging from two

weeks to one year from the date of completion of the original questionnaire. The times will be staggered in order to increase

the range of values, over elapsed time, for which observations are available.

The questions included in the initial questionnaire correspond roughly to the 22 items of the SERVQUAL scale. For each

SERVQUAL item, respondents will be asked to rate their perceptions with scales anchored at very satisfied and very

dissatisfied. They will also be asked how important they consider each item to be in relation to the total service offering. An

individual’s ability to learn can then be operationalized by using measurement scales, the basis of which lie in the

educational learning (Bush and Mosteller, 1955; Diack, 1969). In addition, a number of questions will be asked to establish

the demographic profile of respondents.

The follow up questionnaire shall incorporate the same adapted SERVQUAL scale items as the initial questionnaire, and

respondents will again be asked to rate their perceptions on a scale ranging from 1 to 7. It is assumed that the importance

which respondents will attach to each scale item will not change since the initial administration of the questionnaire.

Questions relating to an individual’s ability to learn will not be asked in the follow up questionnaire. It is, however,

necessary to operationalize a measure of respondents’ exposure to related stimuli since the initial consumption. For this, a

series of questions will be asked relating to a check list of advertisements, mailings, news stories etc., which the respondent

may have been aware of. From this it is hoped, a composite index of exposure to related stimuli will be developed. In

practice, this index measures respondents’ perceptions of stimuli, rather than their exposure to stimuli per se. It is not

considered practical within the confines of a follow up postal questionnaire to explore the complete set of related stimuli

which the respondent will be exposed to, many of which will not have been perceived or subsequently retained.

The survey research is ongoing and data are to be analysed using structural equations modelling.

Conclusion

The study of perceptions has played a secondary role in the service quality literature to the study of expectations. Within the

area of perceptions, very little research has been undertaken to understand the means by which perceptions of service quality

change with the passage of time. The subject is an important one to research, as it can be argued that buyers’ repurchase

intentions are influenced by their perceptions at the time of repurchase, rather than those which prevail immediately

following or during consumption of a service. The desire for short-term results, among both practitioners and academics,

have militated against longitudinal studies of service quality perceptions. This paper has reported on the conceptual

approach to make elapsed time an important component of service quality measurement.

Figure 1 A proposed model of time elapsed perception

References

Abercrombie, M.L.J. (1967, The Anatomy of Judgement: An Investigation into the Process of Perception and Reasoning, 4th

ed., Hutchinson & Co., London.

Andersson, T.D. (1992, "Another model of service quality: a model of causes and effects of service quality tested on a

case within the restaurant industry", in Kunst, P. and Lemmink, J. (Eds), Quality Management in Service, van Goraum, The

Netherlands, pp. 41-8.

Asch, S.E. (1946, "Forming impressions of personality", Journal of Abnormal and Social Psychology, Vol. 46, pp.

1230-40.

Assael, H. (1992, Consumer Behavior & Marketing Action, 4th ed.

Babakus, E. and Boller, G. (1992, "An empirical assessment of the SERVQUAL scale", Journal of Business Research, Vol.

24, May, pp. 253-68.

Bartlett, F.C. (1932, Remembering, Cambridge University Press, Cambridge.

Brown, T., Churchill, G. and Peter, J.P. (1993, "Research note: improving the measurement of service quality", Journal of

Retailing, Vol. 69, Spring, pp. 127-39.

Bruce, V. and Green, P. (1985, Visual Perception: Physiology, Psychology and Ecology, Erlbaum, London.

Bush, R. and Mosteller, F. (1955, Stochastic Models of Learning, John Wiley..

Buzzell, R.D. and Gale, B.M. (1987, The PIMS Principle, The Free Press, New York, NY.

Chisnall, P.M. (1975, Marketing: A Behavioural Analysis, McGraw-Hill, London.

Cronin, J.J. and Taylor, S.A. (1992, "Measuring service quality: a re-examination and extension", Journal of Marketing,

Vol. 56, July, pp. 55-68.

Crosby, P.B. (1984, Quality without Tears, New American Library, New York, NY.

Diack, H. (1969, Reading and the Psychology of Perception, Ray Palmer, Nottingham.

Festinger, L. (1957, A Theory of Cognitive Dissonance, Standford University Press, CA.

Howard, J.H. and Sheth, J.N. (1969, The Theory of Buyer Behaviour, John Wiley & Sons, New York, NY.

Iacobucci, D., Grayson, K.A. and Omstrom, O.L. (1994, "The calculus of service quality and customer satisfaction:

theoretical and empirical differentiation and integration", in Swartz, T.A., Bowen, D.E. and Brown, S.W. (Eds),

Advances in Services Marketing and Management, Vol. 3, JAI Press, Greenwich, CT, pp. 16-84.

Juran, J.M. (1982, Upper Management and Quality, Juran Institute, New York, NY.

Kahneman, D. and Miller, D.T. (1986, "Norm theory: comparing reality to its alternatives", Psychological Review, Vol.

93, pp. 136-53.

Katz, E. (1968, in Abelsonet al.Eds), On Reopening the Question of Selectivity in Exposure to Mass Communications,

Theories of Cognitive Consistency: A Source Book, Rand McNally, Chicago, IL.

Lloyd, P. and Mayers, A. (Eds) (1986, Introduction to Psychology: An Integrated Approach, ASR Manslead, Fontana Press,

London.

Mowen, J.C. (1995, Consumer Behavior, 4th ed., Prentice-Hall, Englewood Cliffs, NJ.

Myers, J.H. (1986, Marketing, McGraw-Hill, USA.

Oliver, R.L. (1980, "A cognitive model of the antecedents and consequences of satisfaction decisions", Journal of

Marketing Research, Vol. 17, November, pp. 460-9.

Osgood, C.E. (1962, Method and Theory in Experimental Psychology, 3rd ed., Oxford University Press, New York, NY.

Reicheld, F.F. and Sasser, W.E.J. (1990, "Zero defections", Harvard Business Review, Vol. 68 No. 5, pp. 105-11.

Rust, R.T. and Zahorik, A.J. (1993, "Customer satisfaction, customer retention, and market share", Journal of Retailing,

Vol. 69 No. 2, Summer, pp. 193-216.

Teas, K.R. (1993, "Consumers’ expectations and the measurement of perceived service quality", Journal of Professional

Services Marketing, Vol. 8 No. 2, pp. 33-53.

Tybout, A.M., Calder, B.J. and Sternthal, B. (1981, "Using information processing theory to design marketing strategies",

Journal of Marketing Research, February, pp. 73-9.

Vernon, M.D. (1966, Experiment in Visual Perception, Penguin, Harmondsworth.

Williams, K.C. (1992, Behavioural Aspects of Marketing, 9th ed., Butterworth-Heinemann, Oxford.

Woodworth, R.S. and Schlosberg, H. (1960, Experimental Psychology, Holt, Reinhart and Winston, New York, NY.

Zeithaml, V.A., Berry, L.L. and Parasuraman, A. (1993, "The nature and determinants of customer expectations of

service", Journal of the Academy of Marketing Science, Vol. 21 No. 1, pp. 1-12.