Services Cape And

8/3/2019 Services Cape And

1/33

Servicescape and loyaltyintentions: an empirical

investigationLloyd C. Harris

Warwick Business School, The University of Warwick, Coventry, UK, and

Chris EzehCardiff Business School, Cardiff University, Cardiff, UK

Abstract

Purpose This paper seeks better to conceptualise, operationalise and subsequently to test a

multi-dimensional and more social view of servicescape and the direct and moderated linkages withloyalty intentions.

Design/methodology/approach A survey research method was used to study servicescapes inthe context of UK restaurants.

Findings In furtherance of conceptualisation efforts, a model is developed to evaluate the linearinfluences of nine servicescape variables on customers loyalty intentions. Additionally, the modelappraises the impact of personal and environmental factors which moderate the servicescape-loyaltyintentions relationship. Analysis of survey responses finds a number of significant associations withloyalty intentions.

Practical implications The results of the study indicate that practitioners should reflect carefullyon a range of servicescape variables and judiciously manage such factors to improve the extent towhich consumers are likely to foster positive intentions to be loyal.

Originality/value The paper contributes a multi-dimensional and more social framework of

servicescape that is subsequently operationalised and tested. It also supplies a measure of servicescapethat future researchers may find useful.

Keywords Services marketing, Marketing environment, Customer loyalty, Psychology, Restaurants

Paper type Research paper

The notion that aspects of servicescapes may have a strong influence on customers

loyalty intentions is an established one in a range of contexts (e.g. Foxall and Greenley,

1999; Cronin, 2003; Foxall and Yani-de-Soriano, 2005). Indeed, the general consensus is

that service settings provoke emotional responses which lead customers to either

choose to continue their association with a particular service provider, or to discontinue

patronage (Foxall and Greenley, 1999; Lovelock, 2001; Hoffman and Turley, 2002;

Tombs and McColl-Kennedy, 2003). Despite the apparent importance of servicescapes,there remains a surprising lack of empirical research addressing its role in

consumption settings (Wakefield and Blodgett, 1999; Tombs and McColl-Kennedy,

2003) and its impact on customers purchase decisions (Cronin, 2003; Foxall and

Yani-de-Soriano, 2005). Generally, the influence exerted by servicescapes has barely

been tapped as a tangible organisational resource (Sweeney and Wyber, 2002; Ezeh

and Harris, 2007). Hence, Hoffman and Turley (2002, p. 33) question the lack of

empirical development in the area, stating that to date little empirical research has

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0309-0566.htm

EJM42,3/4

390

Received September 2006Revised January 2007Accepted February 2007

European Journal of Marketing

Vol. 42 No. 3/4, 2008

pp. 390-422

q Emerald Group Publishing Limited

0309-0566

DOI 10.1108/03090560810852995


2/33

been conducted that investigates the impact of atmospherics as they relate to thecustomers service experience. This leads Chebat and Dube (2000, p. 89) to observethat complex decisions on the design and management of retail atmospherics arefrequently made on the basis of a dearth of information that in other domains would be

unimaginable.While studies focusing on individual environmental factors are relatively common

(e.g. Fiore et al., 2000; Summers and Hebert, 2001; Sweeney and Wyber, 2002), Tombsand McColl-Kennedy (2003, p. 447) note that empirical studies of multipleenvironmental effects are rare and largely limited to the study of two environmentalvariables (a good example being the excellent studies of scent and music by Bone andEllen, 1999, and by Spangenberg et al., 2005). Nevertheless, research intoenvironmental psychology continues to note that studies of the links betweenenvironmental influences and human behaviour should incorporate a wide range ofvariables in order further to understand the dynamics of environmental impacts onhuman attitudes and behaviours (e.g. Cassidy, 1997; Kaya and Erkip, 1999; Moore et al.,2003; Lee and Brand, 2005). This view is echoed in Tombs and McColl-Kennedy (2003)who argue that social spects of servicescape are typically ignored in extantconceptualizations. Indeed, Tombs and McColl-Kennedy (2003) go on to lament the lackof empirical research into such issues and call for rigorous empirical research toaddress this imbalance.

This study has the aim of increasing the current level of theoretical understandingconcerning the influential role of servicescapes in the determination of customersloyalty intentions and highlighting the practical usefulness of the manner in whichservicescape cues can be managed by service organisations within the UK restaurantsector. In this regard, this study is designed to furnish contributions to theory throughadvancing a multi-dimensional framework for modelling servicescape that includesboth tangible and more social aspects, as well as empirical contributions through the

supply of empirical insights into contemporary servicescapes. Throughoperationalising a measure of servicescape, it is intended to provide methodologicalcontributions as well as practical implications.

Beginning with proffered definitions of servicescapes, this paper will reviewprevious research in the area and augment this by presenting a conceptual modelwhich will subsequently be subjected to empirical analysis. Finally, the findings of thestudy are presented and recommendations for both theorists and practitioners aremade.

Definitions of servicescapeThe context of service encounters has been conceptualised by a variety of scholarsusing several different terms. For instance, it has been referred to variously as the

physical environment (Baker, 1987), atmospherics (Kotler, 1973), marketingenvironment (Turley and Milliman, 2000), economic environment (Arnold et al.,1996), interactive theatre (Mathwick et al., 2001), healthscapes (Hutton andRichardson, 1995), environmental psychology (Weinrach, 2000), servicescape(Bitner, 1992), store environment (Roy and Tai, 2003), service environment (Cronin,2003) and social-servicescape (Tombs and McColl-Kennedy, 2003). These differentterms reflect the eclectic nature of the literature in the area and have been used overthe last 30 years of exploration and conceptual development in this literature stream

Servicescape andloyalty

intentions

391


3/33

(Turley and Milliman, 2000, p. 193). However, for the purpose of consistency, this studywill employ the term servicescape, as coined by Bitner (1992), in reference to thedesign of the physical environment and service staff qualities that characterise thecontext which houses the service encounter, which elicits internal reactions from

customers leading to the display of approach or avoidance behaviours.

Classification of servicescape elementsBitner (1992, p. 67) argues that a total configuration of environmental dimensions isresponsible for the constitution of the servicescape. This view builds on earlierenvironmental psychology research arguments that customers respond toenvironments in a holistic manner, after assessing the effects of all the discretestimuli present (see Holahan, 1982). The notion of stimuli congruency is also reflectedin the conceptualisation of servicescapes as originally advanced by Bitner (1992).

Hoffman and Turley (2002, p. 35) suggest that both tangible and intangiblecomponents are essential in creating service experiences. Similarly, Kotler (1973,

pp. 50-1) states that the atmosphere of a particular set of surroundings is describablein sensory terms which include perceptions that are visual, aural, olfactory and tactile.A review reveals that several scholars have advanced broadly similar componentswhich they argue comprise the servicescape and so should be applied in servicescapeanalysis (e.g. Westbrook, 1981; Booms and Bitner, 1982; Bitner, 1992; Berman andEvans, 1995).

The importance of a particular servicescape component is likely to vary acrossdifferent service organisations (Kotler, 1973; Bitner, 1992). However, relevantdimensions of the servicescape can be isolated and general patterns can be explored(Bitner, 1992, p. 65). Therefore, as this paper aims to evaluate the effect of servicescapecues on customers leading to their formation of loyalty intentions, Bakers (1987)typology of ambient factors, design factors and social factors is particularly relevant

because of its sociological bent which acknowledges the influence of humancharacteristics in shaping the service experience. In extending and broadening thisview, Tombs and McColl-Kennedy (2003) argue that existing models are limited by anover-concentration on the tangible characteristics of the servicescape. Consequently,they endorse the re-conceptualisation of servicescape to include more social aspects.While their focus is on customer-to-customer interaction, their central thesis is that thesocial aspects of servicescape need to be incorporated (the nature of which is likely tovary from sector to sector).

Limitations to servicescapes literatureFour major limitations to servicescape study can be identified. Firstly, conceptualliterature universally agrees with the proposition that elements of the servicescape

affect behaviour (see Cronin, 2003; Foxall and Yani-de-Soriano, 2005) yet there has beena significant lack of conceptual development (Bitner, 1992; cf. Tombs andMcColl-Kennedy, 2003), thus necessitating repeated calls for further study (e.g.Foxall and Greenley, 1999; Hoffman and Turley, 2002). Secondly, although there havebeen minimal conceptual advances, there remains even less empirical researchaddressing the role of servicescapes in consumption settings (Bitner, 1992) and itsimpact on customers purchase decisions (Cronin, 2003). Thirdly, most of the studies(both empirical and conceptual) that focussed on the influence of servicescape elements

EJM42,3/4

392


4/33

on purchase decisions have tended to focus on single servicescape elements such asmusic (see Areni and Kim, 1993; Sweeney and Wyber, 2002), aroma (see Spangenberget al., 1996; Fiore et al., 2000) and light (see Areni and Kim, 1994; Summers and Hebert,2001); to the extent that little is known about the global configurations of aspects of the

servicescape (Everett et al., 1994; Tombs and McColl-Kennedy, 2003). Only very fewstudies have incorporated the influence of more than one servicescape element (e.g.Wakefield and Blodgett, 1996; Mattila and Wirtz, 2001; Spangenberg et al., 2005) and ofthese, none has gone further empirically to investigate the potential moderating effectsof the personal and environmental factors of specific importance to customers. Finally,as stated previously, extant models of servicescape are narrowly focussed on physicaldimensions to the detriment of social-interactive dimensions (see Tombs andMcColl-Kennedy, 2003). Although the servicescape concept draws from other broadlysimilar perspectives on the physical environment as advanced in psychology (seeMintz, 1956), landscaping, architecture and interior design (Donovan and Rossiter,1982), commentators have consistently argued that a more rigorous conceptualisationwould be beneficial (e.g. Donovan and Rossiter, 1982; Bitner, 1992; Turley andMilliman, 2000). In this regard, the arguments of Tombs and McColl-Kennedy areaccepted, in that, it is believed that to produce a more rigorous conceptualisation, widerliterature which combines the interaction of human and inanimate variables during theservice experience may prove useful lenses through which servicescapes can beexplored, conceptualised and empirically tested.

These issues indicate that additional conceptual work and empirical research isneeded on the nature and dynamics of servicescapes using multiple servicescapeelements. It is expected that such integration of multiple elements will increase thescant knowledge base on the global configurations of servicescapes.

A conceptual model of servicescape

To aid research efforts towards obtaining a global configuration of the servicescape, aconceptual model is developed (see Figure 1). The model depicts the proposed linearrelationships between nine servicescape elements and loyalty intentions. The first fivevariables reflect physical aspects of servicescape, while the remaining four variablesare more social in nature. In addition, four factors have been positioned in the model to

Figure 1.A conceptual model of

servicescapes


intentions

393


5/33

act as moderators of these proposed linear relationships. The inclusion of moderatingfactors in the model is premised on the need to obtain a deeper and betterconceptualisation of servicescapes dynamics.

Prior to explicating the conceptual rationale for each component of servicescape and

the factors theorised as moderators, it is worthwhile briefly outlining the nature of thedependant variable presented in the model. Loyalty intentions are defined as thedisposition of customers to either repurchase a product/service from a providingorganisation, or go to a competitor (Jacoby and Chestnut, 1978). Wong et al. (1999,p. 54) state that understanding the behaviours and conditions that foster repeatpatronage is an important part of marketing endeavour. Loyalty intentions became avery popular area in marketing research when studies discovered the immense benefitsavailable to organisations through the retention of just one customer. For instance, astudy by Reichheld et al. (2000) reveals that a five percentage-point shift in customerretention consistently results in 25-100 per cent profit increases. Based on this, Zins(2001, p. 288) contends that loyalty management is a key concern in marketing.Similarly, Pritchard et al. (1999, p. 333) postulate that, understanding how or why asense of loyalty develops in customers remains one of the crucial management issues ofour day. Surprisingly however, as Gremler and Brown (1996) point out, there has beena primary focus on customer loyalty in product-related or brand loyalty while loyaltyto service organisations has remained under-researched.

Ambient conditionsMusic is defined by Brunner (1990, p. 94) as a complex chemistry of controllableelements [. . .] rather than a generic sonic mass. The ability of both background andforeground music to affect moods has been demonstrated throughout history by poets,playwrights and composers (Brunner, 1990) and is reflected extensively inservicescape study (see Milliman, 1982, 1986; Yalch and Spangenberg, 1988, 1990;

Baker et al., 1992; Areni and Kim, 1993; Dube et al., 1995; North et al., 1999; Mattila andWirtz, 2001). This is because the music played in a servicescape is capable of not onlystimulating emotions and therefore mood states (Yalch and Spangenberg, 1990; Chebatet al., 1993; Dube et al., 1995), but also impacting on their purchasing behaviour (Areniand Kim, 1993). Thus, both empirical and conceptual studies on the element of musicsuggest that music has become a major component of consumer marketing, both atthe point of purchase and in advertising (Brunner, 1990, p. 94). Indeed, in studies ofdining, North and Hargreaves (1996) highlight that variations in music effectindividuals responses while Milliman (1986) finds that music tempo influences thespeed with which restaurant diners consume their meals. While the tonality, tempo,and texture of music (see Kellaris and Kent, 1993) can be examined, in the currentstudy the focus is on customers perception of music appropriateness (that is, given

subjective interpretations of music, the extent to which consumers consider the musicheard to be suitable or matched to the context of consumption). Therefore, it ishypothesised that:

H1. The greater customers perception of music appropriateness, the greater theintentions of customers to be loyal.

Aroma can be defined as a distinctive smell (especially one that is pleasant). Thesuggestion that aroma affects moods is an ancient one (Hallett and Silver, 2004). This

EJM42,3/4

394


6/33

is reflected in the extensive use of aroma in organisational settings as has been widelyitemised in extant aroma/odour literature (see Mattila and Wirtz, 2001; Chebat andMichon, 2003; Spangenberg et al., 2005). Morrin and Ratneshwar (2003, p. 21) assertthat environmental fragrancing is now becoming a common practice in retailing,

restaurants and other service-oriented businesses because pleasant scents encouragecustomers to spend more time in the servicescape. This is due to the reason thatrespiration deepens in the presence of a pleasant aroma (Takagi, 1989) whereas anunpleasant aroma halts breathing and even causes physical withdrawal (Levine andMcBurney, 1986). Studies of olfactory stimuli within the food service industry havefound that the systematic use of aroma in bakeries increases sales by 300 per cent (seeHirsch, 1991), while coffee chains such have Starbucks have also publicly recognisedthe importance of aroma in service settings (see Hunter, 1995). Thus:

H2. The greater customers appreciation of servicescape aroma, the greater theintentions of customers to be loyal.

Cleanliness can be defined as the absence of dirt (including dust, stains and badsmells). Rosenquist (2005) notes that human demands regarding levels of sanitationand hygiene are linked to a hierarchy of needs, suggesting that consumers in developedeconomies pursuing hedonistic or leisure activities are likely to want (and expect) highstandards of cleanliness. The importance of cleanliness within the service environmenthas been highlighted extensively in the literature (see Fitzsimmons, 2003; Berta, 2005).Indeed, Marinucci (2002, p. 7) contends that marketing-wise, cleanliness makesperfectly good sense. More specifically, Wakefield and Blodgett (1996, p. 49) assertthat cleanliness is an important part of the servicescape. The importance ofcleanliness is manifested in the high number of studies which directly link cleanlinessto customers evaluation of the servicescape, and hence their future behaviour (e.g.Stern and Stern, 2000). Therefore:

H3. The greater customers perception of servicescape cleanliness, the higher theintentions of customers to be loyal.

Design factorsImplicit communicators are visual elements related to aesthetic appeal (Bitner, 1992)or the tangible cues which communicate information about an organisations beliefsand its ways of doing things (Ott, 1989; Buchanan and Huczynski, 1997). In otherwords, they are used to convey and enhance a certain image or mood (Wakefield andBlodgett, 1996). Bitner (1992) highlights that implicit communicators like decor (forinstance the quality of materials used in construction, artwork and floor coverings)give implicit cues to customers on the norms and expectations for behaviour in the

servicescape. In this sense, decor is a visual symbol used to create an appropriateatmosphere within the servicescape (Nguyen and Leblanc, 2002). Indeed, research intothe environmental psychology of decor confirms that variations in decor haveprofound effects on human behaviour, particularly influencing social intimacy (see forexample, Gifford, 1988), so important in the social settings of restaurants. Similarly,decor can influence a customers belief about whether a firm is trustworthy oruntrustworthy, expensive or affordable (Bitner, 1992) and successful or unsuccessful(Nguyen and Leblanc, 2002); thereby providing the customer with an intrinsic cue on


intentions

395


7/33

which to form beliefs about the organisations effectiveness (Bitner, 1992) andthereafter develop their loyalty intentions. Therefore:

H4. The greater customers evaluation of servicescape aesthetic appeal, the

greater the intentions of customers to be loyal.Furnishing is defined as the element which directly affects the comfort andperformance of both customers and employees in the servicescape (Baker, 1987;Bitner, 1992). Arneill and Devlin (2002) use and experimental approach to highlightthat individuals assessment of the service quality is, in part, contingent on their visualinterpretations of the furnishings of a context. Moreover, assessments are also relatedto physical experience, as noted by Wakefield and Blodgett (1996, p. 54) who affirmthat comfort becomes very important when for example, customers remain in thesame seat for extended periods of time. Evaluations of furnishings also encompassissues of personal space and crowding. Thus, Aubert-Gamet (1997, p. 28) state that theinternal layout of buildings may either ease or restrict movement, thereby evoking asense of crowding or spaciousness, while environmental psychology is replete withstudies that find that the extent of crowding of personal space effects humans moodsand behaviours (e.g. Kaya and Erkip, 1999). Hence:

H5. The greater customers evaluation of the comfort of furnishings, the greaterthe intentions of customers to be loyal.

Staff behaviourDefined by Brown et al. (2002, p. 111) as an employees tendency or predisposition tomeet customer needs in an on-the-job context, a customer-oriented organisation isgeared towards a strong emphasis on both internal collaboration and customerfamiliarity (Deshpande et al., 1993; Noble et al., 2002). Customer-orientation is thereforean individual-level construct that is central to a service organisations ability to be

market oriented (Brown et al., 2002, p. 110). The significance of staff behaviour as aproduct of market orientation is related to organisational culture (Deshpande andWebster, 1989; Harris, 1998), which promotes shared values and a strong sense ofcamaraderie (Jaworski and Kohli, 1993), thus implicitly orchestrating the behaviour ofstaff in the servicescape. Surprisingly however, despite the strong link identifiedbetween customer orientation and the implementation of the marketing concept,research in the area remains limited (Brown et al., 2002; Strong and Harris, 2004).Therefore, this study will attempt to add to the literature in the area by empiricallyevaluating the influence of customer orientation on customers loyalty intentions ashypothesised below:

H6. The greater customers perception of staff customer orientation, the greater

the intentions of customers to be loyal.Credibility is defined as the degree to which a source is perceived as trustworthy andcompetent (Rogers and Shoemaker, 1971) or the extent to which a source is perceivedto possess the necessary expertise and can be trusted to provide an objective opinion(Ohanian, 1990). The need for organisations to project their credibility to customers hasencouraged a lot of research in the area of corporate credibility which has beenrevealed to have the ability to influence customers attitudes and purchase intentions(Lafferty and Goldsmith, 1999; Goldsmith et al., 2000a, b). Studies of servicescape

EJM42,3/4

396


8/33

indicate that customers general perceptions of the retail setting is able to influencetheir beliefs about the salespeople representing the organisation, with well-designedservice environments being subconsciously linked to more credible service staff(Bitner, 1992). Thus, customers are more likely to affiliate with salespeople working in

nicer retail environments which should lead to an increase in the perceived level ofcredibility for the salesperson (Sharma and Stafford, 2000, p. 184). Thus:

H7. The greater customers perception of staff credibility, the greater theintentions of customers to be loyal.

Staff imageCompetence can be defined as the expertise of service staff derived from knowledge ofthe subject (Goldsmith et al., 2000a). The importance of staff competence is all themore emphasised within a service environment because the intangibility of thedelivered service accentuates the dependence of customers on service staff to solvetheir problems (Sharma and Stafford, 2000, p. 185). To underpin this, a study by

Cooper and Summer (1990; cited in Berry and Parasuraman, 1991) identified the abilityof salespeople to solve customer problems during and after the selling process to beamong the highest ranked issues of importance to customers. Furthermore, stronglinks have been found to exist between the competence of service staff and competitiveadvantage because most successful strategies are built firmly on inimitablecompetence (Bogner et al., 1999, p. 276). The significance of staff competenceperceptions is further highlighted by Hill et al. (1989), whose study found staffcompetence to be the most important criterion applied by customers in theirsubsequent selection of service providers. Therefore:

H8. The greater customers perception of staff competence, the greater theintentions of customers to be loyal.

Physical attractiveness is the perception of the physical traits of an individual humanperson or group, race, or type of people, as attractive or beautiful. Evidencesuggests thatthe physical attractiveness of the service personnel at the interface with customers cangreatly enhance the service experience (Baker, 1987, p. 81) and serve as motivation forapproach behaviour (Bitner, 1992). This position is supported by the assertion ofAppiah-Adu et al. (2000) that elements of an organisations marketing culture may beobserved from the emphasis placed on the appearance of staff. This is because staffappearance carries a hidden message which imparts meaning through object language(Ruesch and Kees, 1956), thereby helping customers form judgements about the servicebefore and after it has been consumed. Hutton and Richardson (1995, p. 59) maintain thatstaff physical attractiveness forms the major portion of a service organisations imageand so the need arises for the display of a pleasing physical demeanour through cleanand colourful uniforms and proper personal grooming. Hence:

H9. The greater customers evaluation of staff physical attractiveness, the greaterthe intentions of customers to be loyal.

Moderating factorsA review of extant theory reveals a variety of factors that are likely to moderatenegatively the relationship between servicescape and the loyalty intentions of


intentions

397


9/33

consumers (see Figure 1). That is, a number of factors appear likely to reduce theimpact of servicescape variables on loyalty intentions. First, two personal factors(variety-seeking behaviour and perceived sacrifice) emerge as important.Variety-seeking behaviour is defined as low customer involvement with a

particular brand which typically leads to defection out of boredom rather than dueto dissatisfaction (Mitchell, 1992). In contrast, perceived sacrifice refers to the(typically) non-monetary costs such as psychic cost which represents customersmental stress or emotional labour during the shopping experience (Baker et al., 2002),convenience which is the trade-off between what is delivered and the effort required toobtain it (Butcher et al., 2002) and time/effort costs which involves customersperceptions of the time and effort they are likely to expend at a store (Baker et al., 2002).In summary, where consumers have a high degree of variety seeking behaviour and/orperceive repeat patronage as involving significant sacrifice, it is argued that the linkbetween service and loyalty intentions will be moderated. Hence:

H10. (a) The variety-seeking nature of customers and/or (b) the level of customers

perceived sacrifice negatively moderate the relationships between customersperception of the individual servicescape elements of: (i) music, (ii) aroma, (iii)cleanliness, (iv) implicit communicators, (v) furnishing, (vi) customerorientation, (vii) credibility, (viii) competence and (ix) staff physicalattractiveness; and intentions of customers to be loyal.

Two environmental factors are also important. First, economic turbulence can bedefined as the volatility of the aggregate economy in terms of activity changes inemployment, wage levels, interest rates, price levels, inflation and overall customerbuying power (Ercel, 1998; McIntyre and Pencavel, 2004). Competitive intensity isdefined as the degree of competition in an industry (Slater and Narver, 1994) or theabilities, resources and actions of competitors (Jaworski and Kohli, 1993). Statedotherwise, the competitive intensity in an industry reflects the availability ofsubstitutes or switching opportunities open to customers. In summary, whereconsumers perceive a high degree of economic turbulence and/or competitive intensity,it is argued that the link between service and loyalty intentions will be moderated.

H11. The perceived level of (a) economic turbulence and/or (b) competitive intensityby customers negatively moderate the relationships between the individualservicescape elements of: (i) music, (ii) aroma, (iii) cleanliness, (iv) implicitcommunicators, (v) furnishing, (vi) customer orientation, (vii) credibility, (viii)competence and (ix) staff physical attractiveness; and intentions of customersto be loyal.

Research design and methodologyThe research methodology literature indicates that in order to examine the issuesinvolved in this study, a descriptive quantitative research design was appropriate.Consequently, a survey of restaurant patrons was undertaken. The survey was limitedto dining occasions which involved service by frontline staff and consumption of foodand drink in a formal tabled setting (excluding fast-food, takeaway meals, etc.) Thissector was considered particularly appropriate for this study due to a variety ofreasons, including features such as extended contact with the servicescape, closecustomer-staff contact and economic importance (Hartline and Ferrell, 1996; Harris and

EJM42,3/4

398


10/33

Ogbonna, 2002). The focus of the sampling procedure was on developing a sample ofrestaurant patrons that was roughly proportional in terms of gender, occupation andage distribution. Seven hundred and fifty restaurant patrons were approached and atotal of 278 responses were generated, of which 271 were considered eligible (seven

were incomplete). Using the CASRO method of response rate calculation, this generatesa response rate of 37.1 per cent. The majority of restaurants (79 per cent) patronisedwere defined as fine dining or mid-range type restaurants, including nationwidechains as well as independent establishments. The remainder of restaurants weretypically described as bistro-type establishments (although it should be recognisedthat a wide range of such restaurants exist). The average (mean) spend per customerwas between 15 and 20 (although 12 per cent spent over 30 per head and oneinformant spent over 90 per person).

In an effort to improve content validity, response reliability and response rates, thesurvey was conducted in a manner which closely followed the administration anddesign recommendations of Churchill (1991) and Dillman (1978). Such

recommendations encompassed questionnaire design and layout, survey pilotingand pre-notification, and post-survey follow-up reminders. To test for non-responsebias, a comparison of early and late respondents was conducted on a number of keycharacteristics. No significant differences were found between the two groups(Armstrong and Overton, 1977).

A survey research instrument was developed using measures adapted from extantliterature. Aroma was measured using an item adapted from the study bySpangenberg et al. (1996) on the effects of aroma on customers evaluations andbehaviours towards a servicescape; and an item drawn from Mattila and Wirtzs (2001)

study which measured the effect of aroma on customers approach and impulse buyingbehaviour. Two of the music scale items for this study were adapted from Wakefieldand Bakers (1998) study of customers excitement at a mall and two from Mattila andWirtzs (2001) research on the effect of music stimuli on customers approach andimpulse buying behaviour. The four-, five- and two-item measures (respectively) ofcleanliness, implicit communicators and furnishing were adapted from the study byWakefield and Blodgett (1996), which measured customers approach-avoidancebehaviour towards three leisure service settings.

Scale items for the staff behaviour construct were drawn from the areas of customerorientation and credibility. Specifically, the three items measuring customerorientation were adapted from earlier work by Narver and Slater (1990). Inmeasuring credibility, two scale items were adapted from studies by Sweeney andWyber (2002), who test customers perception of service quality and merchandisequality and one item from dAstous (2000), who sought empirical verification of

customers perceived irritability with components of the shopping environment. Thefive scale items for competence were adapted from studies by Sharma and Stafford(2000) which measure the effect of retail atmospherics on customers perceptions ofsalespeople and customer persuasion; and from Sweeney and Wyber (2002), who testcustomers perception of service quality and merchandise quality. The four items forphysical attractiveness were adapted from the study by Ohanian (1990) whose

Expertise, Trustworthiness and Attractiveness Scale was originally developed tomeasure the influence of celebrity endorsers on customers approach behaviour.


intentions

399


11/33

The two items gauging variety-seeking behaviour were adapted from a study byLastovicka and Gardner (1979) on the components of customer involvement. Fouritems from Baker et al. (2002) were adapted to assess the influence of predictorvariables on customers value perceptions and approach-avoidance behaviour. The

two-item perceived economic turbulence scale was developed from an article byMcIntyre and Pencavel (2004) which measured the macroeconomic turbulence effect onreal wage levels and wage structure, while the perceived competitive intensity scaleuses four items adapted from measures of substitutes or switching opportunities in

Jaworski and Kohli (1993). Finally, the six-item scale gauging the intentions ofconsumers to be loyal was adapted from the work of Zeithaml et al. (1996).

After a review of the various types of measurement scales (see Wrenn, 1997), it wasdecided to adopt seven-point Likert-type scoring (Likert, 1932) for all self-reportingitems. Whilst a number of the adopted measures had previously been employed usingfive-point scales, Barnes et al. (1994) argue that the switch to seven-point scales has noimpact on principal components analysis. Seven-point scales as the form ofmeasurement and scoring were used primarily for ease of response and administration(Malhotra, 1993) and for reasons of scale reliability and validity (Churchill and Peter,1984). Responses to the questionnaires were analysed using the SPSS 12 package forWindows. The analysis ranged from univariate and bivariate analyses to moresophisticated multivariate techniques. The following section details the results of theseanalytic procedures.

FindingsIn order to evaluate the developed hypotheses, it was first necessary to constructmeaningful indices of the measures employed. Consequently, principal componentsanalysis with Varimax rotation was conducted of the servicescape, moderating andendogenous variables (see Appendix 1). As expected, these principal components

analyses led to the extraction of multiple dimensions of servicescape, four moderatingfactors and loyalty intentions. Factors were only retained if they possessed aneigenvalue greater than 1, accounted for over 5 per cent of variance and wereconceptually clear and interpretable (Kaiser, 1958; Churchill, 1991).

Prior to the evaluation of the hypotheses, it was considered prudent to establish thereliability and validity of each of the three scales utilised in this study. Reliabilitygauges the degree to which measures are free from error and therefore yield consistentresults (Peter, 1979, p. 6). Slater (1995) argues that the most commonly reportedmeasure of reliability is the Cronbach a and that the usual criterion is 0.7. TheCronbach a coefficient (Cronbach, 1951) test of reliability led to coefficients whichranged from 0.90 to 0.65 (see Table I). The high level of the coefficients and the findingthat deletion of items would reduce the coefficient led to the conclusion that the scales

were acceptably reliable (Peterson, 1994). Table I shows that with the exception of atwo factors (perceived sacrifice and perceived economic turbulence a 0:65), theremaining Cronbach a coefficients exceed the threshold of 0.7. Nevertheless, thecoefficient level of 0.65 is higher than the minimum level of 0.6 suggested by Peterson(1994) and considerably higher than the 0.50 threshold originally advocated byNunnally (1967).

As stated earlier, the questionnaire was piloted, pretested and adjustments made toimprove content validity. However, analyses of the validity of index operationalisation

EJM42,3/4

400


12/33

Item-totalcorrelation

Scale

Factors

Numberofitems

Cronbachsa

Mean

Median

SD

(1)

(2)

(3)

(4)

(5)

(6)

Music

1

4

0.7

3

4.9

0

4.9

2

1.0

8

0.6

1

0.4

1

0.6

7

0.4

1

Aroma

1

2

0.7

5

5.0

0

5.0

0

1.2

8

0.6

0

0.6

0

Cleanliness

1

4

0.7

5

5.2

9

5.3

4

1.0

2

0.6

4

0.6

1

0.5

1

0.4

8

Implicitcommunicators

1

5

0.8

0

4.8

9

4.8

8

1.0

2

0.3

9

0.7

0

0.7

1

0.6

8

0.4

9

Furnishings

1

2

0.7

9

4.7

6

4.7

6

1.3

3

0.6

5

0.6

5

Customerorientation

1

3

0.9

0

4.9

5

4.9

6

1.2

8

0.8

1

0.8

2

0.7

5

Credibility

1

3

0.7

6

5.0

2

5.0

8

1.1

4

0.6

5

0.6

3

0.5

0

Competence

1

5

0.7

8

4.9

6

4.9

2

1.0

9

0.6

8

0.6

4

0.3

9

0.5

6

0.5

4

Physicalattractiveness

1

4

0.8

0

3.9

8

4.0

9

1.0

8

0.5

8

0.6

6

0.7

0

0.5

2

Variety-seekingbehaviou

r

1

2

0.7

1

4.2

6

4.2

6

1.3

8

0.5

5

0.5

5

Perceivedsacrifice

1

4

0.6

5

2.9

1

2.9

3

1.1

2

0.4

5

0.4

8

0.4

2

0.3

7

Perceivedeconomicturbulence

1

2

0.6

5

4.1

9

4.1

9

1.1

9

0.4

8

0.4

8

Perceivedcompetitiveintensity

1

4

0.8

1

4.9

0

4.8

1

1.1

4

0.5

8

0.6

9

0.5

1

0.7

3

Loyaltyintentions

1

6

0.8

9

4.7

8

4.7

6

1.1

6

0.7

5

0.7

8

0.8

0

0.6

3

0.7

1

0.5

7

Table I.Summary of scale

statistics


intentions

401


13/33

and discriminant validity were also undertaken. The validation of the indexoperationalisation test of correlating items to the scale indicated that the correlationswere high and in the expected direction, suggesting an acceptable level of convergentvalidity (see Table I). The level of discriminant validity was gauged via the adoption of

the test recommended by Gaski (1986). This test indicates evidence of discriminantvalidity in that the correlations between factors are lower than either scales acoefficient. Consequently, it was concluded that each of the scales used in the study areacceptably reliable and valid.

Testing of hypothesesA three-stage analytical procedure was conducted with the intention of evaluating thedata by way of:

(1) zero-order correlation analysis;

(2) multiple regression analysis; and

(3) moderated multiple regression analysis.

Correlation analysisZero-order data exploration revealed loyalty intentions to be related to all nineservicescape variables (see Table II). With values ranging from 0.380 to 0.616, theobtained correlation values are in the expected direction (positive), of the anticipatedmagnitude (equal to or greater than 0.10) and strongly related to loyalty intentions(p , 0:01). The variable with the largest correlation with loyalty intentions wascleanliness. This finding is consistent with expectations due to the central role a cleanservicescape plays in restaurant selection. Other major associations with loyaltyintentions are made by the elements of implicit communicators, customer orientation

and competence with coefficients of 0.569, 0.569 and 0.567, respectively. The largecorrelation between these elements and loyalty intentions is demonstrative of theimportance of aesthetic appeal and staff performance in this servicescape, particularlybecause of the amount of time spent by customers in the setting and their high contactintensity with service staff.

Furthermore, in order to ascertain the specific amount of variance shared by eachservicescape variable and loyalty intentions, the coefficient of determination (COD)was calculated. The COD value was obtained by squaring individual r values andsubsequently multiplying them by 100 in order to convert individual results to apercentage of variance. The percentages of variance shared by individual servicescapevariables and loyalty intentions confirm the associative relationships and are asfollows: music (14.44 per cent), aroma (18.40 per cent), cleanliness (37.95 per cent),

implicit communicators (32.38 per cent), furnishing (14.52 per cent), customerorientation (32.38 per cent), credibility (26.32 per cent), competence (32.15 per cent) andphysical attractiveness (18.06 per cent).

Based on the results of zero-order correlation analysis, tentative evidence is found insupport of H1-H9 which represent proposed linear relationships. However, althoughobtained values are highly statistically significant, sole dependence on zero-ordercorrelation analysis limits the ability of the researcher to ascertain the impact of otherexogenous variables in any one zero-order relationship. To overcome this limitation

EJM42,3/4

402


14/33

1

2

3

4

5

6

7

8

9

1.

Music

1.0

0

2.

Aroma

0.5

13**

1.0

0

3.

Cleanliness

0.4

52**

0.5

05**

1.00

4.

Implicitcommunicato

rs

0.4

17**

0.4

50**

0.647**

1.0

0

5.

Furnishing

0.1

78**

0.1

86**

0.368**

0.3

90**

1.0

0

6.

Customerorientation

0.4

44**

0.4

62**

0.579**

0.4

93**

0.3

00**

1.0

0

7.

Credibility

0.3

38**

0.4

59**

0.589**

0.4

13**

0.2

93**

0.6

04**

1.0

0

8.

Competence

0.4

54**

0.4

31**

0.590**

0.5

00**

0.3

04**

0.7

43**

0.6

24**

1.0

0

9.

Physicalattractiveness

0.0

41

0.1

08

0.262**

0.3

21**

0.1

70**

0.2

89**

0.2

45**

0.3

38**

1.0

0

10.

Loyaltyintentions

0.3

80**

0.4

29**

0.616**

0.5

69**

0.3

81**

0.5

69**

0.5

13**

0.5

67**

0.4

25**

Note:

*Correlationissig

nificantatthe0.0

1level(two-tailed)

Table II.Correlation matrix


intentions

403


15/33

therefore, multiple regression analysis was conducted. The result of this multivariateanalysis follows.

Multiple regression analysisMultiple regression analysis was conducted to obtain the entirety of effects betweenservicescape variables and loyalty intentions. This was deemed necessary to overcomethe limitation of correlation analysis. Results show that of the nine possible linearrelationships, only five servicescape variables were found to be significantly directlyrelated to loyalty intentions via the enter method of multiple regression analysis. Itshould be noted that in keeping with established requirements governing theacceptability of regression results, tests for linearity, homoscedasticity, normality andmulticollinearity were performed at each stage of regression. The results of these testsindicate that all the regression assumptions were upheld. Particularly,multicollinearity results reveal that the existing relationship among servicescapevariables is not characterised by large correlations. This is reflected in the VIF scores

and tolerance values, which are less than 10 and greater than 0.10, respectively. Thisresult implies that multicollinearity levels in the regression model are low enough toleave the standardised regression coefficient (b) estimates unaffected, indicating thatthe effect of a particular servicescape variable cannot be explained by one or more ofthe other servicescape variables under study.

Results further reveal high values for R2 (0.543) and adjusted R2 (0.528). Theadjusted R2 value is believed to be the more reliable measure because it provides abetter estimate of the true population value (Pallant, 2001). Therefore, expressed as apercentage, the obtained adjusted R2 value signifies that the regression model explainsnearly 53 per cent of variance in loyalty intentions. In addition, after testing the nullhypothesis that multiple R in the population equals 0, the model attains very highstatistical significance with a value of 0.000 (p , 0:0005). Therefore, the null

hypothesis is rejected.Table III depicts the relationships between individual servicescape variables and

loyalty intentions, while controlling for the effects of other servicescape variables. Asexpected, all the regression coefficients are in the expected direction (positive).However, only five of the nine proposed linear relationships are revealed to bestatistically significant. Three of the elements significant in their linear relationship

Relationship b coefficient t-statistic

Music! LI 0.048 0.904Aroma! LI 0.065 1.199Cleanliness! LI 0.211 3.240 * * *

Implicit communicators! LI 0.138 2.330 * *Furnishing! LI 0.109 2.337 * *

Customer orientation! LI 0.134 1.994 *

Credibility! LI 0.076 1.286Competence! LI 0.070 1.006Physical attractiveness! LI 0.217 4.697 * * *

Notes : *Statistical significance at the 5 per cent level; * *statistical significance at the 1 per cent level;* * *statistical significance at the 0.1 per cent level

Table III.Summary of multipleregression results

EJM42,3/4

404


16/33

with loyalty intentions represent elements related to organisational climate while theremaining two significant elements relate to elements pertaining to organisationalculture. The four non-significant elements of music, aroma, credibility and competencewith t-statistic values of 0.904, 1.199, 1.286 and 1.006, respectively, do not attain

statistical significance.After the variance explained by all other elements in the model was controlled for,

the variable which made the strongest unique contribution to the prediction of loyaltyintentions was physical attractiveness with a b value of 0.217, very high t-statisticvalue of 4.697 and statistically significant contribution to the equation at the 0.1 percent level. This result indicates that customers strongly consider the appearance ofstaff in the servicescape to be a major determinant of their approach-avoidancebehaviour. Consequently, statistical evidence suggests that the physical attractivenessof service staff is a factor that restaurant managers should seriously consider whenmaking their staff hiring and dress decisions for this particular servicescape. Thevariable which made the second strongest unique contribution was cleanliness. Thisresult is congruent with expectations due to the nature of the servicescape under study.With a bvalue of 0.211, t-statistic value of 3.240 and statistical significance at the 0.1per cent level, cleanliness is found to be highly significant in the explanation of loyaltyintentions.

Based on the results of multiple regression analysis presented above, furthersupport is found for H3, H4, H5, H6and H9, while no evidence supports H1, H2, H7and H8 which were previously supported under correlation analysis. The lack ofsupport for hypotheses relating to music, aroma, credibility and competence is possiblydue to the aggregation procedure employed in this study where different types ofrestaurants were assessed simultaneously; so as to identify the servicescape variablesrelevant to all types of restaurants. In other words, a diverse range of restaurantsformed the scope of restaurant analysis in this study.

Owing to the nature of the conceptualised model and the resulting formulatedhypotheses which propose the interaction effects of a number of personal andenvironmental factors, there was a strong need for moderated multiple regressionanalysis to be conducted; in order to obtain the totality of effects concerning not justthe individual and collective influences of servicescape variables on loyalty intentions,but also the interaction effects of the moderating factors on the servicescape-loyaltyintentions relationship. It was believed that due to the measure of interaction effects,via the employment of moderated multiple regression analysis, provides theopportunity to uncover associations that simple regression analysis has erroneouslypreviously indicated to be unrelated.

Moderated multiple regression analysis (MMR)

MMR is particularly suited to the current study because it enables the slope of one ormore of the independent variables to vary across values of the moderator variable (Hairet al., 1998), thereby facilitating the investigation of an extensive range of relationshipsand function forms (Goode and Harris, 2007). This form of analysis is central tomarketing (Irwin and McClelland, 2001, p. 100) because it permits the multiplerelationships between the endogenous variable and exogenous variables to depend onthe levels of the other exogenous variables in the study. As pertains to this researcheffort, such an analysis not only establishes the strength of the relationship between


intentions

405


17/33

the servicescape variables and loyalty intentions, but also augments the additive

multiple regression model (Irwin and McClelland, 2001, p. 101) by allowing for the

effects of the proposed moderating personal and environmental factors to be properly

gauged and their explanatory power adequately established.

The results of MMR are presented in Table IV. However, prior to the discussion ofthese findings, it is prudent to assess the extent to which the MMR equation adheres to

the assumptions of the technique. Consequently, the data was subjected to analysis in

order to gauge the level of rigour to the overall analysis procedure. The test selected for

the assessment of linearity was the Chow test (a more advanced version of the F-test;Chow, 1960), which aims to evaluate the nature of the data by analysing the residual

errors from the model. The strength of the Chow test is its ability to measure the

stability of regression coefficients over two sub-periods of the data by comparing the

sums of squared residuals over the two periods, thereby allowing the researcher

ascertain if the stability of the model as a whole remains unaffected even after the data

is split in half. The result of this test clearly indicates that the model remains stable

even after splitting and thus supports previous statistical findings that the condition oflinearity was not violated.

Nature of association bcoefficient t-statistic

Direct linkagesAroma 0.103 2.119 * *

Cleanliness 0.186 3.042 * * *

Implicit communicators 0.106 1.862 *

Furnishing 0.164 3.638 * * *

Customer orientation 0.295 5.188 * * *

Physical attractiveness 0.228 5.167 * * *

Moderated linkages via personal factorsVariety-seeking behaviour competence 0.095 2.195 * *

Variety-seeking behaviour physical attractiveness 20.142 23.312 * * *

Perceived sacrifice music 0.112 2.310 * *

Perceived sacrifice aroma 20.177 23.365 * * *

Perceived sacrifice customer orientation 20.141 22.316 * *

Perceived sacrifice credibility 0.110 1.928 *

Moderated linkages via environmental factorsEconomic turbulence implicit communicators 0.100 2.235 * *

Economic turbulence physical attractiveness 20.093 22.193 * *

Competitive intensity customer orientation 20.080 21.656a

Competitive intensity physical attractiveness 0.124 2.796 * *

Diagnostic statisticsAdjusted R2 0.585Number of observations 271

F statistic (df1 18, df2 252) 22.182

Notes: *Statistical significance at the 5 per cent level; * *statistical significance at the 1 per cent level;* * *statistical significance at the 0.1 per cent level; atentatively accepted at the 10 per cent levelcongruent to similar exploratory loyalty intentions studies (see Zeithaml et al., 1996)

Table IV.Summary of significantmoderated multipleregression testing results

EJM42,3/4

406


18/33

The test for homoscedasticity was conducted via a covariance matrix estimator calledWhites test (White, 1980). Under this form of testing, the consistent estimator wascompared to the usual covariance matrix estimator in order to detect any significantdivergence in their values. An inspection of the results immediately confirms that both

estimators are about the same, thus signifying that the data meets the condition ofhomoscedasticity.

Normality was tested with the aid of the Jarque-Bera test (Jarque and Bera, 1980).This asymptotic test measures the null hypothesis that the data are normallydistributed. The result of this test revealed that the data easily indicated a normaldistribution. This means that the null hypothesis can be accepted.

Multicollinearity testing was conducted by means of the Maddala test (Maddala,1977). This test is a pooling method designed to obtain more efficient estimates for theparameters of interest by regressing the exogenous variables against the endogenousvariable to serve as a benchmark model. Thereafter, each exogenous variable isregressed against the other exogenous variables resulting in an equal number ofregressions to the number of exogenous variables. The results for the adjusted R2

values for the exogenous regressions are then compared to the previously obtainedadjusted R2 value of the base model and if greater, then there is the suggestion ofmulticollinearity. The results of the Maddala econometric test indicate that of the 16examined relationships, none produced an adjusted R2 value greater than that of thebase model (0.585). Therefore, the assumption of multicollinearity was not violated.

Based on the results of rigorous statistical and econometric testing, the model isshown to conform fully to standard requirements concerning the non-violation of theregression assumptions.

Results of MMR analysis reveal a very high R2 value of 0.613 and correspondinglyhigh adjusted R2 value of 0.585 (see Table V). In percentage terms, the adjusted R2

value obtained signifies that the moderated regression model explains nearly 59 per

cent of variance in loyalty intentions. This percentage value of explained variance is aconsiderable increase from the 53 per cent value explained by the multiple regressionmodel, thus further lending credence to the use of MMR in testing the complex natureof relationships proposed by the conceptual model. Additionally, to confirm the overallexplanatory power of the model and determine its level of robustness, the F-test wasperformed. Neter et al. (1989) posit that the F-test is the ratio of explained tounexplained variance, with an F statistic greater than the critical value of F beingstatistically significant. The F statistic for the current model was found to be highlystatistically significant, with a value of 9.000. This result indicates that all theexplanatory variables in the model have a significantly greater explanatory powerthan the overall mean of loyalty intentions. Furthermore, with a value of 0.000(p , 0:005), the null hypothesis that multiple R in the population equals zero is easily

rejected.The results further demonstrate that 16 relationships are statistically significant.

These 16 relationships are constituted of six linear relationships between servicescapevariables and loyalty intentions, six moderated relationships via personal factors andfour moderated relationships via environmental factors.

Based on the above tabulation of significant relationships, an interestingobservation is that the servicescape element of aroma which was previously foundnon-significant via multiple regression analysis, is shown to be significant when the


intentions

407


19/33

totality of effects is considered. Similarly, the elements of music, competence andcredibility which were non-significant in their linear relationships with loyaltyintentions are found significant by way of moderating factors.

More formally, support is found for each of the hypothesised links if found. Linearassociations are found between aroma (H2), cleanliness (H3), implicit communicators(H4 ), furnishing (H5), customer orientation (H6), physical attractiveness of staff (H9)and intentions to be loyal. Of those servicescape dimensions found not to exhibit linearlinkages with loyalty intentions, non-linear associations are found music (H1),credibility (H7 ) and competence (H8 ) as well as for aroma ( H2 ), implicit

communicators (H4 ), furnishing (H5 ), customer orientation (H6 ) and physicalattractiveness of staff (H9). These in turn demonstrate partial support for H10a-H11b(see Table V).

Conclusions and implicationsThis study represents a concerted attempt to advance a modified theory-basedframework for modelling the role of physical surroundings in the restaurantconsumption setting (see Figure 1). The framework developed builds on a wide range

Relationship typeVariable Linear Non-linear Hypothesis testing

H1 Music No Yes Not supported

H2 Aroma Yes Yes SupportedH3 Cleanliness Yes No SupportedH4 Implicit communicators Yes Yes SupportedH5 Furnishing Yes Yes SupportedH6 Customer orientation Yes Yes SupportedH7 Credibility No Yes Not supportedH8 Competence No Yes Not supportedH9 Physical attractiveness Yes Yes Supported

Moderating personal factorsH10a Variety-seeking behaviour (two relationships) Partially supported

(viii) Competence(ix) Physical attractiveness

H10b Perceived sacrifice (four relationships) Partially supported(i) Music(ii) Aroma(vi) Customer orientation

(vii) Credibility

Moderating environmental factorsH11a Perceived economic turbulence (tworelationships)

Partially supported

(iv) Implicit communicators(ix) Physical attractiveness

H11b Perceived competitive intensity (tworelationships)

Partially supported

(vi) Customer orientation

(ix) Physical attractiveness

Table V.Summary of hypotheses

testing

EJM42,3/4

408


20/33

of literature (including environmental psychology, strategic marketing and consumerbehaviour) to supply an operationalized and tested multifaceted model and measure ofkey servicescape variables. In addition to the methodological contribution made via thetesting of measures of multiple elements of servicescape, a contribution is made

through the supply (and then testing) of a model of the key physical andemployee-oriented aspects of servicescape. In this regard, the current study can beviewed as a direct response to the many commentators on the servicescape that havecalled for additional modelling and empirical testing of developed models (see earlier).Moreover, this study highlights that in addition to the physical surroundings ofexchange (measured in H1-H5 ), a number of staff-related factors are also linked toconsumers subsequent loyalty intentions and constitute key components of thebroader servicescape. Indeed, the inclusion of these variables leads to nearly 59 percent of variations in loyalty being explained by the variables studied. In this regard,the current study builds on the work and recommendations of Tombs andMcColl-Kennedy (2003) who advocate the broadening of the servicescape concept toinclude social variables (their particular focus being on customers), to the extent thatstudies consider what they label the social-servicescape. As such, the current studycan be viewed as supplying empirical evidence n support of this broadening notion.

Unlike the vast majority of studies in the area, this study supplies empiricalevidence for the theoretical bases of servicescapes. Furthermore, dissimilar to otherservicescape studies, this study fully incorporated the moderating effects of personaland environmental factors which study findings proved to be key determinantsunderlying the display of loyalty intentions by customers. Although theMehrabian-Russell PAD model (Mehrabian and Russell, 1974) forwards theimportance of mediators in all environmental stimuli approach-avoidanceresponse relationships (Bellenger et al., 1977; Ward et al., 1992; Wakefield andBlodgett, 1996; Wakefield and Baker, 1998; Lucas, 2000), no servicescape study has

attempted to vary the assessment of servicescapes significantly by integratingmoderating effects related to both customers personalities (variety-seeking behaviourand perceived sacrifice) and the macro environment (perceived economic turbulenceand perceived competitive intensity). The integration of these moderators was a keycomponent to the overall success of the study because their insertion necessitated theuse of moderated multiple regression analysis which revealed a number of interestingassociations. For instance, preliminary data exploration via multiple regressionanalysis indicated that music, aroma, competence and credibility were unrelated toloyalty intentions. However, analysis via moderated multiple regression not onlyrevealed a linear relationship between aroma and loyalty intentions, but also,non-linear relationships between the individual elements of music, competence andcredibility to loyalty intentions. Thus, the results demonstrate the intricate web of

associations between concepts; associations which would have remained hidden hadmoderating effects not been measured. Further, the integration of moderating factorsenabled the presentation of a composite model of servicescapes against which futureempirical findings can be compared.

While both empirical and conceptual servicescape studies have tended to focus onsingle servicescape elements such as music (see Milliman, 1982, 1986; Areni and Kim,1993; Yalch and Spangenberg, 1990; Sweeney and Wyber, 2002), aroma (see Hirsch,1995; Spangenberg et al., 1996; Fiore et al., 2000) and light (see Areni and Kim, 1994;


intentions

409


21/33

Summers and Hebert, 2001), only very few studies have attempted to incorporate theinfluence of more than one servicescape element (see Chebat et al., 1993; Wakefield andBlodgett, 1996; Mattila and Wirtz, 2001). The current research endeavour differssignificantly from these and other previous servicescape studies by positioning nine

servicescape variables and finding empirical support for them all either via linear ornon-linear associations to loyalty intentions. Considered in concert with the fourmoderating personal and environmental factors included in the conceptual model, thepresent study far exceeds the characteristic servicescape study pace. The resultanteffects of the incorporation of a total of 13 elements in the same study are: thesignificant increase in the knowledge base; a targeted response to calls for furtherempirical multi-variable testing; and an accomplishment of the objective of obtaining aglobal configuration of servicescapes.

This study adopted a three-stage analytical process in testing the hypothesisedlinkages between variables. These stages represented:

(1) Correlation analysis to detect potential underlying relationships.

(2) Multiple regression analysis for initial proposition evaluation.(3) Moderated multiple regression analysis for interaction effects estimation.

Additionally, to bolster obtained statistical results, econometric analyses wereconducted. The use of multiple methods of measurement in evaluating both linear andnon-linear linkages ensured a considerable reduction of the potential problems withshared-method variance and thus advanced far from spurious results on which otherservicescape studies can build.

Notwithstanding the limitations of the study, the significant findings allow for theadvancement of implications for practitioners. The obvious implication of the findingsof this empirical study is that restaurant managers must recognise the overwhelmingbenefits to be derived from paying particular attention to the servicescape variables.

These, for clarity and ease of presentation, were grouped under the broad categories oforganisational climate and organisational culture. As these elements lie largely withinthe area of management control and can be manipulated to affect customers loyaltyintentions positively, management should actively promote the desired performance ofservice staff and involve a more systematic approach to cue placement rather thancontinue to rely on the haphazard selection and placement of the tangible andintangible cues that litter the servicescape.

All the moderators in the conceptual model were found to be strong facilitators ofrelationships in the servicescape. Therefore, service managers should take these intoconsideration when designing the servicescape and implementing the overallmarketing function. For instance, where variety-seeking behaviour is a significantdeterminant of customers behaviour, theme nights could be organised; to reduce

perceived sacrifice levels, crowd control measures such as table reservation could makea significant difference; to counter high levels of perceived economic turbulence,projections of value are required from the decor of the servicescape to the quality ofmeals as this would address customers need to stretch their disposable incomes;finally, in cases where customers perceive a high level of competitive intensity to exist,restaurant managers may want to consider a highly differentiated service strategyakin to that adopted by Hard Rock Cafe, whose composite service offering cannot bereadily matched by competitors.

EJM42,3/4

410


22/33

A number of limitations related to this research are evident and thus the conclusionsand implications drawn from study findings should be interpreted in light of these.First, as is typical of most exploratory studies, this study was cross-sectional in nature.Therefore, in testing the hypothesised relationships, inferences could only be made in

associative terms and not in terms of cause and effect. Second, significant difficultlywas encountered in the construction of scale items, development of hypotheses andinterpretation of findings; given the paucity of servicescape literature and inadequacyof empirical evidence of direct relevance to servicescape study. Finally, data collectionefforts were focused on customers. However, the drawback of this focus was that asdata was not simultaneously collected from customer-contact staff and/or servicemanagers, the study was unable to detect potential gaps between customersperceptions of the received service and the perception service organisations wereattempting to project.

A number of other research avenues arise from the findings of this study. While thisstudy was aggregate in nature and thus identified elements of importance to all typesof restaurants, other studies may focus on specific classes of restaurants in order for

elements significant to particular restaurants to be identified. Similarly, other studiescould make attempts to apply this global configuration to other service settings likehotels, airports, banks and hospitals and to other countries. Such studies could benefitfrom the incorporation of a wide range of literatures. However, environmentalpsychology seems an especially fruitful source of theory on which subsequent studiescould profitably build. As the elements studied were found to link significantly toloyalty intentions through linear and/or non-linear relationships, it is proposed thatservicescape scholars attempt to test a wide variety of elements (such as perceivedvalue, trust and satisfaction) for which theoretical or empirical support suggests eithertheir relationship to loyalty intentions, or to the other stages of loyalty particularlyaction loyalty. To this end, studies of a longitudinal nature are required to establish the

causes for linkages between loyalty intentions and action loyalty. This servicescapestudy focused its data collection efforts on customers because their perceptions of theservice experience were believed to determine their loyalty intentions. However, otherservicescape studies can make attempts to simultaneously incorporate responses fromcustomer-contact staff and/or service managers so as to detect potential gaps betweencustomers perceptions and the image which service organisations are trying to project.It is believed that such detection would not only enable further development ofservicescape literature, but also ensure that a more comprehensive recommendationcan be made to service organisations on how best to meet identified customer needsmore effectively and thus improve the overall efficiency and marketability of theirservice operations.

ReferencesAlvesson, M. (2002), Understanding Organisational Culture, Sage Publications, London.

Appiah-Adu, K., Fyall, A. and Singh, S. (2000), Marketing culture and customer retention in thetourism industry, The Service Industries Journal, Vol. 20 No. 2, pp. 95-113.

Areni, C.S. and Kim, D. (1993), The influence of background music on shopping behaviour:classical versus top-40 music in a wine store, in McAlister, L. and Rothschild, M.L. (Eds),

Advances in Consumer Research, Association for Consumer Research, Provo, UT,pp. 336-40.


intentions

411


23/33

Areni, C.S. and Kim, D. (1994), The influence of in-store lighting on consumers examination of

merchandise in a wine store, International Journal of Research in Marketing, Vol. 11,

pp. 117-25.

Armstrong, J.S. and Overton, T.S. (1977), Estimating non-response bias in mail surveys,

Journal of Marketing Research, Vol. 14, August, pp. 396-403.

Arneill, A.B. and Devlin, A.S. (2002), Perceived quality of care: the influence of the waiting room

environment, Journal of Environmental Psychology, Vol. 22, pp. 345-60.

Arnold, S.J., Handelman, J. and Tigert, D.J. (1996), Organisational legitimacy and retail store

patronage, Journal of Business Research, Vol. 35 No. 3, pp. 229-39.

Aubert-Gamet, V. (1997), Twisting servicescapes: diversion of the physical environment in a

re-appropriation process, International Journal of Service Industry Management, Vol. 8

No. 1, pp. 26-41.

Baker, J. (1987), The role of the environment in marketing services: the consumer perspective,

in Czepiel, J., Congram, C.A. and Shanahan, J. (Eds), The Services Challenge: Integrating for

Competitive Advantage, American Marketing Association, Chicago, IL, pp. 79-84.

Baker, J., Levy, M. and Grewal, D. (1992), An experimental approach to making retail store

environmental decisions, Journal of Retailing, Vol. 68 No. 4, pp. 445-60.

Baker, J., Parasuraman, A., Grewal, D. and Voss, G.B. (2002), The influence of multiple store

environment cues on perceived merchandise value and patronage intentions, Journal of

Marketing, Vol. 66 No. 2, pp. 120-41.

Barnes, J., Audhem, P. and Gilbert, F. (1994), Number of factors obtained by chance: a situation

study, in Wilson, K.J. and Black, W. (Eds), Developments in Marketing Science, Vol. 16,

Academy of Marketing Science, Nashville, TN.

Bellenger, D.N., Robertson, D.H. and Greenberg, B.A. (1977), Shopping centre patronage

motives, Journal of Retailing, Vol. 53 No. 2, pp. 29-38.

Berman, B. and Evans, J.R. (1995), Retail Management: A Strategic Approach, 6th ed.,

Prentice-Hall, Englewood Cliffs, NJ.

Berry, L. and Parasuraman, A. (1991), Marketing Services: Competing through Quality, The Free

Press, New York, NY.

Berta, D. (2005), Focus on restaurant cleanliness may spruce up sales figures, Nations

Restaurant News, Vol. 39 No. 30, p. 122.

Bitner, M. (1992), Servicescapes: the impact of physical surroundings on customers and

employees, Journal of Marketing, Vol. 56 No. 2, pp. 57-71.

Bogner, W.C., Thomas, H. and McGee, J. (1999), Competence and competitive advantage:

towards a dynamic model, British Journal of Management, Vol. 10 No. 4, pp. 275-90.

Bone, P.F. and Ellen, P.S. (1999), Scents in the marketplace: explaining a fraction of olfaction,

Journal of Retailing, Vol. 75 No. 2, pp. 243-62.Booms, B.H. and Bitner, M.J. (1982), Marketing services by managing the environment, Cornell

Hotel & Restaurant Administration Quarterly, Vol. 23, May, pp. 35-9.

Brown, T.J., Mowen, J.C., Donavan, D.T. and Licata, J.W. (2002), The customer orientation of

service workers: personality trait effects on self- and supervisor performance ratings,

Journal of Marketing Research, Vol. 39 No. 1, pp. 110-19.

Brunner, G.C. II (1990), Music, mood and marketing, Journal of Marketing, Vol. 54 No. 4,

pp. 94-104.

EJM42,3/4

412


24/33

Buchanan, D. and Huczynski, A. (1997), Organisational Behaviour, Prentice-Hall, Hemel

Hempstead.

Butcher, K., Sparks, B. and OCallaghan, F. (2002), Effect of social influence on repurchase

intentions, Journal of Services Marketing, Vol. 16 No. 6, pp. 503-14.

Cassidy, T. (1997), Environmental Psychology: Behaviour and Experience in Context, Psychology

Press, Hove.

Chebat, J.C. and Dube, L. (2000), Evolution and challenges facing retail atmospherics:

the apprentice sorcerer is dying, Journal of Business Research, Vol. 49 No. 2, pp. 89-90.

Chebat, J.C. and Michon, R. (2003), Impact of ambient odours on mall shoppers emotions,

cognition, and spending: a test of competitive causal theories, Journal of Business

Research, Vol. 56, pp. 529-39.

Chebat, J.C., Gelinas-Chebat, C. and Filiatrault, P. (1993), Interactive effects of music and visual

cues on time perception: an application to waiting lines in banks, Perceptual and Motor

Skills, Vol. 77, pp. 995-1020.

Chow, G.C. (1960), Tests for equality between sets of coefficients in two linear regressions,Econometrica, Vol. 28, July, pp. 591-605.

Churchill, G.A. (1991), Marketing Research: Methodological Foundations, McGraw-Hill, London.

Churchill, G.A. and Peter, J.P. (1984), Research design effect on the reliability of rating scales:

a meta analysis, Journal of Marketing Research, Vol. 21 No. 4, pp. 360-75.

Cronbach, L.J. (1951), Coefficient alpha and the internal structure of tests, Psychometrica,

Vol. 16, pp. 297-334.

Cronin, J.J. Jr (2003), Looking back to see forward in services marketing: some ideas to consider,

Managing Service Quality, Vol. 13 No. 5, pp. 332-7.

dAstous, A. (2000), Irritating aspects of the shopping environment, Journal of Business

Research, Vol. 42 No. 2, pp. 149-56.

Deshpande, R. and Webster, F. (1989), Organisational culture and marketing: defining the

research agenda, Journal of Marketing, Vol. 53 No. 1, pp. 3-15.

Deshpande, R., Farley, J.U. and Webster, F.E. Jr (1993), Corporate culture, customer orientation

and innovativeness in Japanese firms: a quadrad analysis, Journal of Marketing, Vol. 57

No. 1, pp. 23-37.

Dillman, D.A. (1978), Mail and Telephone Surveys: The Total Design Method, Wiley, New York,

NY.

Donovan, R.J. and Rossiter, J.R. (1982), Store atmosphere: an environmental psychology

approach, Journal of Retailing, Vol. 58 No. 1, pp. 34-57.

Dube, L., Chebat, J. and Morin, S. (1995), The effects of background music on consumers desire

to affiliate in buyer-seller interactions,Psychology and Marketing, Vol. 12No. 4,pp. 305-19.

Ercel, G. (1998), The Turkish case in global economic turbulence, paper presented at the

Deutsche Bank Emerging Market Conference, October.

Everett, P.B., Pieters, R.G.M. and Titus, P.A. (1994), The consumer-environment interaction:

an introduction to the Special Issue, International Journal of Research in Marketing,

Vol. 11 No. 2, pp. 97-105.

Ezeh, C. and Harris, L.C. (2007), Servicescape research: a review and a research agenda,

Marketing Review, January.


intentions

413


25/33

Fiore, A., Yah, X. and Yoh, E. (2000), Effects of product display and environmental fragrancing

on approach responses and pleasurable experiences, Psychology and Marketing, Vol. 17

No. 1, pp. 27-54.

Fitzsimmons, B. (2003), Restroom cleanliness: the last detail, Restaurant Hospitality, Vol. 87

No. 10, pp. 100-2.

Foxall, G.R. and Greenley, G.E. (1999), Consumers emotional responses to service

environments, Journal of Business Research, Vol. 46, pp. 149-58.

Foxall, G.R. and Yani-de-Soriano, M.M. (2005), Situational influence on consumers attitude and

behaviour, Journal of Business Research, Vol. 58, pp. 518-25.

Gaski, J.F. (1986), Interrelations among a channel entitys power sources: impact of the exercise

of reward and coercion on expert, referent and legitimate power sources, Journal of

Marketing Research, Vol. 23, pp. 62-77.

Gifford, R. (1988), Light, decor, arousal, comfort and communication, Journal of Environmental

Psychology, Vol. 8, pp. 177-89.

Goldsmith, R.E., Lafferty, B.A. and Newell, S.J. (2000a), The impact of corporate credibility and

celebrity credibility on consumer reaction to advertisements and brands, Journal of

Advertising, Vol. 29 No. 3, pp. 43-54.

Goldsmith, R.E., Lafferty, B.A. and Newell, S.J. (2000b), The influence of corporate credibility on

consumer attitudes and purchase intent, Corporate Reputation Review, Vol. 3 No. 4,

pp. 304-18.

Goode, M.M.H. and Harris, L.C. (2007), Online behavioral intentions: an empirical investigation

of antecedents and moderators, European Journal of Marketing, Vol. 41 Nos 5/6,

pp. 512-36.

Gremler, D.D. and Brown, S.W. (1996), Service loyalty: its nature, importance and implications,

in Edvardsson, B., Brown, S.W., Johnston, R. and Scheuing, E. (Eds), Advancing Service

Quality: A Global Perspective, International Service Quality Association, New York, NY,

pp. 171-80.

Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1998), Multivariate Data Analysis,

5th ed., Prentice-Hall, Englewood Cliffs, NJ.

Hallett, V. and Silver, M. (2004), Scents and sensibility, US News & World Report, Vol. 137

No. 18, pp. D2-D3.

Harris, L.C. (1998), Cultural domination: the key to a market oriented culture?, European

Journal of Marketing, Vol. 32 Nos 3/4, pp. 354-73.

Harris, L.C. and Ogbonna, E. (2002), Exploring service sabotage: the antecedents, types and

consequences of frontline, deviant, antiservice behaviors, Journal of Service Research,

Vol. 4 No. 3, pp. 163-83.

Hartline, M.D. and Ferrell, O.C. (1996), The management of customer-contact service employees:an empirical investigation, Journal of Marketing, Vol. 60 No. 4, pp. 52-70.

Hill, C.J., Garner, S.J. and Hanna, M.E. (1989), Selection criteria for professional service

providers, Journal of Services Marketing, Vol. 3 No. 4, pp. 61-9.

Hirsch, A.R. (1991), Nostalgia: a neuropsychiatric understanding, paper presented at the

Annual Meeting of the Association for Consumer Research, Chicago, IL, October.

Hirsch, A.R. (1995), Effects of ambient odours on slot-machine usage in a Las Vegas casino,

Psychology and Marketing, Vol. 12 No. 7, pp. 585-94.

EJM42,3/4

414


26/33

Hoffman, K.D. and Turley, L.W. (2002), Atmospherics, service encounters and consumer

decision making: an integrative perspective, Journal of Marketing Theory and Practice,Vol. 10 No. 3, pp. 33-47.

Holahan, C. (1982), Environmental Psychology, Random House, New York, NY.

Hunter, B.T. (1995), The sales appeals of scents (using synthetic food scents to increase sales),

Consumer Research Magazine, Vol. 18 No. 10, pp. 8-10.

Hutton, J.D. and Richardson, L.D. (1995), Healthscapes: the role of the facility and physical

environment on consumer attitudes, satisfaction, quality assessments and behaviours,

Health Care Management Review, Vol. 20 No. 2, pp. 48-61.

Irwin, J.R. and McClelland, G.H. (2001), Misleading heuristics and moderated multiple

regression models, Journal of Marketing Research, Vol. 38 No. 1, pp. 100-9.

Jacoby, J. and Chestnut, R. (1978), Brand Loyalty Measurement and Management, Wiley, NewYork, NY.

Jarque, C.M. and Bera, A.K. (1980), Efficient test for normality, homoscedasticity and serial

independence of regression residuals, Economics Letters, Vol. 6, pp. 255-9.

Jaworski, B.J. and Kohli, A.K. (1993), Market orientation: antecedents and consequences,

Journal of Marketing, Vol. 57 No. 3, pp. 53-71.

Kaiser, H.F. (1958), The Varimax criterion for analytic rotation in factor analysis,

Psychometrika, Vol. 23 No. 3, pp. 187-200.

Kaya, N. and Erkip, F. (1999), Invasion of personal space under the condition of short-term

crowding: a case study on an automatic teller machine, Journal of EnvironmentalPsychology, Vol. 19 No. 2, pp. 183-9.

Kellaris, J.J. and Kent, R.J. (1993), An exploratory investigation of responses elicited by music

varying in tempo, tonality, and texture, Journal of Consumer Psychology, Vol. 2 No. 4,pp. 381-401.

Kotler, P. (1973), Atmospherics as a marketing tool,Journal of Retailing, Vol. 49 No. 4, pp. 48-64.

Lafferty, B.A. and Goldsmith, R.E. (1999), Corporate credibilitys role in consumers attitudes

and purchase intentions when a high versus a low credibility endorser is used in the ad,

Journal of Business Research, Vol. 44 No. 2, pp. 109-16.

Lastovicka, J.L. and Gardner, D.M. (1979), Component of involvement, in Maloney, J.C. and

Silverman, B. (Eds), Attitude Research Plays for High Stakes, American MarketingAssociation, Chicago, IL, pp. 53-73.

Lee, S.Y. and Brand, J.L. (2005), Effects of control over office workspace on perceptions of the

work environment and work outcomes, Journal of Environmental Psychology, Vol. 25,pp. 323-33.

Levine, J.M. and McBurney, D.H. (1986), The role of olfaction in social perception and behavior,

in Herman, C.P., Zanna, M.P. and Higgins, E.T. (Eds), Physical Appearance, Stigma and

Social Behavior: The Ontario Symposium, Lawrence Erlbaum Associates, Hillsdale, NJ,pp. 179-217.

Likert, R. (1932), A technique for the measurement of attitudes,Archives of Psychology, Vol. 140,pp. 1-55.

Lovelock, C.H. (2001), Services Marketing: People, Technology, Strategy, 4th ed., Prentice-Hall,Englewood Cliffs, NJ.

Lucas, A.F. (2000), The determinants and effects of slot servicescape satisfaction in a Las Vegas

hotel casino, PhD thesis, University of Nevada, Las Vegas, NV.


intentions

415


27/33

McIntyre, F. and Pencavel, J. (2004), The effects of macroeconomic turbulence on real wage

levels and the wage structure: Brazil, 1981-1999, Economic Development & CulturalChange, Vol. 52 No. 4, pp. 681-717.

Maddala, G.S. (1977), Econometrics, McGraw-Hill, New York, NY.

Malhotra, N.K. (1993), Marketing Research: An Applied Orientation, Prentice-Hall, HemelHempstead.

Marinucci, D. (2002), Cleanliness: a key part of service, Tire Business, Vol. 19 No. 20, p. 7.

Mathwick, C., Malhotra, N. and Rigdon, E. (2001), Experiential value: conceptualisation,measurement and application in the catalogue and internet shopping environment,

Journal of Retailing, Vol. 77 No. 1, pp. 39-56.

Mattila, A.S. and Wirtz, J. (2001), Congruency of scent and music as a driver of in-storeevaluations and behaviour, Journal of Retailing, Vol. 77 No. 2, pp. 273-89.

Mehrabian, A. and Russell, J.A. (1974), An Approach to Environmental Psychology, MIT Press,Cambridge, MA.

Milliman, R. (1982), Using background music to affect the behaviour of supermarket shoppers,


Milliman, R. (1986), The influence of background music on the behaviour of restaurant patrons, Journal of Consumer Research, Vol. 13 No. 2, pp. 286-9.

Mintz, N.L. (1956), Effects of esthetic surroundings: prolonged and repeated experience in a

beautiful and an ugly room, Journal of Psychology, Vol. 41, pp. 459-66.

Mitchell, V.W. (1992), Understanding consumers behaviour: can perceived risk theory help?,

Management Decision, Vol. 30 No. 3, pp. 26-31.

Moore, K.D., Van Haitsma, K., Curyto, K. and Saperstein, A. (2003), A pragmatic environmental

psychology: a methatheoretical inquiry into the work of M. Powell Lawton, Journal ofEnvironmental Psychology, Vol. 23, pp. 471-82.

Morrin, M. and Ratneshwar, S. (2003), Does it make sense to use scents to enhance brand

memory?, Journal of Marketing Research, Vol. 40 No. 1, pp. 10-25.

Narver, J.C. and Slater, S.F. (1990), The effect of market orientation on business profitability,


Neter, J., Wasserman, W.

Services Cape And

Documents

Transcript of Services Cape And