The current issue and full text archive of this journal is available at www.emeraldinsight.com/1361-2026.

htm

JFMM 15,4

Atmosphere in fashion stores: do you need to change?

Andrew G. Parsons
Auckland University of Technology, Auckland, New Zealand
Abstract
Purpose The aims of this paper are to establish: a typical womens apparel store environment as a realistic base for measuring the effects of changes; effective environmental stimuli levels; and the effect of repeated exposure on affect. Design/methodology/approach A schema of typical stimuli is developed through literature, observing 212 stores, and surveying 39 womens fashion retail GMs. An experiment (n 489) establishes the set of stimuli and effective levels for creating affect for a womens fashion store. Shoppers (n 62) were repeatedly exposed to combinations of the two sets (industry standard and ideal) to examine whether decay in affect occurs, and whether changes can mitigate this. Findings Interactions between sensory stimuli have a significant effect on fashion shoppers affect for a store. Fashion retailers are less differentiated in their use of sensory stimuli than they could be to achieve the responses they expect. Stagnation from repeated exposure can diminish affect for the store whereas small changes in stimuli levels can revitalise and increase affect. Research limitations/implications A model of sensory stimuli-based relationships with shopper affect should incorporate interactions. Empirically, different stimuli can easily be added or substituted within a dimension to test its effect within a factorial design. A model of fashion store atmosphere is likely to require a mediating influence of repeated exposure. Practical implications It is worthwhile getting the correct package of stimuli for a fashion stores atmosphere. This does not necessarily require wholesale changes; rather small changes in stimuli level can enhance a store. Originality/value The paper presents the only study to use a holistic approach to store atmosphere and base effectiveness measures against the fashion industry norm, and consider the effect of repeated exposure. Keywords Store atmosphere, Retail environment, Shopper response, Consumer behaviour, Marketing environment, Retail management Paper type Research paper

428
Received September 2010 Revised September 2010 Accepted May 2011

Journal of Fashion Marketing and Management Vol. 15 No. 4, 2011 pp. 428-445 r Emerald Group Publishing Limited 1361-2026 DOI 10.1108/13612021111169933

1. Introduction Store atmosphere, created by the layout and environment, is important in the fashion sector of retail (Birtwistle and Shearer, 2001; Newman and Patel, 2004), as it is crucial for success and a key determinant of patronage (Newman and Patel, 2004). Scarpi (2006), for example, finds that in specialty apparel stores the environment can influence the shopper orientation, resulting in different behaviours, while Moye and Giddings (2002) show that components of store atmospheres can have a direct influence on shopping behaviours in a variety of apparel stores. Paulins and Geistfeld (2003) suggest that womens apparel retailers must be prepared to implement changes to decor to remain competitive, but casual observation of stores suggest that many apparel retailers rely heavily on only display fixtures to enhance the visual merchandising without carefully considering the other elements of the store design, such as the walls, flooring, scents, music, temperature, etc. While there are those fashion retailers that have adopted a conceptual direction such as a residential theme, a minimalistic concept, or a thematic approach (Diamond and Diamond, 1996) and have therefore utilised their store environment as part of the selling strategy, many store environments remain as functional and basic areas, with typically plain walls, flooring,

ceilings, lighting, etc. Fashion textbooks nevertheless regularly mention the backdrop, store environment, or atmospherics (e.g. Drake et al., 1992; Oelkers, 2004; Solomon and Rabolt, 2009), as tools that that can be used to help influence shopper behaviour and affect (liking) for the store. If there is a typical functional fashion store environment, does it achieve what could be done in terms of influencing shoppers behaviours and affect for the store, or are there elements that could be changed to enhance the store effectiveness? And, if change does occur, as Paulins and Geistfeld (2003) suggest it needs to, how will shoppers respond? In our study we compare the affect responses of shoppers exposed repeatedly to an industry standard or typical store atmosphere to those exposed to an experimentally derived ideal atmosphere and identify the impact a change in atmospherics has on affect. To establish the typical industry atmosphere we use store observation based on a set of dimensions derived from the literature to then create a survey measurement of retail managers. From the survey and observation we create a schema of typical industry stimuli. To establish the ideal levels we conduct an experiment to identify the levels of stimuli that elicit the greatest affect for the store. Following this a second experiment is conducted where shoppers were repeatedly exposed to a typical store, the ideal store, or a combination changing from one to the other halfway through the repetitions. Thus, the conceptual framework involves developing a construct called industry standard, made up of items that are the typical levels of sensory stimuli found in the marketplace. We also develop a construct called ideal standard, which are the levels of sensory stimuli that are shown to result in the most positive affect for the store. We then take these constructs and test whether repeated exposure to them, or a combination of them, impacts upon the affect shoppers reveal for stores. 2. Background Store atmospherics has generated widespread interest by academics and practitioners since Kotler (1974), Baker (1986), and Bitner (1992) encapsulated the concept of olfactory, aural, tactile, and visual dimensions creating an environment that affects shoppers emotions and behaviour. The Donovan and Rossiter (1982) approachavoidance model focused much of the ensuing attention on behavioural responses to individual or limited interactions of stimuli with less attention paid to affect or emotional responses to the store environment. For example, Areni and Kim (1994) establish a link between the brightness of lighting and increased examination of merchandise, while Milliman (1982) shows that slower music encourages shoppers to spend more time in-store. Similarly, dAstous (2000) suggests that loud music results in less time spent in-store. The problem with these studies is that they treat the stimuli (and resultant response) as isolated events directed solely at behaviour, whereas the store environment provides an holistic cognitive experience. Recent studies have begun to rectify this (e.g. Baker et al., 2002; Kottasz, 2006; Wirtz et al., 2007; Oakes and North, 2008a), but general frameworks persist in examining single-period responses to stimuli (e.g. Fiore and Kim, 2007; Noad and Rogers, 2008; Soars, 2009), continuing support for the single stimulus single-period response approach (e.g. Eroglu et al., 2005; Michon et al., 2005; Ward et al., 2007; Oakes and North, 2008b). Babin and Attaway (2000) asked whether re-patronage is perhaps in part a function of affect created by store atmospherics, with others looking at the enjoyment of the shopping experience (as created by the environment) influencing re-patronage (e.g. Ogle et al., 2004; Hart et al., 2007; Michon et al., 2008;

Atmosphere in fashion stores

429

JFMM 15,4

430

Joseph-Mathews et al., 2009; Seock, 2009). In this light, it is perhaps more appropriate to examine the effects of a store atmosphere or environment when the shopper is presented with all the sensory stimuli, as the single stimulus approach lacks face validity for the shopper. What has also not been examined is whether repeated exposure to the sensory stimuli alters the affective response of shoppers. Is there perhaps a decay effect or is there a compound effect? The approach in the past has also been to vary levels of stimuli to determine likely responses with little regard to industry norms for stimuli and expected responses. In an economic era necessitating caution with expenditure, and a shopper era demanding sustainability, it is pertinent to consider what effect the investment in atmosphere may have over time, and whether changes are warranted. This study seeks to first establish what a typical environment is in a womens apparel store, to use as a realistic base for measuring the effects of any changes to the environment; this is done by determining the common levels of environmental stimuli in the marketplace. Next, the study establishes the set of ideal (in terms of retailer desired shopper responses) environmental stimuli levels. Finally the study establishes whether repeated exposure to the same levels of environmental stimuli affects shopper responses, or whether change stimulates the shopper. 3. Typical levels of environmental stimuli Literature-based set of stimuli A review of the store atmospherics literature established a set of individual sensory stimuli that had been previously studied, ranging from scents (e.g. Bone and Jantrania, 1992; Gulas and Bloch, 1995; Mitchell et al., 1995; Spangenberg et al., 1996; Bone and Ellen, 1999; Morrin and Ratneshwar, 2000; Mattila and Wirtz, 2001; Parsons, 2009) through music (e.g. Grewal and Baker, 1994; Baker et al., 1995; Sharma and Stafford, 2000), colour and lighting (e.g. Crowley, 1993; Babin et al., 2003), to temperatures and surface coverings (e.g. Kerin et al., 1992; Machleit et al., 1994; Sharma and Stafford, 2000; dAstous, 2000). This set was used as a field check for observation of stores (n 212) throughout New Zealand. The literature-based sensory constructs were identified as aural stimuli (music style, volume, tempo, familiarity, and other sounds); olfactory stimuli (scent pleasantness and congruence); tactile stimuli (temperature, flooring); and visual stimuli (lighting sources, level, colours, and fixtures/fitting). These were then overlaid with observed levels of the items; for example, lighting source was observed to include natural light, exposed fluorescent, ceiling panel fluorescent, halogen, non-halogen spotlights, etc. Note that the normal fifth sense of taste is omitted, as it is in most atmospheric studies. This is because taste is almost invariably a functional aspect that is product oriented, rather than a non-utilitarian element that is part of an atmosphere. In womens apparel stores there is no expectation of taste either functional or non-utilitarian being present. Of major concern to any study examining the interactions that may occur between stimuli in creating an atmosphere and having an effect on affect is the possibility that there may be critical components of sensory stimuli that have not been identified by the literature. Omission of these parts may adversely affect the impact the stimuli may have in the experiment leading to false results. Furthermore, in constructing the schema, by not having a clearer idea of what stimuli and stimuli components there are in store atmospheres, the study would open itself to either missing crucial information

because it was not asked for, or creating greater potential for error through interpretation of open-ended answers. Refinement of stimuli set through field observation Field observation was selected as a means of gaining insights and ideas about what other sensory stimuli components may exist in creating store atmospheres. Boote and Mathews (1999) suggest observation is an appropriate methodology when at least one of four criteria is met: (1) (2) (3) (4) the phenomenon is easily observable; the phenomenon is a social process or a mass activity; the phenomenon occurs at a subconscious (sic) level; and the objects under investigation are either unable or unwilling to communicate directly with the researcher.

Atmosphere in fashion stores

431

Sensory stimuli are by definition required to be able to be sensed. In this respect, they are clearly observable (condition 1). Furthermore, observation allows the stimuli to be directly recorded, rather than relying on third-party recollection of what is present. Second, sensory stimuli can be classified into the five senses, and then into sub-groups based on the delivery of the stimuli. Thus, patterns of presence and level (of the phenomenon) are sought in the observation (condition 2). Whilst observation has been regularly used in store atmosphere research because of the subconscious effects on consumers (e.g. Milliman, 1982; Crowley, 1993; Areni and Kim, 1994; Herrington and Capella, 1996; Yalch and Spangenberg, 2000) the stimuli themselves are not subconscious and are in fact directly observable/measurable. Thus condition 3 is not satisfied. Finally, in this situation, the objects under investigation are not people, but sensory stimuli such as music, colours, construction materials, temperature, etc. and thus are unable to communicate directly with the researcher as to their presence (satisfying condition 4). Observation is clearly an appropriate methodology from a technical view. As Dodd et al. (1998) note, observation can be used as both a qualitative and a quantitative research methodology and where the research is exploratory in nature it is likely to be qualitative. However, unlike exploratory studies that use unstructured observation because of the early stage they occur at in the research process (Boote and Mathews, 1999; Churchill, 1999), the observation in this study is able to be reasonably structured because of the clear classifications that already exist for sensory stimuli based on the five senses and the work of Kotler (1974), Baker (1986), and Bitner (1992) that created typologies of stimuli. Effectively therefore the objective of the observation research is to identify gaps in the classes (e.g. are there sounds other than music that create store atmosphere), and gaps in the measures of existing classes (e.g. does music vary only on tempo, volume, and pitch, and what are the levels of tempo, volume, and pitch that are employed). Boote and Mathews (1999) list several widely used classifications for observational research that operate in terms of dichotomies. Human vs machine observation (e.g. Hair et al., 2003; Zikmund, 2003) is the first of these. The observation in this study makes use of both human and machine observation. The stores were digitally videotaped, including full audio recording. At the same time, the researcher observed any scent and the ambient temperature. The video was used because of the subconscious effects detailed in Boote and Mathews (1999) third condition, and shown

JFMM 15,4

432

to exist in store atmosphere studies (e.g. Milliman, 1982; Crowley, 1993; Areni and Kim, 1994; Herrington and Capella, 1996; Yalch and Spangenberg, 2000). Video could capture stimuli that might otherwise not be noticed and recorded by a human observer in the environment, and checked for later by a researcher that is not distracted by all the stimuli. However, video can only capture visual and aural stimuli, and record information for the haptic sense. It is unable to record scent, nor temperature or related tactile stimuli. Thus, the researcher noted these stimuli. However, because senses are inherently subjective, store staff were also asked about the presence of scent, and the ambient temperature. The second dichotomy of overt vs covert (e.g. Hair et al., 2003; Zikmund, 2003) was not relevant as the stimuli would not change in either case, and there were no ethical issues involved with observing the stimuli (Boote and Mathews, 1999). Similarly the third dichotomy of natural vs contrived (e.g. Hair et al., 2003; Zikmund, 2003) was not applicable as the observation had to be in the natural environment. Finally, the dichotomy of participatory vs non-participatory observation (e.g. Hair et al., 2003; Zikmund, 2003) is only applicable in terms of the researcher using their own senses to observe stimuli (the scent and ambient temperature) but store staff corroborated these. Each store observed was selected on the basis of three criteria. The first criterion was simply that it was an example of a womens apparel store. The second was an evolving criterion. Once examples of stores that exhibited the stimuli suggested in the literature had been observed (thus confirming the existence of such stimuli), the criterion changed to being examples of stores that exhibited either unusual combinations of stimuli (in terms of the existing literature) or presented stimuli that had not been suggested by the literature. This was initially determined through storefront examination (looking through the door/window, etc.) similar to Ward et al.s (1992) proto-typicality approach, followed by entry into the store and confirmation of the stimulus. The third of the criteria was that management would allow access for stimuli recording. The store was observed by video and human sense confirmed by management interview. Video was taken of the entire store. Videoing the entire store was normally by either walking through the store, or by centrally locating the camera and panning around the store. The approach was determined by the layout and size of the store. Even if there was only a single stimulus that varied in the store from other store examples, the whole store was videoed to capture any potential interactions, by recording the combinations of stimuli used. In some stores the layout required positioning the camera in other spots such as part way up stairs, on balconies, and other raised areas in order to capture the entire store. During the visit, any scent was recorded on the observation form along with any information on ambient temperature and any unstructured information on stimuli that may not be captured by the video. The store was often discussed with management and any further information gleaned also noted on the store sheet. Subsequent to the visit, the video was viewed looking and listening for the stimuli to be recorded in the structured section of the store sheet, along with any new stimuli identified noted in the unstructured section. Where doubt existed as to the nature of stimuli, a second viewer was asked to observe and suggest their interpretation. The original observer was the researcher who has a number of years experience in this technique. If this did not resolve the issue, the store management was contacted. Stores were observed in four cities as coverage of retail in major centres that also incorporate the multicultural diversity of New Zealand; four towns as examples of

tourism-influenced retail centres; two small centres as examples of unusual period or culturally influenced retail; and seven further towns as examples of smaller, ruralbased retail centres. It was expected that the main centre stores would reflect the literature as these retailers are more international in character, and that the other centres may offer insight into other possible stimuli combinations as these have unique influences impacting on the retail environment. Refinement of stimuli and discovery of desired outcomes through survey To establish marketplace practice the general managers of 39 New Zealand womens clothing retailers were surveyed using a set of fixed response questions derived from the observation. There were three types of measurement, all designed to provide characteristics. One type consisted of presence questions respondents were asked to indicate whether they had such stimuli present in-store and were simple, pre-coded lists that were mutually exclusive and collectively exhaustive, using descriptors that discussions with retailers in the observation stage had suggested. These descriptors were in laymens terms rather than precise measures (e.g. bright light rather than lux level, with bright described as brighter than outside daylight), which potentially introduces a degree of subjectivity to responses. This was pre-tested for consistency, however, and a sample was also drawn from the final responses and the relevant stores visited to observe levels; responses were deemed to be consistent in measurement across all the stimuli descriptors, so if there was any error, it was at least systematic. A second type of measurement consisted of scales measuring levels of stimuli. These scales were linear numeric scales with full descriptors based on retailer discussions, designed to elicit distributions for each stimuli. The midpoints were the mode from observations. The third types of measurement were expectation questions, asking respondents to select from a list one or more expected responses to their in-store stimuli. These were centred on the key stimuli dimension constructs identified in the literature. Pre-tests of the sections, the whole and the revised questionnaire were undertaken (Alreck and Settle, 1995; Churchill and Iacobucci, 2002). A schema, following the classic guidelines of Sokal and Sneath (1963), Frank and Green (1968), Harvey (1969), and Hunt (1991), of the four sensory dimensions was constructed from the responses to the survey. This forms the typical industry environment set of stimuli (Table I), along with the expected responses to the stimuli (Table II). 4. The ideal store atmosphere We designed a two-stage experiment to establish the ideal set of stimuli (cf. industry standard). Treatments were presence, and levels of stimuli, with measurement taken of affective responses. Subjects (n 489) were randomly assigned to treatment groups, or a control group, with a GLM used for analysis. A digital movie was made of a clothing store with segments lasting approximately 90-120 seconds (depending on the stimuli incorporated). This approach has proven effective for environmental representation of visual and aural aspects (Chebat et al., 1993; Voss et al., 1998; Machleit and Mantel, 2001; Baker et al., 2002), providing a high degree of ecological validity (Bateson and Hui, 1992), where a specific environmental setting is required (Sweeney and Wyber, 2002). Parsons (2009) demonstrates the ability of a digital representation to capture an audiences affective and cognitive flow for sensory stimuli with visual and aural stimuli portrayed electronically, and tactile and olfactory stimuli physically present. A measure used by Voss et al. (1998) was employed asking a sub-sample (n 12) if the

Atmosphere in fashion stores

433

JFMM 15,4

Stimuli dimension Music style Music volume Music tempo Music familiar and preferred by customers Other sounds Scent present Scent pleasant Scent congruence Temperature controlled Temperature Flooring Lighting source Lighting level Wall colour Ceiling colour Floor colour Lighting colour

Retailer reported level of stimuli Easy listening (40.0%); popular (top 40) (28.6%) Average (2.94: 0.072) Average (3.03: 0.079) Yes (91.2%) No other sounds (80%) No (76.5%) Pleasant (2.13: 0.350) Somewhat congruent (2.00: 0.267) Yes (85.3%) 18-251C (75.8%); 10-171C (18.2%) Carpet (73.0%); wood (32.4%); concrete (21.6%); vinyl (16.2%); tiles (13.5%) Halogen (55.3%); natural light (50.0%); exposed fluorescent (42.1%); ceiling panel fluorescent (36.8%); non-halogen spotlights (23.7%) Bright (2.08: 0.095) White (34.3%); cream (25.7%) White (74.3%); cream (14.3%) Grey (21.2%); brown (12.1%); blue (12.1%); natural wood (12.1%) Translucent (86.0%); white (14.0%)

434

Notes: Proportion stating presence of this level shown in parentheses, or if five-point scale used, mean and standard deviation shown in parentheses; multiple levels permitted; only levels that were listed by Table I. Schema of sensory stimuli 10 per cent or more are given

Stimuli dimension Music

Retailer expectation of shopper response Stay longer (41.2%); take time to examine merchandise (32.4%); increase sales (23.5%); examine the merchandise more (14.7%); slow down shoppers move through store (14.7%); no change (14.7%); spend less time in store (14.7%) Evaluate store more positively (38.9%); like store (33.3%); take time to examine merchandise (27.8%); no change (27.8%); increase sales (22.2%); stay longer (22.2%); slow down shoppers move through store (22.2%); examine the merchandise more (16.7%) Stay longer (72.7%); like store (51.5%) Examine the merchandise more (64.3%)

Scent

Table II. Shopper responses to stimuli, expected by retailers

Temperature Lighting/colour

Notes: Proportion stating this expectation shown in parentheses; multiple expectations permitted

videotaped scenario was realistic (on a seven-point scale anchored by very unrealistic [1] and very realistic [7]: mean 6.4), to assess ecological validity, which was considered acceptable. The video was also pre-tested to ensure a reasonable level of proto-typicality (Ward et al., 1992). A manipulation check was conducted with a subsample (n 10), identifying the occurrence of temperature, lighting, music, and scent differences correctly.

Subjects entered a room that was fully blacked out with theatre-quality curtains. A single light was on illuminating a black, adjustable leather chair. The subject was seated and told they were about to see a 90-120 second video clip of a store. They were given a set of wireless headphones, and were told that at the same time as the video starting, some music would be heard. The video projector was rear of the subject out of sight. The projection filled the wall in front of the subject (approximately 4 m 3 m). During the video, a scent level was present and the temperature in the room was controlled. Immediately upon termination of the video, the subject was presented with the measurement instrument. The levels of each stimulus were applied as follows: music: slow tempo; mid-tempo; fast tempo; temperature: extremes (hot/cold); mid level (comfortable); lighting: bright; dull (neutral/dull); scent: congruent; incongruent, based on the schema, forming a 3 2 2 2 design. The music was classical/popular mixes, with tempo pre-tested to be slow, mid, and fast tempo. Heating/cold air units, monitored using a thermostat, controlled temperature. The two levels were the 18-25 degree level (mid) and 10-17 (low). Lighting was adjusted in the video by altering the contrast and brightness levels in the digitisation. Again, lighting was pre-tested to ensure subjects saw the contrasts as reflective of the industry descriptors. Finally, scent was perfume (congruent associated Parsons, 2009) and coffee (incongruent) presented through a spray. The sample was approximately one-third general population and two-thirds students. As the store was a type likely to be shopped at by both groups the inclusion of a student sample for convenience was not seen as a potential problem; however the split sample responses were compared, with no significant difference found. As discussed by Sternthal et al. (1994), the need for a sample that is truly representative of the population of interest arises only for research aimed at generalisations. For the theoretical explanation this paper seeks, a more homogenous sample is appropriate as it lessens the possibility of diversity undermining the chances of finding theoretically predicted outcomes. To examine the responses to stimuli, a measure of affect was developed based on scales originally developed by Fisher (1974) for measuring affect in terms of perceived environmental state. The 13 relevant items from Fishers (1974) environmental quality scale that have been used in similar environmental marketing studies (e.g. Crowley, 1993; Spangenberg et al., 1996; Parsons, 2009) were employed. The mean of the 13 items is taken for each subject to give a score for affect. Each item uses a seven-point scale. Cronbachs a for the store 0.7795, an acceptable level (with the common threshold held as 0.7000 Hair et al., 1995). It is conceivable that normal experience with the store type may have an effect upon responses to stimuli, particularly in light of Mitchell et al.s (1995) discussion of expected stimuli having different effects from unexpected. On this principle, subjects treated with a store that, according to the schema developed, was typical and expected, are likely to have less positive affect responses than those for whom the store was typical, but because of inexperience, unexpected. Three items were used to measure subjects levels of normal exposure to the store type; typical frequency of visits to the store type, typical expenditure at such a store, and typical length of time spent in such a store and used as a check of potential difference in affect; no significant difference was found. MANOVA was employed to analyse the dependence relationship represented as the differences in a set of dependent measures (affect and four behavioural measures that are not discussed in this paper) across a series of groups formed by the four categorical independent measures (the sensory stimuli). Hair et al.s (1995) requirements for

Atmosphere in fashion stores

435

JFMM 15,4

436

MANOVA are met as follows: the sample in each cell exceeds the number of dependent variables; the interactions are ordinal; the use of covariates to account for systematic bias has been taken care of by the random assignment of respondents to treatments; the three assumptions of MANOVA, independence (among observations), equality of variance-covariance matrices, and normality (of the dependent measures) are all considered met. There is no discernible time-order amongst respondents, treatments were randomly shuffled throughout each day, and respondents were randomly assigned to experiment treatments, and there does not appear to be any pattern of dependence amongst respondents. The equality of the matrices is tested by the box test and there are no significant differences (F 0.951, significance 0.735). Group sizes are approximately equal so even if the assumption were violated, there would be minimal impact (Hair et al., 1995). While there is no test for multivariate normality, univariate normality across the measures suggests that any departure from multivariate normality would be inconsequential[1]. Outliers were not a problem as we confined data to limited scales. Pillais criterion was used; it is the most robust of the tests available when sample sizes are not large (as in this case). Effect size and the power of the test were examined and considered acceptable. We interpret the results in light of observed causality based on three common criteria (e.g. Churchill and Iacobucci, 2002); our estimated marginal means support concomitant variation, time order of occurrence is integral to the after-only experimental design, and the elimination of other possible causal factors is obtained by the random assignment of subjects and the controlled laboratory setting. The result of the multivariate test is shown in Table III. Examination of the marginal means (the population estimates rather than the sample means) allows us to see the directions of the relationships established. These means are shown in Table IV. The main effects are all positive with the exception of temperature. Brighter lighting results in greater affect, as does faster music. These are findings new to the literature on apparel stores. Congruent scent increases affect,

Effect

Pillais trace 0.000a 0.000a 0.000a 0.033a 0.000a 0.000a 0.246 0.000a 0.005a 0.855 0.000a 0.164 0.000a 0.741 0.000a 0.435 level

Affect (liking) 0.000a 0.000a 0.000a 0.001a 0.000a 0.001a 0.062 0.000a 0.000a 0.390 0.000a 0.021a 0.000a 0.355 0.000a 0.270

Intercept Main effects Lighting Temperature Music Scent Interactions Lighting temperature Lighting music Temperature music Lighting temperature music Lighting scent Temperature scent Lighting temperature scent Table III. Music scent Multivariate test p-value Lighting music scent for main effects and Temperature music scent interactions; Pillais trace Lighting temperature music scent and between subjects Notes: Pillais trace criterion used; asignificant at the 0.05 effect

Stimuli Lighting Bright Dull Temperature Extreme Mid level Music Slow Mid Fast Scent Congruent Incongruent Note: Means are shown followed by the standard error enclosed in parentheses)

Affect EMM

Atmosphere in fashion stores

4.67 (0.03) 4.43 (0.04) 4.10 (0.04) 4.99 (0.04) 4.47 (0.04) 4.49 (0.04) 4.67 (0.04) 5.10 (0.04) 3.99 (0.04) Table IV. Estimated marginal means significant main effects

437

strengthening the general findings of Spangenberg et al. (1996), Mitchell et al. (1995), Mattila and Wirtz (2001), and Parsons (2009). Temperature increases affect when it is mid-level rather than the extreme, which is supported by the general retail findings of Machleit et al. (1994) and dAstous (2000). 5. Repeated exposure The significant main effects and interactions were used to run the second stage of the experiment with repeated measures over a four-day period (n 62 subjects; sampled from the same population as the previous experiment). The same procedures were followed as before, except that subjects were exposed to the typical industry standard or the ideal set of main effects and interactions (derived from the earlier experiments results), or a combination of both over the period (changeover occurring period three). The ideal levels were bright lighting, fast music, mid-level temperature, and presence of a congruent scent. The industry norm is bright lighting, mid-tempo music, mid-temperature, and no scent. A repeated measures GLM was used to examine the within-subjects effects of repeated exposure to stimuli. The ideal atmosphere group was exposed to the same ideal stimuli in each of the four periods. The industry atmosphere group was exposed to the same industry stimuli in each of the four periods. The industry/ideal group was exposed to two periods of industry atmosphere followed by two periods of ideal atmosphere, while the ideal/ industry group had the reverse. Mauchlys test of sphericity suggests that for the industry, industry/ideal, and ideal/industry groups, sphericity can be assumed for the analysis, but for the ideal group, the e correction needs to be applied, so the Greenhouse-Geisser e corrected and Huynh-Feldt e corrected significance levels are used. The ideal group had no significant effect for affect over time (F 1.848, Greenhouse-Geisser significance 0.172; Huynh-Feldt significance 0.165), and neither did the ideal/industry group (F 2.291, significance 0.095), but the other two did industry group (F 5.910, significance 0.002); industry/ideal group (F 5.977, significance 0.002). If we go with the premise that there is likely to be a decay effect from repeated exposure to the same stimuli, the directions of the repeated exposure trends in

JFMM 15,4

Figures 1-4 suggest that if the store atmosphere is less than ideal there is a significant decline in affect for the store. If the atmosphere is ideal (in terms of retailer desired responses) the negative effects of repeated exposure could be mitigated. Furthermore, if the atmosphere is changed positively i.e. the stimuli levels become ideal or closer to ideal there can be significant increase in the affect for the store.
7 6 Mean affect 5 4 3 2 1 Day 1 Day 2 Day 3 Period Day 4

438
Figure 1. Ideal atmosphere bright lighting, fast music, mid-level temperature, presence of a congruent scent

7 6 Mean affect 5 4 3 2 1 Day 1 Day 2 Day 3 Day 4 Period 7 6 Mean affect 5 4 3 2 1 Day 1 Day 2 Day 3 Day 4 Period 7 6 Mean affect 5 4 3 2 1 Day 1 Day 2 Day 3 Day 4 Period

Figure 2. Industry atmosphere bright lighting, mid-tempo music, mid-level temperature, no scent

Figure 3. Industry/ideal atmosphere industry atmosphere for two periods followed by ideal atmosphere for two periods

Figure 4. Ideal/industry atmosphere ideal atmosphere for two periods followed by industry atmosphere for two periods

6. Conclusions and implications for managers and researchers The first of the conclusions drawn is that interactions between sensory stimuli belonging to different dimensions (Kotler, 1974; Baker, 1986; Bitner, 1992) do occur and do have significant effects on shopper affect. On this basis, the literature-based and dimensions-based conceptualisations of research-to-date that have concentrated on main effects only are rejected as accurate representations of the influence of in-store, non-utilitarian sensory stimuli on shopper affect. The conceptualisation provided in the interactions-based approach provides a better guide for the operationalisation of the store atmosphere as a package as encapsulated by the construct approach (e.g. Baker, 1986; Bitner, 1990, 1992; Kerin et al., 1992; Ward et al., 1992; Grewal and Baker, 1994; Baker et al., 1995; Spies et al., 1997; Babin and Attaway, 2000). There is also a clear conclusion that arises from the schema developed. Kotler (1974) and subsequent researchers view atmospherics as the conscious designing of space to create certain effects in buyers specifically emotional effects that enhance their purchase probability. Increasingly, however, researchers are seeing stimuli being used as a means of differentiating the retail environment from competitors, and enhancing the shopping experience by creating exciting surroundings (e.g. Wakefield and Baker, 1998; Parsons, 2003, 2009), or at the very least, ambient congruity (e.g. Grewal and Baker, 1994; Spies et al., 1997). Yet the marketplace shows, through the schema of stimuli used and the expected responses established a surprising degree of uniformity within the fashion sector. This is particularly surprising in light of Ward et al.s (1992) showing that consumers perceptions of proto-typicality and their attitude towards stores is strongly influenced by environmental cues. Whether this is a lack of design on the part of retailers in New Zealand, a cost issue (e.g. white paint is cheaper than coloured; fluorescent lighting is cheaper than halogen, etc.), or a false perception amongst retailers that the stimuli packages they are using are the best combination, the empirical results indicate that retailers could do better in achieving desired affective responses. The conclusion drawn therefore is that retailers are far less differentiated in their use of sensory stimuli than they could be to achieve the responses they expect. As Parsons (2003) discusses in the shopping mall context, there is a greater need for retailers to differentiate themselves in an environment of disaffected shoppers bored by the sameness of offerings, yet the observation and survey of retailers in this research shows a general consensus of sensory stimuli usage within the fashion sector, at odds with a context of shoppers seeking not only utilitarian succour but hedonic gratification as well (Babin et al., 1994). The third conclusion concerns repeated exposure to stimuli. After spending sometimes hundreds of thousands of dollars, a retailer is faced with a shopper quickly becoming bored or disaffected with the sensory stimuli and the benefits of positive affect are diminished or lost altogether. Ward et al.s (1992) perceptions of prototypicality requires the environmental cues to suggest a matching of the shoppers self-concept and the store image as portrayed (in part) by store atmosphere, but as Sirgy et al. (2000) suggests, the person may develop or the store may develop (or stagnate) and the match may become less appealing. If having the same stimuli is seen as stagnation, and is considered a negative event (in terms of the shopping environment), Stoltman et al. (1999) suggest such an event could lead to switching. The results of the experiment on repeated exposures effect on affect suggest that stagnation with repeated exposure can diminish affect for the store. The results also suggest that change in stimuli, particularly positive change in terms of moving to levels that are more likely to induce positive affect, can revitalise the shopper affect.

Atmosphere in fashion stores

439

JFMM 15,4

440

Thus, the conclusion is that repeated exposure to the same stimuli can result in reduced affect, whereas a change in stimuli can revitalise and even increase affect. The change does not have to be wholesale nor dramatic. The differences between the industry standard set of influential stimuli and the ideal set is a matter of level. By altering the level of the stimuli retailers can achieve a meaningful affective response at a fraction of the cost of a full refit. The retail manager must appreciate that small changes can have large effects. The findings show that a change from mid tempo music to faster tempo music can significantly enhance shopper affect in a clothing store. The dilemma a manager faces when designing the atmosphere of their store is that having invested considerable budget into creating the ambience that appeals to the shopper the shopper can then become bored and disaffected, as the findings about repeated exposure revealed. However, a simple change of music played, a dimming of the lights temporarily, the introduction of an alternative scent all these changes in stimuli are simple and small, but can have significant positive effects on responses. Thus, a retailer who has invested in a store design expected to last five or six years, is not faced with a redundant design after the first visits by shoppers. By constantly manipulating the sensory stimuli levels and even presence, the retailer can instil a revitalisation of the store. Whilst not examined in this research, the logical interval between changes would be tied to the normal interval between visits for typical shoppers. Empirically this is the first study to take a stimuli from each of the dimensions or factors conceptually described by Kotler (1974), Baker (1986), and Bitner (1992) and test them in a full factorial design, creating a representation of a holistic store atmosphere, rather than simply single or multiple stimuli. By doing so the research shows that it is possible to empirically test combinations of stimuli and/or levels. The implication therefore for researchers is that different stimuli could easily be added or substituted within a dimension to test its effect within a factorial design, rather than relying on main effect results only. Conceptually the first implication for researchers is that in any model of sensory stimuli-based store atmosphere relationships with shopper affect and/or behaviour should incorporate interactions. By omitting interactions researchers may be missing potentially influential stimuli that individually may not have significant effects, but in combination with other stimuli may greatly influence shopper affect. This also means that it may be necessary to revisit studies that have contradictory results concerning main effects, such as the debate over whether scent should be congruent (e.g. Bone and Jantrania, 1992; Mitchell et al., 1995; Bone and Ellen, 1999; Fiore et al., 2000), or even present (Bone and Ellen, 1999; Gulas and Bloch, 1995; Mattila and Wirtz, 2001) to generate responses, or whether music tempo influences behaviour (e.g. Herrington and Capella, 1996; Milliman, 1982). It may be that the music, scent, or whatever stimuli has been studied is interacting with other stimuli or is not interacting with other stimuli when it should be, to provide what has been observed as a main effect. The second conceptual implication is that a model of store atmosphere is likely to require a mediating influence in the form of repeated exposure. Clearly this is only relevant in studies concerning patronage, but it would be rare for a researcher to be investigating a retail situation where a single visit is common. The findings from this study are not comprehensive in their determination of a need for the mediating variable, but there is a clear indication that such a variable is likely to have a part to play in any model of store atmosphere. Conceptually this then raises the question of whether other mediating, or moderating variables may exist that should also be included.

Methodologically the strength of the findings in this research shows two things. First, the combination of field observation and sample survey research, based on an extensive literature analysis, to produce a schema of industry norms, allows for the construction of a valid and reliable set of experiments to empirically test the conceptual model and related propositions. Such an approach (the use of a schema) is unusual in marketing research, but this study shows how effective such an application can be. Second, the use of a controlled setting can be applied in environmental studies of marketing, achieving ecological validity but allowing the researcher to control external influences. The use of digitally enhanced displays of visual and aural stimuli in combination with physical manifestations of olfactory and tactile stimuli allow the researcher to conduct studies of store and other market environments without the expense of constructing artificial sets, or the trade-off of placing the study in a field setting. Finally, with the ability to measure components of store atmosphere directly, but as part of a package, it may be that a conceptualisation like Bitners (1990, 1992) can begin to be tested, integrating store atmospherics with other store environment factors, or with other determinants of patronage (e.g. Kerin et al., 1992; Sirgy et al., 2000).
Note 1. Kolmogorov-Smirnov tests give the following univariate results: affect: Z 2.236; time: Z 3.440; money: Z 3.259; frequency: Z 2.968; likely: Z 3.243. In all cases the asymptotic significance was 0.000. References Alreck, P.L. and Settle, R.B. (1995), The Survey Research Handbook, 2nd ed., Richard D. Irwin, Boston, MA. 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 No. 2, pp. 117-25. Babin, B.J. and Attaway, J.S. (2000), Atmospheric affect as a tool for creating value and gaining share of customer, Journal of Business Research, Vol. 49 No. 2, pp. 91-9. Babin, B.J., Darden, W.R. and Griffin, M. (1994), Work and/or fun: measuring hedonic and utilitarian shopping value, Journal of Consumer Research, Vol. 20 No. 4, pp. 644-56. Babin, B.J., Hardesty, D.M. and Suter, T.A. (2003), Color and shopping intentions: the intervening effect of price fairness and perceived affect, Journal of Business Research, Vol. 56 No. 7, pp. 541-51. Baker, J. (1986), The role of the environment in marketing services: the consumer perspective, in Czepiel, J.A., Congram, C.A. and Shanahan, J. (Eds), The Services Challenge: Integrating for Competitive Advantage, American Marketing Association, Chicago, IL, pp. 79-84. Baker, J., Grewal, D. and Parasuraman, A. (1995), The effect of store atmosphere on customer quality perceptions and store image, Stores, Vol. 77 No. 7, p. RR5. 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 (April), pp. 120-41. Bateson, J.E.G. and Hui, M.K. (1992), The ecological validity of photographic slides and videotapes in simulating the service setting, Journal of Consumer Research, Vol. 19 No. 2, pp. 271-81. Birtwistle, G. and Shearer, L. (2001), Consumer perception of five UK fashion retailers, Journal of Fashion Marketing and Management, Vol. 5 No. 1, pp. 9-18.

Atmosphere in fashion stores

441

JFMM 15,4

442

Bitner, M.J. (1990), Evaluating service encounters: the effects of physical surroundings and employee responses, Journal of Marketing, Vol. 54 (April), pp. 69-82. Bitner, M.J. (1992), Servicescapes: the impact of physical surroundings on customers and employees, Journal of Marketing, Vol. 56 No. 2, pp. 57-71. 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. Bone, P.F. and Jantrania, S. (1992), Olfaction as a cue for product quality, Marketing Letters, Vol. 3 No. 3, pp. 289-96. Boote, J. and Mathews, A. (1999), Saying is one thing; doing is another: the role of observation in marketing research, Qualitative Market Research: An International Journal, Vol. 2 No. 1, pp. 15-21. Chebat, J.-C., Gelinas-Chebat, C. and Filiatrault, P. (1993), Interactive effects of musical and visual cues on time perception: an application to waiting lines in banks, Perceptual and Motor Skills, Vol. 77 No. 3, pp. 995-1021. Churchill Jr, G.A. (1999), Marketing Research: Methodological Foundations, 7th ed., The Dryden Press, Fort Worth, TX. Churchill Jr , G.A. and Iacobucci, D. (2002), Marketing Research, Methodological Foundations, 8th ed., Harcourt College Publishers, Fort Worth, TX. Crowley, A.E. (1993), The two-dimensional impact of colour on shopping, Marketing Letters, Vol. 4 No. 1, pp. 59-69. dAstous, A. (2000), Irritating aspects of the shopping environment, Journal of Business Research, Vol. 49 No. 2, pp. 149-56. Diamond, J. and Diamond, E. (1996), Fashion Advertising and Promotion, Delmar Publishers, New York, NY. Dodd, C.A., Clarke, I. and Kirkup, M.H. (1998), Camera observations of customer behaviour in fashion retailing: methodological propositions, International Journal of Retail and Distribution Management, Vol. 26 No. 8, pp. 311-17. Donovan, R.J. and Rossiter, J.R. (1982), Store atmosphere: an environmental psychology approach, Journal of Retailing, Vol. 58 No. 1, pp. 34-57. Drake, M.F., Spoone, J.H. and Greenwald, H. (1992), Retail Fashion Promotion and Advertising, Macmillan Publishing Co., New York, NY. Eroglu, S.A., Machleit, K.A. and Chebat, J.-C. (2005), The interaction of retail density and music tempo: effects on shopper responses, Psychology and Marketing, Vol. 22 No. 7, pp. 577-89. Fisher, J.D. (1974), Situation-specific variables as determinants of perceived environmental aesthetic quality and perceived crowdedness, Journal of Research in Personality, Vol. 8 (August), pp. 177-88. Fiore, A.M. and Kim, J. (2007), An integrative framework capturing experiential and utilitarian shopping experience, International Journal of Retail & Distribution Management, Vol. 35 No. 6, pp. 421-42. Fiore, A.M., Yah, X. and Yoh, E. (2000), Effects of a product display and environmental fragrancing on approach responses and pleasurable experiences, Psychology and Marketing, Vol. 17 No. 1, pp. 27-54. Frank, R.E. and Green, P. (1968), Numerical taxonomy in marketing analysis: a review article, Journal of Marketing Research, Vol. 5 No. 1, pp. 83-94. Grewal, D. and Baker, J. (1994), Do retail store environmental factors affect consumers price acceptability? An empirical examination, International Journal of Research in Marketing, Vol. 11 No. 2, pp. 107-15.

Gulas, C.S. and Bloch, P.H. (1995), Right under our noses: ambient scent and consumer responses, Journal of Business and Psychology, Vol. 10 No. 1, pp. 87-98. Hair Jr, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1995), Multivariate Data Analysis, 4th ed., Prentice Hall, Englewood Cliffs, NJ. Hair Jr, J.F., Bush, R.P. and Ortinau, D.J. (2003), Marketing Research: Within a Changing Information Environment, 2nd ed., McGraw-Hill Irwin, New York, NY. Harvey, D. (1969), Explanation in Geography, St Martins Press, New York, NY. Hart, C., Farrell, A.M., Stachow, G., Reed, G. and Cadogan, J.W. (2007), Enjoyment of the shopping experience: impact on customers re-patronage intentions and gender influence, The Services Industries Journal, Vol. 27 No. 5, pp. 583-604. Herrington, J.D. and Capella, L.M. (1996), Effects of music in service environments: a field study, Journal of Services Marketing, Vol. 10 No. 2, pp. 26-41. Hunt, S.D. (1991), Modern Marketing Theory: Critical Issues in the Philosophy of Marketing Science, South-Western Publishing Co., Cincinnati, OH. Joseph-Mathews, S., Bonn, M.A. and Snepenger, D. (2009), Atmospherics and consumers symbolic interpretations of hedonic services, International Journal of Culture, Tourism, and Hospitality Research, Vol. 3 No. 3, pp. 193-210. Kerin, R.A., Howard, D.J. and Jain, A. (1992), Store shopping experience and consumer price-quality-value perceptions, Journal of Retailing, Vol. 68 No. 4, pp. 376-97. Kotler, P. (1973/1974), Atmospherics as a marketing tool, Journal of Retailing, Vol. 49 No. 4, pp. 48-64. Kottasz, R. (2006), Understanding the influences of atmospheric cues on the emotional responses and behaviours of museum visitors, Journal of Non-Profit and Public Sector Marketing, Vol. 16 Nos 1/2, pp. 95-119. Machleit, K.A. and Mantel, S.P. (2001), Emotional response and shopping satisfaction: moderating effects of shopper attributions, Journal of Business Research, Vol. 54 No. 2, pp. 97-106. Machleit, K.A., Kellaris, J.J. and Eroglu, S.A. (1994), Human versus spatial dimensions of crowding perceptions in retail environments: a note on their measurement and effect on shopper satisfaction, Marketing Letters, Vol. 5 No. 2, pp. 183-94. Mattila, A.S. and Wirtz, J. (2001), Congruency of scent and music as a driver of in-store evaluations and behaviour, Journal of Retailing, Vol. 77 No. 2, pp. 273-89. Michon, R., Chebat, J.-C. and Turley, L.W. (2005), Mall atmospherics: the interaction effects of the mall environment on shopping behaviour, Journal of Business Research, Vol. 58 No. 5, pp. 576-83. Michon, R., Yu, H., Smith, D. and Chebat, J.-C. (2008), The influence of mall environment on female fashion shoppers value and behaviour, Journal of Fashion Marketing and Management, Vol. 12 No. 4, pp. 456-68. Mitchell, D.J., Kahn, B.E. and Knasko, S.C. (1995), Theres something in the air: effects of congruent or incongruent ambient odor on consumer decision making, Journal of Consumer Research, Vol. 22 No. 2, pp. 229-38. Milliman, R.E. (1982), Using background music to affect the behaviour of supermarket shoppers, Journal of Marketing, Vol. 46 No. 3, pp. 86-91. Morrin, M. and Ratneshwar, S. (2000), The impact of ambient scent on evaluation, attention, and memory for familiar and unfamiliar brands, Journal of Business Research, Vol. 49 No. 2, pp. 157-65. Moye, L.N. and Giddings, V.L. (2002), An examination of the retail approach-avoidance behaviour of older apparel consumers, Journal of Fashion Marketing and Management, Vol. 6 No. 3, pp. 259-76.

Atmosphere in fashion stores

443

JFMM 15,4

444

Newman, A.J. and Patel, D. (2004), The marketing directions of two fashion retailers, European Journal of Marketing, Vol. 38 No. 7, pp. 770-89. Noad, J. and Rogers, B. (2008), The importance of retail atmospherics in B2B retailing: the case of BOC, International Journal of Retail & Distribution Management, Vol. 36 No. 12, pp. 1002-14. Oakes, S. and North, A.C. (2008a), Using music to influence cognitive and affective responses in queues of low and high crowd density, Journal of Marketing Management, Vol. 24 Nos 5-6, pp. 589-602. Oakes, S. and North, A.C. (2008b), Reviewing congruity effects in the service environment musicscape, International Journal of Service Industry Management, Vol. 19 No. 1, pp. 63-82. Oelkers, D.B. (2004), Fashion Marketing, Thomson South-Western, Mason, OH. Ogle, J.P., Hyllegard, K.H. and Dunbar, B.H. (2004), Predicting patronage behaviours in a sustainable retail environment, Environment and Behaviour, Vol. 36 No. 5, pp. 717-41. Parsons, A.G. (2003), Assessing the effectiveness of shopping mall promotions, International Journal of Retail & Distribution Management, Vol. 31 No. 2, pp. 74-9. Parsons, A.G. (2009), Use of scent in naturally odourless stores, International Journal of Retail & Distribution Management., Vol. 37 No. 5, pp. 440-52. Paulins, V.A. and Geistfeld, L.V. (2003), The effect of consumer perceptions of store attributes on apparel store preference, Journal of Fashion Marketing and Management, Vol. 7 No. 4, pp. 371-85. Scarpi, D. (2006), Fashion stores between fun and usefulness, Journal of Fashion Marketing and Management, Vol. 10 No. 1, pp. 7-24. Seock, Y.-K. (2009), Influence of retail store environment cues on consumer patronage behaviour across different retail store formats: an empirical analysis of US Hispanic consumers, Journal of Retailing and Consumer Services, Vol. 16 No. 5, pp. 329-39. Sharma, A. and Stafford, T.F. (2000), The effect of retail atmospherics on customers perceptions of salespeople and customer persuasion: an empirical investigation, Journal of Business Research, Vol. 49 No. 2, pp. 183-91. Sirgy, M.J., Grewal, D. and Mangleburg, T. (2000), Retail environment, self-congruity, and retail patronage: an integrative model and a research agenda, Journal of Business Research, Vol. 49 No. 2, pp. 127-38. Soars, B. (2009), Driving sales through shoppers sense of sound, sight, smell, and touch, International Journal of Retail & Distribution Management, Vol. 37 No. 3, pp. 286-98. Sokal, R.R. and Sneath, P. (1963), Principles of Numerical Taxonomy, W. H. Freeman and Co., San Francisco, CA. Solomon, M.R. and Rabolt, N.J. (2009), Consumer Behavior in Fashion, 2nd ed., Pearson Education, Upper Saddle River, NJ. Spangenberg, E.R., Crowley, A.E. and Henderson, P.W. (1996), Improving the store environment: do olfactory cues affect evaluations and behaviours?, Journal of Marketing, Vol. 60 No. 2, pp. 67-80. Spies, K., Hesse, F. and Loesch, K. (1997), Store atmosphere, mood and purchasing behaviour, International Journal of Research in Marketing, Vol. 14 No. 1, pp. 1-17. Sternthal, B., Tybout, A.M. and Calder, B.J. (1994), Experimental design: generalisation and theoretical explanation, in Bagozzi, R.P. (Ed.), Principles of Marketing Research, Basil Blackwell, Oxford, pp. 195-223. Stoltman, J.J., Morgan, F.W. and Anglin, L.K. (1999), An investigation of retail shopping situations, International Journal of Retail & Distribution Management, Vol. 27 No. 4, pp. 145-53.

Sweeney, J.C. and Wyber, F. (2002), The role of cognitions and emotions in the music-approachavoidance behaviour relationship, Journal of Services Marketing, Vol. 16 No. 1, pp. 51-69. Voss, G.B., Parasuraman, A. and Grewal, D. (1998), The roles of price, performance, and expectations in determining satisfaction in service exchanges, Journal of Marketing, Vol. 62 No. 4, pp. 46-61. Wakefield, K.L. and Baker, J. (1998), Excitement at the mall: determinants and effects on shopping response, Journal of Retailing, Vol. 74 No. 4, pp. 515-39. Ward, J.C., Barnes, J. and Bitner, M.J. (1992), Measuring the prototypicality and meaning of retail environments, Journal of Retailing, Vol. 68 No. 2, pp. 194-220. Ward, P., Daves, B.J. and Kooijman, D. (2007), Olfaction and the retail environment: examining the influence of ambient scent, Service Business, Vol. 1 No. 4, pp. 295-316. Wirtz, J., Mattila, A.S. and Tan, R.L.P. (2007), The role of arousal congruency in influencing consumers satisfaction evaluations and in-store behaviours, International Journal of Service Industry Management, Vol. 18 No. 1, pp. 6-24. Yalch, R.F. and Spangenberg, E.R. (2000), The effects of music in a retail setting on real and perceived shopping times, Journal of Business Research, Vol. 49 No. 2, pp. 139-47. Zikmund, W.G. (2003), Business Research Methods, 7th ed., South Western, Cincinnati, OH. Corresponding author Andrew G. Parsons can be contacted at: andrew.parsons@aut.ac.nz

Atmosphere in fashion stores

445

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com Or visit our web site for further details: www.emeraldinsight.com/reprints