The role of the perceived servicescape in a supermarket

The case of ICA MAXI, Jnkping, Sweden

Masters thesis within International Marketing Author: Tutor: Jnkping 2012 Egle Cicenaite, Maria Maciejewska Prof. Adele Berndt

Masters Thesis in Business Administration

Title: Author: Tutor: Date: Subject terms: The role of the perceived servicescape in a supermarket. The case of ICA MAXI, Jnkping, Sweden Egle Cicenaite, Maria Maciejewska Prof. Adele Berndt 2012-05-14 Mehrabian & Russell Stimulus-Organism-Response model, Bitners framework of servicescapes, environmental factors in a supermarket, customers in Sweden, ICA MAXI

Abstract
Introduction: As many buying decision are made in retail stores, it is interesting to investigate which factors, such as noise, lighting, music, colour, layout or visual communication have a significant influence on customers behaviour in a supermarket. Since only a few studies have been conducted among customers in Sweden, this paper examined their behaviour in the supermarket. The purpose of this study was to examine the role of the various environmental factors in the supermarket as well as how the perceived servicescape influences customers behaviour in the store. The research approach used was that of a single case study, based on survey instrument (personal interviews and electronic collection) in order to collect data on facets in regards to the perceived servicescape and customers behaviour in the supermarket. A supermarket chain the ICA group, precisely ICA MAXI in Jnkping was used for this purpose. The collected data was analysed with the means of factor analysis and multiple regression methods in order to obtain factors that can influence customers behaviour in the supermarket. The conducted study showed that lighting, noise, colours, signs and symbols as well as space/function conditions such as layout and equipment are the factors that create emotionally pleasant environment in the supermarket. Thus, these factors influence customers moods, attitudes or certain beliefs about the supermarket. Moreover, customers may feel happier, more satisfied or relaxed in the supermarket, which lead to arousal stimulation or excitement. Furthermore, the environmental factors explain the approach behaviour such as exploring the supermarket, spending more time on browsing the products which, consequently, refer to an increased number of items bought.

Purpose:

Method:

Findings and Conclusion:

Acknowledgments
This research would not be completed without the great help and support from the people who contributed their time in different ways to our research. We specially would like to thank our supervisor as well as our fellow-students for their suggestions and benevolent attitude. Furthermore, we are kindly grateful for the respondents in Jnkping, Sweden who spent their time to take part in our survey. We want to say thank you as well to managers of ICA MAXI, Jnkping, Sweden for allowing us to collect the data in the supermarket. Last but not least, we highly appreciate our families for their warm and effortless support and encouragement to our study. Jnkping May, 2012 Egle Cicenaite and Maria Maciejewska

ii

Table of Contents
1 Introduction .................................................................................... 7
1.1 1.2 1.3 1.4 1.5 1.6 Background .............................................................................................................. 7 Problem .................................................................................................................... 9 Purpose ...................................................................................................................10 Research questions ................................................................................................10 Delimitations..........................................................................................................11 Definitions..............................................................................................................11 Servicescapes ..........................................................................................................12 Dimensions of the perceived servicescape ........................................................14 2.2.1 Ambient conditions .................................................................................14 a) Music .........................................................................................................14 b) Noise..........................................................................................................15 c) Lightning ...................................................................................................15 d) Colour ........................................................................................................16 2.2.2 Signs, symbols and artifacts....................................................................17 a) Visual communication (Signage, personal artifacts) ...........................17 2.2.3 Space/function conditions .....................................................................17 a) Layout, equipment and furnishing ........................................................17 2.2.4 Connected stimuli ....................................................................................18 Internal responses: cognitive, emotional and physiological ............................18 2.3.1 Cognition and environment ...................................................................18 2.3.2 Emotion and environment .....................................................................18 2.3.3 Physiology and environment ..................................................................19 2.3.4 Variations in internal responses .............................................................19 Individual behaviours ...........................................................................................19 2.4.1 The M-R Stimulus-Organism-Response model ..................................19 2.4.2 Environmental stimulus (S) ....................................................................20 2.4.3 Emotional states (O) ...............................................................................20 2.4.4 Behaviour responses (R) .........................................................................22 2.4.5 Studies based on the M-R model ..........................................................22 The research design ..............................................................................................23 Quantitative research method .............................................................................23 Descriptive research..............................................................................................24 The target population ...........................................................................................24 Sampling technique ...............................................................................................24 Data collection .......................................................................................................25 Questionnaire.........................................................................................................26 3.7.1 Pre-Test .....................................................................................................26 Data analysis...........................................................................................................27 Trustworthiness and relevance of information .................................................27 Reversing negatively worded questions .............................................................29 Reliability ................................................................................................................29

Literature Review ........................................................................... 12

2.1 2.2

2.3

2.4

Methodology ..................................................................................23
3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9

Findings .........................................................................................28
4.1 4.2

iii

4.3 4.4 4.5 4.6

Descriptive statistics .............................................................................................31 Analysing the significant differences ..................................................................32 4.4.1 T-test..........................................................................................................32 4.4.2 One-way ANOVA...................................................................................33 Factor analysis........................................................................................................38 Multiple regression ................................................................................................42 4.6.1 Correlations and multicollinearity .........................................................43 4.6.2 Evaluating the regression model of customers behaviour in ICA MAXI.........................................................................................................44 4.6.3 Evaluating contribution of the dependent variable in customers behaviour in ICA MAXI ..................................................................45

5 6

Analysis ..........................................................................................45 Conclusion .....................................................................................49

6.1 6.2 6.3 Managerial implications ........................................................................................50 Limitations .............................................................................................................51 Future research ......................................................................................................51

7 8

References ......................................................................................53 Appendices .....................................................................................57

iv

Figures
Figure 1. A Framework for Environment-Behaviour Relationship ....................................13 Figure 2. The Mehrabian and Russell (1974) model ..............................................................20

Graphs
Graph 1. The food industry segmentation - Sweden .............................................................. 8 Graph 2. Retail Food Market in Sweden................................................................................... 8 Graph 3. Scree Plot ....................................................................................................................75

Tables
Table 1. Donovan and Rossiters (1982) study ......................................................................22 Table 2. Demographics ..............................................................................................................29 Table 3. Cronbachs Alphas ......................................................................................................30 Table 4. Correlations ..................................................................................................................30 Table 5. Descriptive Statistics ...................................................................................................31 Table 6. T-tests for gender and ambient conditions, space/function, signs, symbols and artifacts, behaviour, and the statement How often do you visit ICA MAXI? .........33 Table 7. T-tests for nationality and ambient conditions, space/function, signs, symbols and artifacts, behaviour, and the statement How often do you visit ICA MAXI? .........33 Table 8. One-way ANOVA age and ambient conditions .....................................................34 Table 9. One-way ANOVA age and space/function............................................................34 Table 10. One-way ANOVA age and signs, symbols and artifacts.......................................34 Table 11. One-way ANOVA age and behaviour .................................................................35 Table 12. One-way ANOVA disposable income and ambient conditions, space/function, signs, symbols and artifacts, and behaviour ..................................................................35 Table 13. T-test monthly disposable income and signs, symbols and artifacts ...................36 Table 14. T-test monthly disposable income and behaviour .............................................36 Table 15. One-way ANOVA monthly disposable income and the statement How often do you visit ICA MAXI? .............................................................................................37 Table 16. One-way ANOVA age and the statement How often do you visit ICA MAXI?37 Table 17. T-test age and the statement How often do you visit ICA MAXI? ....................38 Table 18. One-way ANOVA age and the statement I would prefer if ICA MAXI would play music ......................................................................................................................38 Table 19. KMO and Bartletts Test..........................................................................................39 Table 20. Eigenvalues ................................................................................................................39 Table 21. Rotated Component Matrix .....................................................................................40 Table 22. Factors retained .........................................................................................................40 Table 23. Cronbachs Alphas for all factors ...........................................................................41 Table 24. Multiple regression variables entered ..................................................................43 Table 25. Multiple regression - correlations ...........................................................................43 Table 26. Coefficient multiple regression ............................................................................44 Table 27. Multiple regression R and R square ....................................................................44 Table 28. Multiple regression - ANOVA ................................................................................44 Table 29. Multiple regression coefficients ...........................................................................45

Appendices
Appendix 1 Tables ......................................................................................................58 Table A. 1. Item-Total Statistics ambient conditions.............................................................57 Table A. 2. Item-Total Statistics space/function..................................................................57 Table A. 3. Item-Total Statistics signs, symbols and artifacts ...............................................58 Table A. 4. Item-Total Statistics behaviour ..........................................................................58 Table A. 5. Item-Total Statistics all factors .........................................................................58 Table A. 6. T-test gender and ambient conditions .................................................................59 Table A. 7. T-test gender and space/function .......................................................................60 Table A. 8. T-test gender and signs, symbols and artifacts ....................................................60 Table A. 9. T-test gender and behaviour ..............................................................................60 Table A. 10. T-test gender and behaviour (detail information) .........................................61 Table A. 11. T-test gender and the statement How often do you visit ICA MAXI? ........62 Table A. 12. T-test nationality and all factors .....................................................................62 Table A. 13. T-test nationality and the statement How often do you visit ICA MAXI? ..63 Table A. 14. One-way ANOVA age and ambient conditions (detail information) ...........63 Table A. 15. One-way ANOVA age and space/function (detail information) .................63 Table A. 16. One-way ANOVA age and space/function (detail information) .................64 Table A. 17. One-way ANOVA age and space/function (detail information) .................64 Table A. 18. One-way ANOVA age and signs, symbols and artifacts (detail information)66 Table A. 19. One-way ANOVA age and signs, symbols and artifacts (detail information)66 Table A. 20. One-way ANOVA age and signs, symbols and artifacts (detail information)66 Table A. 21. One-way ANOVA age and behaviour (more details) ..................................68 Table A. 22. One-way ANOVA age and behaviour (more details) ..................................68 Table A. 23. One-way ANOVA age and behaviour (more details) ..................................68 Table A. 24. One-way ANOVA monthly disposable income and signs, symbols and artifacts and behaviour .................................................................................................70 Table A. 25. One-way ANOVA monthly disposable income and signs, symbols and artifacts (detail information).......................................................................................71 Table A. 26. One-way ANOVA monthly disposable income and signs, symbols and artifacts (detail information).......................................................................................71 Table A. 27. One-way ANOVA monthly disposable income and behaviour .................72 .Table A. 28. One-way ANOVA monthly disposable income and signs, symbols and artifacts ..........................................................................................................................72 Table A. 29. One-way ANOVA monthly disposable income and the statement How often do you visit ICA MAXI? (detail information) ..................................................73 Table A. 30. One-way ANOVA age and the statement How often do you visit ICA MAXI? (detail information) .....................................................................................73 Table A. 31. One-way ANOVA age and the statement I would prefer if ICA MAXI would play music (detail information) .........................................................................74 Table A. 32. Eigenvalues Factor analysis .............................................................................74 Appendix 2 - Questionnaires................................................................................................75 In Swedish (a paper version)................................................................................75 In English (an e-version)..................................................................................... 79

vi

Introduction

This chapter aims at giving the reader an overview of the paper. Short background information and the problem are presented. Furthermore, the purpose of the thesis as well as the research questions are discussed. Finally, delimitations and definitions end the chapter.

1.1

Background

The global food and beverage sector, which includes farming, food production, distribution, retail and catering, in 2008 was valued at $5,7 trillion. Hence, it can be said that this industry is one of the main contributors to growth of all economies. The forecast for 2014 is an increase of 3,5 per cent to reach a value of $7 trillion. The largest share in the global food and beverage industry is Europe where generated revenues reached $1,4 trillion in 2007, and over 4 million workers were employed (IMAP, 2010). The retail food industry in 2008 was valued at $3,9 trillion. It is expected to increase 4,8 per cent between the years 2008 and 2013 to reach a value of $5 trillion. Hypermarkets, supermarkets, and discounters contributed 48,8 per cent of total revenues (IMAP, 2010). Based on this fact, it can be concluded that the food and beverage sector as well as the retail food industry have a big growing potential in the near future. When it comes to the Swedish retail food industry, it cannot be characterized by a low degree of internationalization since in 1999 Dutch giant Ahold took over 50 per cent interest in the ICA group. Subsequently, successive consolidations occurred which resulted in a Nordic, rather than domestic, focus to marketing (USDA Foreign Agricultural Service, 2010). The Swedish food retail industry grew by 2,2 per cent in 2009 to reach a value of $30,6 billion. The forecast for 2014 is an increase of 7,5 per cent since 2009, to reach a value of $32,9 billion. What is more, 64,8 per cent of the industrys overall value of the Swedish food retail industry (which is $19,8 billion) is generated by hypermarkets, supermarkets, and discounters sales (Dataminor, 2010). Therefore, it can be said that the Swedish food retail industry follows the same trend as the worldwide food industry. Graph 1 presents the Swedish food retail industry segmentation in details.

Graph 1. The food industry segmentation - Sweden

The food retail industry segmentation

0,5% 16,6% 16,7% 64,8% 0,3% 1,1% Hypermarkets, Supermarkets & Discounters Convenience Stores & Gas Stations Food and Drinks Specialists Drug Stores & Health and Beauty Stores Cash & Carries & Warehouse Clubs Other Source: Dataminor. (2010). Food Retail in Sweden (reference code 0181-2058) [Brochure].

The Swedish food retail market is the most concentrated market in Europe, and is dominated by chains of either hypermarkets or supermarkets. The three most powerful groupings have 86 per cent of a total market share. While the six largest chains are taken into account, the total market share increases to 98 per cent. The strongest chain is the ICA group which has almost half of the market share. However, a worldwide expansion trend of hard discount stores can be noticed as well in Sweden. The first hard discount store in Sweden Danish Netto was introduced in 2002, followed by German Lidl in 2003. Graph 2 shows the Swedish retail market in details.
Graph 2. Retail Food Market in Sweden

Retail Food Market in Sweden

3% 2% 2% 8% 16% 49% ICA Group Coop Axfood Bergendahls 20% Lidl Netto Other

Source: Gullstrand, J., & Jrgensen, Ch. (2011). Local Price Competition: The Case of Swedish Food Retailers [Working Paper].

Over the last decades the retail food industry has undergone changes. Fierce competition as well as technological change have had a significant influence on the current shape of the industry. Supermarkets have transformed from small, independently owned establishments

to large, administratively centralized chains (Davis, Freedman, Lane, McCall, Nestoriak & Park, 2009). According to the Food Marketing Institute in 2010 the total supermarket sales was $562 746 billion. In 36 149 supermarkets ($2 million or more in annual sales) 3,4 million employees were employed (FMI Facts). The above mentioned figures confirm that supermarkets have become a crucial entity in the entire economy due to the considerable number of employees and immense sales value that power the economy. Supermarkets are not just buildings filled with products on the shelves. The supermarket consists as well of a certain atmosphere inside the store. There are many factors such as noise, lighting, music, colour, layout or visual communication that can be taken into consideration in order to build an environment as customer-friendly as possible. Since the suitable environment of the supermarket can influence customers to spend more time inside the store, this can lead to an increased number of items bought. Therefore, many marketers are trying to adapt supermarkets interiors to increase customer satisfaction, loyalty, and by doing so, to increase the sales volume.

1.2

Problem

Wysocki (1979) claims that retailers can manipulate the atmosphere of the supermarket by creating a particular layout (Eroglu & Machleit, 1990; Wakefield & Blodgett, 1994; Ang & Leong, 1997; Lin, 2004; Countryman & Jang, 2006; Ballantine, Jack & Parsons, 2010), lightning (Birren, 1969; Taylor & Sucov, 1974; Areni & Kim, 1994; Countryman & Jang, 2006; Quartier, Christiaans & Van Cleempoel, 2008), colour (Bellizzi & Hite, 1992; Valdez & Mehrabian, 1994; Babin, Hardesty & Suter, 2003; Yildirim, Akalin-Baskaya & Hidayetoglu, 2007; Ballantine et al., 2010), and music (Yalch and Spangenberg, 1988; Yalch & Spangenberg, 1990; Yalch & Spangenberg, 1993; Herrington & Capella, 1996; Oakes, 2000; Mattila & Wirtz, 2001; Ballantine et al., 2010; Jain & Bagdare, 2011; Morrison, Gan, Dubelaar & Oppewal, 2011). According to Vida, Obadia and Kunz (2007) the atmosphere of the supermarket also consists of smells (Morrison et al., 2011), physical environment, sales promotions, shelf space, food sample stations, and visual communication (Davis, 1984; Wener, 1985; Bitner, 1992; Ang & Leoing, 1997). These determinants can influence both customers emotions and their behaviour in the supermarket. According to Kotler (1973), designing an efficient buying environment in order to invoke specific emotional effects, may result in enhancing the chance of a purchase. Additionally, Hansen and Deutscher (1977) state that shopping hours, parking access, prices, services or product diversification are objective variables of the atmosphere of the store. Many studies have been conducted in relation to the effect of the store environment and customers behaviour, however only a few of them were made among customers in Sweden. Taken into consideration the fact that the choice of the supermarket might precede brand choice and that many buying decision are made in the retail store, the lack of attention in both environmental psychology and marketing is surprising. For instance, the biggest supermarket chain in Sweden, the ICA group, does not play background music in their stores. However, researchers (such as Yalch & Spangenberg, 1990; Oakes, 2000; Ballantine et al., 2010) found out that background music results in customers interest as well as arousal. In addition, according to Yalch and Spangenberg (1990), people tend to spend more time in the supermarket when background music is played. Thus, the environment of the supermarket has big potential to be a powerful and an effective

marketing tool if marketers would better understand how to utilize it. Since marketers and retailers want their customers to spend more time in the supermarket, it is necessary to create a relaxed and comfortable environment. What is more, the environmental factors are considered to be crucial elements that determine success for the marketers. Even if one has not been in a particular supermarket, ones first impression may have a great influence on buying intentions (Nguyen & Leblanc, 2002). Moreover, since buying decisions made in the supermarket can be influenced by the environmental factors, retailers and marketers should take as much advantage as possible (Converse & Spencer, 1942). The authors discovered plethora of studies upon the environmental factors and its affect on customers behaviour, which were conducted in various countries worldwide. However, not many studies concerned customers in Sweden. Since the lack of studies within this field in Sweden is surprising, the authors decided to examine the perception of the customers in Sweden towards the environmental factors in the supermarket.

1.3

Purpose

Retailers spend large amounts of money every year to build, design and refurbish stores. However, so far marketers are not engaged in systematic research that enables them to inspect the appropriate mix of the environmental factors that affect customers behaviour in supermarkets (Baker, Levy & Grewal, 1992). Thus, this paper extends the current literature review with the express purpose of providing results of quantitative research that can be utilized by marketers to understand which and how environmental factors influence customers behaviour in the supermarket. Moreover, the Mehrabian and Russell (1974) Stimulus-Organism-Response model, which helps to understand how different environmental factors influence customers behaviour in the supermarket, is going to be examined. Furthermore, another model that is going to be applied is the Bitners (1992) framework of servicescapes. This model has been used extensively in various studies, since it presents a framework of understanding the environment and behaviour relationships. The factors going to be examined are ambient (music, lightening, colour, noise), space/function (layout, equipment, furnishing) and signs, symbols and artifacts (visual communication such as: signage, personal artifacts) (Zeithmal, Bitner & Gremler, 2009). The paper will focus on the strongest supermarket chain in Sweden the ICA group, in precise ICA MAXI in Jnkping. The authors decided to conduct the study among ICA MAXI customers in Jnkping, since it is the largest supermarket of the ICA group in this city.

1.4

Research questions

Based on the problem discussion and in order to fulfil the thesis purpose, the authors formulated the following research questions: 1. What is the role of the perceived servicescape in the supermarket? 2. How does the perceived servicescape influence customers behaviour in the supermarket?

10

1.5

Delimitations

This paper presents a literature review on the environmental factors that influence customers behaviour only in supermarkets. The paper will analyse customers of the ICA group, since this supermarket chain has almost 50 per cent of the market share in Sweden. Other retailing stores will not be taken into consideration. Precisely, the empirical research is going to be conducted among ICA MAXI customers in Jnkping, Sweden. The role of the perceived servicescape that influences customers behaviour in the supermarket is going to be presented. Moreover, the way it influences the customers behaviour is going to be examined. Since this paper will focus only on customers perspective, the way employees respond to the environment is not going to be taken into account. Based on the Bitners (1992) framework of servicescapes, only ambient conditions, space/function and signs, symbols and artifacts factors are going to be examined upon this study.

1.6

Definitions

Axfood AB (Axfood) is a retailer and wholesaler of food products in Sweden. Axfood sells food products through both its subsidiaries and a collaboration of 650 proprietor-run stores (Dataminor, 2010). Coop Norden oversees the operation of approximately 1 100 discount stores, supermarkets, and hypermarkets throughout the Nordic region (Dataminor, 2010). Food retail market can be defined as the retail sales of all products, packed and unpacked, as well as beverages (Dataminor, 2010). Grocery store is a retail store that sells a variety of products, including some perishable items and general merchandise (FMI Glossary). Hypermarket is defined as a combined supermarket and discount store, at least 200 000 square feet (18 580,6 square meters) or larger, that sells a wide variety of food and general merchandise at a low price (FMI Glossary). ICA AB (ICA) is a Sweden based retail group that operates chain grocery stores in Sweden, Norway, and Latvia. The company is a joint venture group (Hakon Invest and Royal Ahold of the Netherlands) (Dataminor, 2010). Lidl (Lidl & Schwarz Stiftung) operates a chain of 6 800 grocery stores throughout Europe (in Germany it operates 3 100 stores). Main products in the stores are mostly under Lidls own brand. Lidl expanded to Denmark, Sweden, Hungary, Norway, Poland, and Slovenia (Dataminor, 2010). Perceived servicescape upon this paper includes such factors as music, noise, lightening, colour, visual communication (signage, personal artifacts), and layout, equipment and furnishing. Servicescape is the physical surroundings of a service company. It includes the exterior and interior design, ambient conditions such as temperature, noise, odour, as well as tangible parts of the service such as business cards, brochures, and other communication material (Reimer & Kuehn, 2005, p. 786).

11

Supermarket is a conventional grocery store, but not a warehouse club or mass merchant, with annual sales of $2 million or more per store (FMI). Visual communication consists of such elements as name, logo, retail identity, institutional signage, directional signage, departmental signage, category signage, POS signage, and lifestyle graphics situated in the supermarket. Visual communication program includes a range of messages, from large and bold directives, used sparingly to provide cues to the gross organization of the space, to the smaller, more specific, and plentiful messages that describe actual merchandise (Dunne & Lusch, 2008, p. 461).

Literature Review

The second chapter presents the theoretical framework that the method as well as the analysis are based on. The literature review contains the Mehrabian and Russell (1974) Stimulus-Organism-Response model and the Bitners (1992) framework of servicescapes with the description and a role of ambient conditions, space-function and signs, symbols and artifacts factors. The Mehrabian and Russell (1974) Stimulus-Organism-Response model (the M-R model) states that individuals respond emotionally to various environmental stimuli. The stimulus factors are physical features and can be described as store layout, colour, lightning, music, noise, sales promotions, shelf spacing, etc. According to the authors of this theory, an emotional response leads to an approach avoidance behaviour which includes emotional evaluations of the environment, evaluations of the service experience, longterm patronage decisions, attitudes toward others in the environment, spending behaviour, shopping time and exploration of the environment (Herrington & Capella, 1996, p. 27). Emotional states that are induced by the physical environment are pleasure and arousal. Pleasure refers to the degree where individual feels good, happy or satisfied in the environment. Arousal refers to the extent to which an individual feels stimulated, excited or active. Approach behaviour refers to the extent to which an individual is free to act or willingness to move such as propensity to buy (Baker et al., 1992). Avoidance behaviour, conversely, can be seen as a desire to leave the environment (Tombs & McColl-Kennedy, 2003). Approach avoidance behaviour will be described in more details further in sequence while analyzing individual behaviours in the Bitners (1992) framework of servicescapes (paragraph 2.4).

2.1

Servicescapes

Customers quite often rely on physical evidence or tangible cues in order to evaluate a company or services provided before the purchase. General elements of the physical evidence consist of companys physical facility or, so called, the servicescape and other forms of tangible communication. The servicescape elements that influence customers include facility exterior design (e.g. signage, parking, surrounding environment, landscape) and facility interior design (e.g. layout, equipment, signage, music and lighting) (Zeithaml et al., 2009). According to Zeithaml et al. (2009) the servicescape has an effect on the customers experience. It does not matter whether experience is mundane (e.g. bus ride), spectacular

12

(e.g. long adventure) or personally meaningful (e.g. child birth in a hospital), in all cases, the servicescape will affect the flow of the experience that leads to customers satisfaction and their emotional connection with the company delivering that experience. That is why, many companies rely on the servicescape in order to communicate and create customers experiences. Many articles offer various terms to define the servicescape. For example, Baker (1987) refers to the servicescape as physical environment, Turley and Milliman (2000) as marketing environment, Roy and Tai (2003) as store environment (Harris & Ezeh, 2008, p. 392). These different definitions have been used in the literature for many years. However, this paper will focus on the Bitners (1992) definition of the servicescape that refers to a design of physical environment which elicits internal reactions from customers such as approach or avoidance (cited in Harris & Ezeh, 2008, p. 392). Retailers want customers to spend a lot of time in supermarkets. In such cases, while shopping, customers also explore physical surroundings of the store. It goes without saying that the perceived quality of the servicescape is playing a substantial role in affecting whether customers are satisfied, which influences how long they stay in the supermarket. For example, while doing shopping, customers observe the layout as well as interior design of the supermarket, which might influence how long they are willing to stay in the store and if they return (Wakefield & Blodgett, 1996).
Figure 1. A Framework for Environment-Behaviour Relationship

Ambient Conditions Lightning, Music, Colours, Noise, Temperature, etc.

Space/Function Layout, Equipment, Furnishing, etc.

Signs, Symbols, and Artifacts Signage, Personal artifacts, etc.

Perceived servicescape

Customers responses

Cognitive Beliefs Categorization Symbolic meaning

Emotional Mood Attitude

Psychological Pain Comfort Movement Physical fit

Individual Behaviours Attraction Stay/explore Spend money Return Carry out plan

Source: Zeithaml, V. A., Bitner, M. J., & Gremler, D. D. (2009). Services Marketing. Integrating Customer Focus Across the Firm (5th ed.).

13

The Bitners (1992) servicescapes model is one of the most extensively used typologies of the service environment and its affect (Tombs & McColl-Kennedy, 2003). The framework (see Figure 1) explains how different environmental factors influence internal responses and external behaviours of customers. It suggests that various environmental factors are perceived by customers and they may respond cognitively, emotionally or physiologically. Internal responses influence customers behaviour and can be explained by the Mehrabian and Russell (1974) Stimulus-Organism-Response model (Bitner, 1992). Below, each part of the framework is going to be described.

2.2

Dimensions of the perceived servicescape

Bitner (1992) classifies three dimensions of the physical environment that influence customers holistic perception of the environment (e.g. perceived quality) and their internal responses (e.g. satisfaction with the environment) as well as external responses (e.g. approach-avoidance). Those dimensions are: (1) ambient conditions, (2) space/function, (3) signs, symbols and artifacts. In this study Ambient conditions consist of such factors as noise, lighting, music and colour. They influence the way people think, feel and respond to the particular factors. Spatial layout and functionality consist of factors such as layout, equipment, furnishings, and play an important role since, generally speaking, a main function of supermarkets is to accomplish exact purposes and needs of customers. Layout refers to how the equipment is placed, and functionality refers to how easy it is to navigate in the supermarket. According to Zeithaml et al. (2009) layout may influence customers satisfaction, their searching behaviour as well as the supermarket performance. Finally, signs, symbols and artifacts provide cues for customers and can be used as labels, for directional purposes and to inform about the behaviour in the supermarket. 2.2.1 a) Ambient conditions Music

When it comes to music, three dimensions can be taken into consideration, namely: (1) physical dimension, which consists of volume (loudness), pitch (succession of notes occurring overtime throughout a song), tempo (speed or rate of the rhythm), rhythm (pattern of accents given to beats or notes), (2) an emotional tone, and (3) the degree to which a customer likes the background music a preferential dimension (Mattila & Wirtz, 2001; Jain & Bagdare, 2011). Various researchers tackled music played in the retail store in relation to the influence on a customer, however, the results of their studies are mixed. In a study conducted by Ballantine et al. (2010), the authors find that medium volume music with well spaced sound sources has the desired result in increasing interest as well as arousal. Moreover, they state that music can also have a negative effect on customers when it is not used correctly. It is shown by Smith and Curnow (1966) that loud music causes that customers spend less time in the store. Additionally, dAstous (2000) claims that loud music is irritating (cited in Ballantine et al., 2010). Furthermore, slow tempo and low volume make retail patrons shop and spend more money in comparison with fast tempo and high volume music (Herrington & Capella, 1996).

14

Yalch and Spangenberg (1990) show that shoppers spent more time in the store than they planned while background music was played (cited in Morrison et al., 2011). Additionally, Yalch and Spangenberg (1993) show that customers can spend more money when the music closely matches their musical tastes (cited in Herrington & Capella, 1996). The Morrison et al. (2011) study shows that music and aroma influence on both customers arousal by increasing pleasure levels, and customers behaviour money and time spend in the store, satisfaction as well as approach behaviour. Other studies have confirmed that the influence of music can differ when it comes to a customers age. Yalch and Spangenberg (1988) claim that younger customers perceived that they had shopped longer when instrumental easy-listening (background music) was played, whereas elder customers thought that they had spent more time shopping when foreground music was played (Top 40). Therefore, the authors state that unfamiliar music may have adversely impact on customers time perception (cited in Mattila & Wirtz, 2001). Moreover, according to Linsen (1975) people like hearing music while shopping as they feel that the store with background music care about their shoppers. Furthermore, they can feel that time spend while shopping as well as in the checkout queue is shorter (cited in Herrington & Capella, 1996). Music is able to evoke behavioural and affective responses in customers. Milliman (1982; 1986) shows that in order to create pleasurable atmosphere in the store, background music should be smoothing (cited in Baker et al., 1992). In addition, the result of one of his experiment confirms that while slow tempo music is played in the retail store, customers spend more money. The other experiment conducted in the restaurant environment shows that customers spend more time on consuming food while slow tempo music than in the fast-music condition (cited in Mattila & Wirtz, 2001). Jain and Bagdare (2011) find that pleasurable shopping experience brings desired customer responses: (1) cognitive (e.g. spending, expectations, perceptions of time), and (2) behavioural (e.g. patronage, recommendations, speed or purchase decisions). Moreover, music effect on shopping experience is moderated by the store profile, customer profile, ambient factors as well as time of purchase. Oakes (2000) provides an extension of the Bitners (1992) model of servicescapes, namely musicscape. This framework focuses only on one of the ambient conditions the musical variable. Unlike the servicescape, musicscape examines only customers response and behaviour in the certain environment. b) Noise

Noise, or loudness of a sound can be seen very annoying and irritating among customers. Kryter (1985) presents two components of loudness such as amplitude or height of sound waves and frequency of sound waves. Too much noise is associated with negative stimulation. Moreover, Kryter (1985) states that too much noise can lead to irritability, decreased concentration or tension. Conversely, silence can be also seen very problematic (cited in Lin, 2004). c) Lightning

The first studies upon lightning in the retail environment confirm that customers are drawn towards light (Taylor & Sucov, 1974). Other studies show that light can draw attention to

15

products. Moreover, bright light causes products to be examined and touched more often than under dim lightning. Therefore, following Magnum (1998), lightning seems to emphasise the attractiveness of a product in the store (cited in Quartier et al., 2008). Additionally, Mehrabian and Russells (1974) study present that individuals are attracted by light sources. According to Baker et al. (1994), and Sharma and Stafford (2000), soft incandescent lighting is associated with a higher quality environment, whereas bright fluorescent lighting gives a discount image. In addition, researches present that people tend to be more positive and have lower levels of boredom in places with windows, in comparison to places without windows. Moreover, Areni and Kim (1993) claim that brighter stores encourage to analyze products more, but do not affect time spend in the store and sales (cited in Countryman & Jang, 2006). The influence of lightning in terms of its ability to arouse customers was investigated by Mehrabian (1976). He states, that on the one hand, an individual who is aroused, becomes stimulated and alert. On the other hand, a person can feel relaxed, calm or even sleepy in non-arousing environments. Moreover, in the first case, the aroused person will exhibit increased heart rate, muscle tension, and lowered skin temperature (Areni & Kim, 1994, p.118). According to Birren (1969), really bright light is recommended in order to stimulate purchases. Retail shop managers are recommended to employ soft lightning to reduce customer stimulation (such as arousal), and thus, slow the pace at which they move in the store. Therefore, managers can manipulate lightning in order to influence the time that customers spend while shopping (cited in Areni & Kim, 1994). d) Colour

According to Ballantine et al. (2010), limited colour pallet has a stronger impact on customer than a wide range of colours. Furthermore, using many colours at once may cause low effectiveness at highlighting displays. Colour can influence customers evaluation-related affect affective tone as well as activation-related affect arousal. Valdez and Mehrabian (1994) claim that blue (short wavelength, cool-coloured) environments are perceived to evoke better feelings than orange (long wavelength, warm-coloured) environments (cited in Babin et al., 2003; Yildirim et al., 2007). Thus, it can be concluded that interiors painted in cool colours will evoke more positive effects in customers, which may result in increase purchase decisions. Stones experiments (2003) show that stimulating colours which can interact with other instore factors are red colours, and blue colours are calming colours. Additionally, Canter (1974) mentions that red colours make things seem nearer and blue colours make them farther away (cited in Yildirim et al., 2007). Moreover, warm colours are associated with higher blood pressure, eyeblink frequency and respiratory rate (Gerard, 1957), inaccurate judgements of length, size (Goldstein, 1939), time (Smets, 1969), as well as speed of movement (Nakshian, 1964). Emotionally, red colours are associated with elated mood states (Schaie & Heiss, 1964), and arousal (Cahoon, 1969). Cool colours have been found to have the opposite effects. When it comes to the retail store, colours are used mainly to draw customers attention. Based on Bellizzi et al. (1983) study, warm colours environments draw customers attention, however, customers find warm colour environments as unpleasant, tense, negative as well as less attractive than cool colour environments (cited in Bellizzi & Hite, 1992).

16

2.2.2 a)

Signs, symbols and artifacts Visual communication (Signage, personal artifacts)

Items in the physical environment may serve either explicit or implicit signals in order to communicate about the place to its users. Among explicit communicators, signs displayed on the exterior and interior of a structure can be found. For example, labels such as the name of the company or department, signs for directional purposes (such as entrances or exits) as well as signs which communicate behaviour or rules (e.g. no smoking) (Bitner, 1992). Davis (1984) and Wener (1985) find this dimension as very complex since it includes numerous items in the environment that give different signals. Wener and Kaminoff (1982), and Nelson-Schulman (1984) state that different signs decrease perceived crowding and stress. Bitner (1992) claims that signs, symbols and artifacts affect approach behaviour (cited in Ang & Leoing, 1997). 2.2.3 a) Space/function conditions Layout, equipment and furnishing

When it comes to the spatial layout, one of the most important attribute is the effect of crowding. According to Eroglu and Machleit (1990) and Wakefield and Blodgett (1994), the spatial layout should minimize crowding that leads to a positive impression among customers (cited in Countryman & Jang, 2006). Empirical researches conducted by Harrell and Hutt (1976) and Hui and Bateson (1990 1991) find out that when layout is crowded, customers experience stress (cited in Ang & Leong, 1997). The furniture placement in the supermarket should assign to spatial movement, function as walls or communicate as boundaries. For instance, changes in ceiling heights can influence spatial perception more than changing the place of furnishings. According to Ching (1996), high ceilings are associated with a feeling of spaciousness while low ceilings convey feelings of cosiness and intimacy (cited in Lin, 2004). The width of the aisle is another in-store factor that has an impact on the customer. Large open spaces encourage customers to wandering around and make them more comfortable. However, too little or too much space has a negative outcome (Ballantine et al., 2010). Following Thang and Tan (2003), decorations as well as display of merchandise, ease of movement in the store or layout have an important affect on the store preference (cited in Ballantine et al., 2010). When it comes to layout, the main task is to guide the customer throughout the store, thus it fails to be recognized as a positive attractor cue (Ballantine et al., 2010). Product display features in connection with other factors such as space, colour or lightning can draw the customers attention. If the spacing of merchandises in a display is too small, then products seem to be cluttered and cause a negative impact on customers (Ballantine et al., 2010).

17

2.2.4

Connected stimuli

Since the customer does not perceive only one cue (only colours or music) in the store, a combination of factors has a greater influence on a shopper. For example, the combination of fluorescent, bright lights and popular background music is perceived by the customers as a discount image, on the other hand soft lights and classical background music are associated with a prestige image and customers are aware to expect higher prices in this store. Studies suggest that bright light in connection with warm colours create a discount store concept (Babin et al., 2003). Moreover, low-image stores are perceived to have grid layout, uncooperative as well as sloppily dressed salespeople. High-image stores, on the other hand have the opposite features cooperative and nicely dressed sales people and open layout (Grewal & Baker, 1994).

2.3

Internal responses: cognitive, emotional and physiological

According to the Bitners (1992) framework of servicescapes, customers respond to the environment either cognitively, emotionally or physiologically. Perceptions of the servicescape might lead to a certain beliefs or emotions that influence individuals behaviour. Thus, cognitive, emotional and physiological responses will be presented independently. 2.3.1 Cognition and environment

The perceived servicescape might cause cognitive responses that have an effect on peoples beliefs about a place and products found in that place. According to Broadbent, Bunt and Jencks (1980) and Rapoport (1982), the servicescape can be seen as a part of nonverbal communication (cited in Bitner, 1992). For instance, specific cues in the environment such as interior design or decor may influence customers beliefs whether the store is expensive or successful. Bitner (1990) in a customer study finds out that decor in travel agents office influences customers attribution for the travel agents behaviour (cited in Bitner, 1992). What is more, the perception of the servicescape helps customers to distinguish companies by the way it is categorized. Research shows that in the restaurant industry, customers distinguish them according to fast food or elegant restaurants (Zeithaml et al., 2009). In relation to the field of this paper, it can be said that customers classify supermarkets for example into those that have relatively high prices and variety of products, offer great customer service and those that have comparatively lower prices and low diversification of products. 2.3.2 Emotion and environment

In addition to cognitive responses, the perceived servicescape can also elicit emotional responses (Zeithaml et al., 2009). For example, when being in a particular place, one can feel happy and relaxed, whereas other places can make one depressed or gloomy. Therefore, the environmental factors such as music, colours, lighting, decor may have a very powerful and unexplainable influence on customers moods in the particular place. Environmental psychologists Mehrabian and Russell (1974) classify emotional responses to environment into two dimensions such as pleasure/displeasure and degree of arousal (cited in Bitner, 1992). For example, places that cause feeling of pleasure are ones where people spend more time and money, while unpleasant places are usually avoided. Likewise, an arousing environment is seen positively, unless the excitement is connected with

18

unpleasantness. Moreover, unpleasant places that are high in arousal such as too noisy or confusing are usually avoided as well (Bitner, 1992). According to Obermiller and Bitners studies, people who buy products in emotionally pleasant environment evaluate products more positively (cited in Bitner, 1992). 2.3.3 Physiology and environment

The perceived servicescape can elicit physiological responses as well. Too much noise or inappropriate temperature in the place may cause physical discomfort, the quality of the air can result in a difficulty to breathe, low lighting might decrease ability to see products and refer to eyes pain. All physical responses may influence customers behaviour whether they stay and enjoy the environment of the store (Zeithaml et al., 2009). 2.3.4 Variations in internal responses

As it was explained above, people respond to the environment in three ways cognitive, emotional and physiological and those responses affect the way people behave in the particular place. However, it should be also taken into consideration that not all individuals act the same way. Cultural and personality differences, temporary conditions such as the purpose for being in the store or mood can result in variations, or in other words, how people respond to the servicescape (Zeithaml et al., 2009). Arousal seeking is one personality trait that influences how people respond to the servicescape. Usually, arousal seekers take pleasure in and look for high levels of stimulation, while other people rather choose lower levels of stimulation and are arousal avoiders. In addition, the purpose for being in the particular environment can also influence customers. Of course, when it comes to supermarkets, it goes without saying that people do shopping there. However, some individuals may come just to buy milk or bread, whereas others can spend hours and enjoy the shopping. A person who is in the supermarket for a couple of minutes, probably will be less affected by the environment. Finally, cultural differences may also affect customers responses in the particular environment. For instance, in China there is a strong cultural preference for a red colour. However, Western countries would not use this colour in that extent. Moreover, Americans prefer to shop in more quiet and orderly supermarkets, while Indians like more chaotic environment (Zeithaml et al., 2009).

2.4

Individual behaviours

According to environmental physiologists Mehrabian and Russell (1974), there are two forms of individuals behaviour: (1) approach and (2) avoidance. In this section, approach avoidance behaviour will be described more precisely. 2.4.1 The M-R Stimulus-Organism-Response model The Stimulus-Organism-Response (S-O-R) paradigm relates features of the environment (S) to approach-avoidance behaviour (R) in the environment. There are two assumptions in this model: (1) persons emotions and feelings determine what they do as well as how they do it, and (2) persons respond with various sets of emotions to a certain environment and

19

respond either by approach or avoid the environment (Tai & Fung, 1997; Ward, Davies & Kooijman, 2007). The Mehrabian and Russells (1974) model has been widely used by various researchers (e.g. Donovan & Rossiter, 1982; Donovan, Rossiter, Marcoolyn & Nesdale, 1994; Tai & Fung, 1997; Foxall, 1997; Ward et al., 2007; Kenhove & Desrumaux, 1997; Spangenberg, Sprott, Grohmann & Tracy, 2006; Kearney, Kennedy & Coughlan, 2007). Mehrabian and Russell (1974) adopted a Stimulus-Organism-Response (S-O-R) approach (see Figure 2). This model proposes a measure of the environment (S) in terms of information rate a measure of novelty and complexity (Donovan et al., 1994, p. 284).
Figure 2. The Mehrabian and Russell (1974) model

Environmental Stimuli (S) e.g. colour, temperature, music

Emotional States (O) Pleasure Arousal Dominance

Behavioural Response (R) Approach Avoidance

Source: Tombs, A., & McColl-Kennedy, J. (2003). Social-servicescape conceptual model. Marketing Theory, Vol. 3, No. 4, p. 447-475.

Figure 2 shows that environmental stimuli (S) have an effect on customer behaviour (R) by organism (O) which is mediated by the customers internal state (Quartier, Vanrie & Cleempoel, n.d.). In other words, environmental factors are said to have an influence on emotional states (O) pleasure, arousal and dominance. The emotional states component has an influence on the (R) factor, namely on the approach-avoidance behaviour (Kearney et al., 2007). All components are presented in more details below. 2.4.2 Environmental stimulus (S)

Stimuli can be defined as those determinants that have an influence on an individuals internal states (Chang, Eckman & Yan, 2011). Store image and atmosphere components can be describe by using environmental stimuli such as design, odours, music, colours, or layout (Dennis & Newman, 2005). However, it is worth to mention, that it is not effortlessly to measure the stimuli factors, since in the certain environment there are many stimuli involved (Tai & Fung, 1997). 2.4.3 Emotional states (O)

Organism refers to internal processes and structures intervening between stimuli external to the person and the final actions, reactions, or responses emitted (Chang et al., 2011, p. 235). According to the M-R model, there are three emotional states (PAD). Pleasure refers to the extent to which a person feels good, joyful, happy or satisfied in the environment. The term arousal relates to the extent to which a person feels excited or stimulated, alert

20

or active in a situation (Tai & Fung, 1997, p. 317). Based on this model, it can be said that every environment determines different emotional states in the individual. In the Mehrabian and Russells (1974) study, pleasure was measured by respondents verbal evaluation of their responses to environment as following (Soriano, Foxall & Pearson, 2001): Happy Unhappy, Pleased Annoyed, Satisfied Unsatisfied, Contented Melancholic, Hopeful Despairing, Relaxed Bored. Arousal was indicated by respondents feelings of being (Soriano, Foxall & Pearson, 2001): Stimulated Relaxed, Excited Calm, Frenzied Sluggish, Jittery Dull, Wide-awake Sleepy, Aroused Unaroused. Dominance was verbally assessed as the extent to which respondents report feeling (Soriano, Foxall & Pearson, 2001): Controlling Controlled, Influential Influenced, In control Cared-for, Important Awed, Dominant Submissive, Autonomous Guided. Mehrabian and Russell (1974) state that (1) pleasure could be crucially related to approachavoidance behavioural measures, (2) arousal could be negatively related in unpleasant environments and positively related in pleasant environments, and (3) dominance could be positively related to approach behaviour. However, the dominance dimension state has been deleted in subsequent studies based on the M-R model since there is a lack of empirical support (Tai & Fung, 1997; Sweeney & Wyber, 2002). Based on Donovan and Rossiter (1982) findings, pleasure is a significant factor of approach-avoidance behaviour when it comes to the store. Pleasure may result for example in spending more than anticipated (Sweeney & Wyber, 2002). Baker et al. (1992) claim that both pleasure and arousal have a positive impact on the willingness to buy (cited in Sweeney & Wyber, 2002). An example when arousal might be evoked is using fast tempo music, additionally customers can feel pleasure while observing pleasant video images (Dennis & Newman, 2005).

21

2.4.4

Behaviour responses (R)

Responses are the final results of the process and represent the final customers decisions (Chang et al., 2011). Responses can be divided into two types: (1) approach, and (2) avoidance behaviour. More precisely, a customers response is the extent to which customers either approach or avoid perceived stimuli. As far as shopping behaviour is concerned, approach can be measured for example by (1) spending, (2) time in a store, (3) intention to revisit, or (4) prices number of items bought (Dennis & Newman, 2005). 2.4.5 Studies based on the M-R model

The first who applied the Mehrabian and Russell (1974) model to the retail store environment were Donovan and Rossiter (1982). According to Kearney et al. (2007) this study made a significant contribution to the literature as the relationship between behavioural intentions, emotional states and the store environment have been established. Donovan and Rossiters (1982) study shows that only two dimensions, namely pleasure and arousal, are able to predict customers behaviour in the retail store (Mattila & Wirtz, 2001). Table 1 presents response taxonomy in Donovan and Rossiters (1982) study in the retail environment.
Table 1. Donovan and Rossiters (1982) study Approach The desire to physically stay in the environment The willingness to explore the environment Avoidance Relation The desire to leave the Store patronage intentions environment Tendency to remain The degree of in-store searching and the inanimate with the range of merchandise to which the surroundings customer allows themselves to be exposed The desire to communicate Tendency to avoid any The interaction with the sales staff and with others in the store interaction with others other customers The degree of enhancement The degree of hindrance that Satisfaction and repeat shopping that the environment gives to the environment gives to frequency, as well as the amount of time solving problems solving problems and money spent in the store Source: Tombs, A., & McColl-Kennedy, J. (2003). Social-servicescape conceptual model. Marketing Theory, Vol. 3, No. 4, p. 447-475.

When it comes to an arousal state, studies present different results. According to Sherman, Mathur and Smith (1997), arousal increases spending and purchases intentions. Interestingly, Millimans (1982) research shows that arousal has the opposite effect and Smith and Curnow (1966) do not find any effect. Moreover, Sherman et al. state that arousal increases also visit duration, on the other hand Smith and Curnow (1966) find the opposite. In Donovan and Rossiters (1982) study conducted in the retail stores, it is confirmed that arousal has a positive effect on dependent variables in the pleasant environment in the retail store. However, Donovans (1994) research in discount stores does not confirm this relationship (cited in Kaltcheva & Weitz, 2006). Thus, it can be concluded that pleasantness has a positive effect on customers buying behaviour, however, the impact of arousal differ across studies.

22

Methodology

The third chapter aims to describe the chosen method quantitative research. Relevant theories, used as the framework for the method, are provided in order to gain a deeper understanding of the chosen research method. Therefore, secondary and primary data collection, non-probability sampling technique and data analysis are presented. Moreover, the population is defined and the structure of the questionnaire is provided. Finally, validity and reliability of the research are discussed.

3.1

The research design

According to Green and Tull (1970), a research design is the specification of methods and procedures for acquiring the information needed. It is the over-all operational pattern or a framework of the project that stipulates what information is to be collected from which source by what procedures (cited in Beri, 2008, p. 61). Malhotra and Birks (2007) describe the research design as a framework or blueprint for conducting a marketing research project. The research design specifies which procedures are necessary to be carried out in order to solve a research problem. It indicates precise details and the practical aspects of implementing approach to a problem. In simple words, the research design can be described as a work plan. The plan which identifies what has to be done in order to solve the research problem. The function of the research design is to ensure that the evidence obtained enables us to answer the initial question as unambiguously as possible (Vaus, 2005, p. 9). A good research design guarantees an effective and efficient research project. In addition, it gives a direction for a researcher and creates so-called time table. The time table ensures that different procedures will be completed within a certain time limit.

3.2

Quantitative research method

According to Creswell (1994), a quantitative research is an explanation of a phenomenon by gathering numerical data and analyzing it based on mathematical methods. Creswells definition basically states that numerical data is collected in the quantitative research and can be analyzed by using statistics (cited in Sukamolson, n.d.). The main purpose of the quantitative research is to come up with objective and valid descriptions of the phenomenon. In the thesis, the quantitative method was chosen, since it is used to explain some kind of phenomenon. In relation to the purpose of this thesis, the phenomenon is what is the role of the perceived servicescape that affect customers behaviour in the supermarket? What is more, the quantitative research method is widely used while examining customers attitudes and opinions. Since the authors wanted to investigate customers perception of the servicescape in ICA MAXI, the quantitative research method was justified to be used in this study. Additionally, one of the main advantages of using quantitative research is its clarity, since numbers can be easier to understand and interpret in comparison with hundreds of coding categories. Moreover, quantitative studies are usually designed around the use of measures where validity and reliability have been already established (Taylor, 2005).

23

3.3

Descriptive research

Descriptive research is used to draw a picture of the topic and describe characteristics of the phenomenon. In other words, it describes and interprets the present (Taylor, 2005). Descriptive research can be also used to investigate and solve many current problems which may occur in a company (e.g. a decrease in sales). Solutions of those problems can be very useful for the directions in the future as well as relevant information how to reach designated goals. Usually descriptive research gives an adequate description of the phenomenon that supposed to be studied (Taylor, 2005). Descriptive research is mainly used when specific research questions or hypotheses are formulated. Additionally, it is a pre-planned and structured research (Malhotra & Birks, 2007).

3.4

The target population

A population is elements which share several common set of characteristics. In order to obtain primary data, the population should be precisely defined, as imprecise definition of the population can result in an ineffective as well as misleading research (Malhotra & Birks, 2007). In this research, the target population was customers of ICA MAXI in Jnkping, Sweden. More precisely, the population consisted of female and male ICA MAXI customers aged from 18 years old. The authors did not make distinction when it comes to nationalities, as the main objective of this research was to examine the role of the perceived servicescape that can influence customers behaviour regardless of cultural differences. Since the questionnaire was available in both English and Swedish versions, all ICA MAXI customers could be potential respondents.

3.5

Sampling technique

The research was a cross-sectional study, which means it involved the collection of information from any given sample of population elements only once. The cross-sectional study is one of the most often used descriptive design in marketing research. There are two sampling techniques: probability and non-probability techniques. In probability techniques units are selected by chance. In this technique, it is possible to specify the probability of selecting a particular sample of a given size. The second techniques is non-probability sampling. This sampling technique relies on the personal judgment of the researcher more than on the chance to select sample elements. Among non-probability techniques Malhotra and Birks (2007) enumerate: (1) convenience sampling, (2) judgemental sampling, (3) quota sampling, and (4) snowball sampling. In this research, two non-probability sampling were used, a judgemental and a snowballing sampling in precise. In the judgemental sampling, the units were selected based on the researchers judgement about which units would be the most suitable, useful or representative in the research. In the snowball sampling, initial respondents who possess the desired characteristic of the target population were selected. After filling in the questionnaire, they were asked to send the link to the questionnaire to people who might also have the desired characteristic. Therefore, subsequent respondents were selected based on the referrals (Babbie, 2007). When it comes to personal interviewing, the judgemental

24

sampling was used, whereas the snowball sampling was implemented as a method of collecting the responses via electronic collection.

3.6

Data collection

In the thesis, both primary and secondary data were retrieved. According to Malhotra and Birks (2007), primary data is data collected by a researcher for a specific purpose of addressing the research problem at hand. Moreover, primary data is used when available secondary data is insufficient to fulfil the scientific research purpose (Ghauri & Grnhaug, 1995). Primary data in comparison with secondary data is more expensive to obtain and more time-consuming while collecting as well as analyzing (Malhotra & Birks, 2007). Secondary data is data collected for purposes other than the research problem at hand. It is used not only in order to find information to solve the research problem but as well as to understand and explain the problem (Ghauri & Grnhaug, 1995). Secondary data used in this paper was external data. The authors gathered the theoretical framework found in books, journal articles, online databases etc. Since the authors attempted to examine the role of the perceived servicescape in the supermarket, more precisely in ICA MAXI in Jnkping, and the approach was conducted from customers perspective, obtaining primary data required interaction with ICA MAXI customers. Primary data was collected based on a survey, a structured questionnaire, created by means of Qualtrics. As the authors of this paper decided to conduct both personal interviews and electronic collection, non-probability sampling techniques described above seem to be justified. When it comes to personal interviews, street interviews were used. The responses from ICA MAXI customers in Jnkping were collected by the authors in the middle of April 2012. The second technique that was conducted was electronic collection. The Qualtrics link was mailed via the Internet, and shared in social media tool, Facebook in precise. In order to ensure that the right people participated in the survey, a filter question was asked. In order to obtain more answers from Swedish respondents, the authors decided to collect data using both versions of the questionnaire among ICA MAXI customers Swedish and English. Collecting data via personal interviews does not require an advanced level of the language as in the other methods such as focus groups or depth interviews. Since the authors do not speak Swedish fluently, the questionnaire was translated into Swedish by Swedish speaking persons. The questionnaire was created in both languages, as the authors assume that younger generations have a satisfactory knowledge of English language and it is easier for them to fill in the questionnaire via Qualtrics link, whereas elder customers might feel more comfortable while filling in the printed version of the questionnaire translated into Swedish. The authors decided to collect a total sample size of at least 100 responses from ICA MAXI customers, from both personal and electronic collection. The number of responses was also conditioned by the recommended number of answers while conducting factor analysis in SPSS.

25

3.7

Questionnaire

According to Dillman (2007), three types of data can be obtained through collecting questionnaires responses, namely: opinion, behaviour, and attribute (cited in Saunders, Lewis & Thornhill, 2009). Since this research is going to examine ICA MAXI customers behaviour, the research based on the questionnaire seems to be justified. According to Malhotra and Birks (2007) there are ten steps that a researcher has to take into consideration while designing a questionnaire: (1) specify the information needed, (2) specify the type of interviewing method, (3) determine the content or individual questions, (4) overcome the respondents inability and unwillingness to answer, (5) choose question structure, (6) choose question wording, (7) arrange the questions in proper order, (8) identify the form and layout, (9) reproduce the questionnaire, and (10) eliminate problems by pilot-testing. The questionnaire within the thesis consisted mainly of structured questions. Moreover, a filter question as well as dichotomous questions were asked. The questionnaire was composed of five types of questions which were created by the authors of this study. Questions were related to: (1) ambient conditions, (2) space/function, (3) signs, symbols and artifacts, (4) customers behaviour in the supermarket, and (5) demographic questions. Questions in the first section were asked in order to understand what respondents think about colours, lighting and the noise level in ICA MAXI. What is more, respondents also had to express their opinion whether they would like to hear music in ICA MAXI. The purpose of the second section of the questionnaire was to examine whether customers are satisfied with the layout of ICA MAXI. Questions included in the third section were formed to analyze respondents opinion when it comes to directional signs and other types of communication, displayed in ICA MAXI. The main idea of the fourth section was to analyze general customers attitude towards ICA MAXI. The demographic questions were asked in order to draw a profile of customers of ICA MAXI. The used scale was non-comparative, since respondents did not have to compare the object being rated with, for example, another object. What is more, the respondents evaluated only one object at a time. Among non-comparative scales, itemised rating scales, more precisely, Likert scale was used. Likert scale allows the respondent to choose a degree of disagreement or agreement with each of the statements when it comes to the stimulus object (Malhotra & Birks, 2007). In the research, 7-point Likert scale was adopted. Respondents were asked to indicate in each statement a degree they agree or disagree with particular statements from strongly disagree (=1) to strongly agree (=7). The questionnaire was created in Qualtrics. The authors decided to create the questionnaire in two versions, English and Swedish see Appendices p. 75-82. The printed version of the questionnaire that was used during personal interviews was available only in Swedish. 3.7.1 Pre-Test

Pilot-testing is used in order to test the questionnaire among a small sample of respondents. The main objective of a pre-test is to eliminate potential problems. The questionnaire can be tested in terms of question content, wording, sequence, instructions, questions difficulty, form and layout. According to Malhotra and Birks (2007), the

26

respondents in the pilot-test should be chosen based on the same background characteristics as well as familiarity with the topic being investigated. In this research, the pilot-test was conducted among 10 students from Jnkping International Business School as they possess similar background characteristics and familiarity with the examined topic as defined before population (they are customers of ICA MAXI). Since the questionnaire was provided in both Swedish and English versions, the authors decided to conduct the pilot-test among 5 international students and 5 Swedish students in order to eliminate inexact, ambiguous and translations issues.

3.8

Data analysis

The data was analysed using SPSS data entry software. In the thesis, two multivariate techniques were conducted. These techniques are suitable for analyzing data when there is more than one measurement of each element and the variables are analyzed simultaneously (Malhotra & Birks, 2007). More precisely, factor analysis and multiple regression were conducted. Factor analysis is an interdependence technique that measures the relationship between variables since there is no distinction between independent and dependent variables. It is often use for reduction and summarisation. The variables that were investigated are following: ambient conditions, space/function and signs, symbols and artifacts. Multiple regression is a general as well as an flexible data analytic system. According to Cohen, Cohen, West and Aiken (2003), it can be conducted when a quantitative variable (a dependent variable) is to be studied in relationship to any factors of interest (independent variables). In this case the dependent variable was customers behaviour in ICA MAXI, and the rest of the variables were independent variables.

3.9

Trustworthiness and relevance of information

In order to gather trustworthiness and relevant information, the reliability and validity of the research should be taken into account. As an important issue in the research is obtaining valid and reliable information, the research should be free of errors. Hence, these two issues are presented below. Validity refers to the degree to which a tool measures what it is supposed to measure (Malhotra & Birks, 2007). Furthermore, in terms of validity, a comparison of the purpose of the conducted research and the conclusion should be done. In this case the purpose is to examine the role of the perceived servicescape that can influence customers behaviour in the supermarket. Reliability refers to the degree which the tool (used upon similar conditions) measures the feature under investigation (Carter & Porter, 2000). A tool that was used to measure reliability was coefficient alpha (Cronbachs alpha). The coefficient varies from 0 to 1, and a score above 0,6 indicates satisfactory internal consistency reliability (Malhotra & Birks, 2007).

27

A relationship between validity and reliability exists. Firstly, if a measure is perfectly valid, it is perfectly reliable. Secondly, if a measure is unreliable, it cannot be perfectly valid. Finally, when a measure is perfectly reliable, it may or may not be perfectly valid. Therefore, reliability is a necessary, but not sufficient condition for validity (Carter & Porter, 2000).

Findings

The fourth chapter gives detailed information about the findings of the conducted research. Firstly, the reliability of ambient conditions, space/function, sign, symbols and artifacts, and behaviour factors is presented as well as the relationship between these factors. Furthermore, the results of descriptive statistics, T-tests and one-way ANOVAs are shown. Finally, the findings based on the factor analysis and the multiple regression end the chapter. A total of 162 responses were collected in two ways: online responses 61, and printed questionnaires 101. However, 20 surveys were deleted from the data set due to incompleteness or respondents were not customers of ICA MAXI in Jnkping. Thus, 142 (87,65 per cent) responses were used for the further analysis. In general, two major age groups were respondents aged 18-29 (47,9 per cent) and above 49 years old (21,7 per cent). As was stated before, the authors assumed that filling in the online questionnaire by a young generation would be more probable than by elder customers of the supermarket. The research showed that among 41 responses collected via Qualtrics link, 35 respondents were aged between 18-29 (85 per cent of the electronic collection and 21,6 per cent of the total number of the respondents), whereas only 3 respondents were above 40 years old (7 per cent of the electronic collection and 1,85 per cent of the total number of the respondents). Additionally, respondents above 40 years old (45 respondents) that filled in the paper version of the questionnaire accounted for 44,55 per cent of the personal interviews and 27,78 per cent of the total number of the respondents. When it comes to the employment status, 60,1 per cent of the participants were part time workers, 40,6 per cent were full time workers and 30,8 per cent were students. Due to the fact that the respondents were able to choose more than one option within this question, the total percentage exceed 100 per cent. Furthermore, female respondents accounted for 52,4 per cent, whereas male respondents 47,6 per cent. In relation to monthly disposable income and nationality, only 3 questionnaires had missing data. The most common answers of monthly disposable income were below 10.000 Swedish crowns (35,7 per cent), between 20.001-30.000 Swedish crowns (21,7 per cent), followed by 10.000-20.000 Swedish crowns (20,3 per cent). Swedish respondents scored 69,9 per cent, whereas international participants 28 per cent. Table 2 presents detailed demographics of the respondents.

28

Table 2. Demographics
Demographics Age Total Missing data Full time worker Part time worker Retired Student Unemployed Answers 18-29 years old 30-39 years old 40-49 years old >49 years old Frequency 68 27 17 31 143 58 23 21 44 9 155 68 75 143 51 29 31 25 4 140 3 100 40 140 3 Percent (%) 47.6 % 18.9 % 11.9 % 21.7 % 100 % 40.6 % 60.1 % 14.7 % 30.8 % 6.3 % 152.5 % 47.6 % 52.4 % 100 % 35.7 % 20.3 % 21.7 % 17.5 % 2.8 % 97.9 % 2.1 % 69.9 % 28 % 97.9 % 2.1 %

Employment status

Total Missing data Gender Total Missing data < 10.000 kr 10.000-20.000 kr 20.001-30.000 kr 30.001-40.000 kr > 40.000 kr Male Female

Monthly disposable income

Total Missing data Nationality Swedish International

Total Missing data Source: Own composition based on the questionnaire outcome.

4.1

Reversing negatively worded questions

Wording in some questions has to be reversed in order to prevent response bias (Pallant, 2005). For example, the question I enjoy shopping in ICA MAXI is worded in a positive way. However, the question such as This is a place that I would avoid shopping in the future is negatively worded. The negatively worded questions should be reversed in order to ensure that all questions will be scored at the high scores, that indicates high level of agreement. In this case, three questions were reversed: The lighting in ICA MAXI is too dark, Colours used in ICA MAXI are too bright and This is a place that I would avoid shopping in the future.

4.2

Reliability

The most popular tool to measure internal consistency, hence, how the items are connected with each other, is a Cronbachs Alpha coefficient. This indicator is sensitive to a number of items included in the scale, and provides an average correlation of the all items used in the scale. The more items are included in the factor, the higher the Cronbachs Alpha, which leads to higher reliability (Pallant, 2005). In order to check whether data is reliable, Cronbachs Alphas were calculated for each factor: ambient conditions,

29

space/function, signs, symbols and artifacts, behaviour as well as for all variables in total. When the value of the Cronbachs Alpha is above 0,6, the scale can be considered as reliable within the sample (Malhotra & Birks, 2007).
Table 3. Cronbachs Alphas Factor Cronbach's Alpha N of Items Interpretation High reliability High reliability High reliability High reliability High reliability % of the deviance in the score 80,2 71,9 88,1 77,9 89,5

Ambient 0,802 11 conditions Space/function 0,719 6 Signs, symbols 0,881 5 and artifacts Behaviour 0,779 5 All factors 0,895 27 Source: Own composition based on the SPSS outcomes.

The scores of the Cronbachs Alphas for the ambient conditions, space/function, signs, symbols and artifacts, behaviour, and all components in total (Table 3) formed a high reliability which leads to acceptable scales (respectively 0,802; 0,719; 0,881; 0,779; 0,895). It can be concluded that the higher the respondents scored the questions about ambient conditions, space/function, signs, symbols and artifacts, behaviour, and all components in total, the higher their intensions to stay or explore ICA MAXI. Reliability statistics of ambient conditions, space/function, signs, symbols and artifacts, behaviour, and all components in total (Table 3) showed respectively 80,2 per cent, 71,9 per cent, 88,1 per cent, 77,9 per cent, and 89,5 per cent of the deviance in these scores by combining respectively 11, 6, 5, 5, and 27 items which will be considered as a true score variance or internally consistent reliable variance. Tables A. 1 A. 5 in the Appendix 1 provide detailed information about correlation between different items within ambient conditions, space/function, signs, symbols and artifacts, behaviour, and all components in total. Table 4, shows that the factors described above are correlated with each other and create a positive correlation all scores are between 0 (no correlation) and 1 (a positive correlation), which means that as one variable increases, so too does the other. The size of the absolute value provides an indication of the strength of the relationship. A perfect correlation of 1 or -1 indicates that the value of one variable can be determined exactly by knowing the value on the other variable (Pallant, 2005, p. 121).
Table 4. Correlations Ambient conditions Ambient conditions Space/ function Signs, symbols and artifacts Behaviour Spearmans rho Correlation Coefficient Correlation Coefficient Correlation Coefficient Correlation Coefficient 1,000 ,442** ,407** ,518** Space/ function ,442** 1,000 ,501** ,592** Signs, symbols and artifacts ,407** ,501** 1,000 ,595** Behaviour ,518** ,592** ,595** 1,000

Source: Own composition based on SPSS outcome.

**. Correlation is significant at the 0.01 level (2-tailed).

30

According to Table 4 the strongest correlation exists between signs, symbols and artifacts and behaviour factors (0,595) and between space/function and behaviour factors (0,592). On the other hand, the weakest relationship is between signs, symbols and artifacts and ambient conditions factors (0,407). According to Cohen (1988) guidelines, the first and the second scores create large relationships between variables, and the third one stands for a medium strong relationship (cited in Pallant, 2005). Moreover, the first relationship shows that signs, symbols and artifacts factor helps to explain 35,4 per cent (0,595x0,595x100%=34,4%) of the variance in respondents scores on the behaviour scale.

4.3

Descriptive statistics

Descriptive statistics is a good way to summarize numerical data to make it easier to interpret, moreover, it includes mean and standard deviation. The 7-point Likert scale was used in the questionnaire, where 1 means strongly disagree and 7 strongly agree. Table 5 presents means and standard deviations of all the statements.
Table 5. Descriptive Statistics Descriptive Statistics I would prefer if ICA MAXI would play music. The lighting in ICA MAXI is-Appropriate The lighting in ICA MAXI is-Inviting The lighting in ICA MAXI is-Comfortable The lighting in ICA MAXI is-Too dark Colours used in ICA MAXI are-Appropriate Colours used in ICA MAXI are-Soothing Colours used in ICA MAXI are-Pleasant Colours used in ICA MAXI are-Attractive Colours used in ICA MAXI are-Too bright The noise level in ICA MAXI is acceptable for me. It is easy to move around ICA MAXI. It is easy to move in ICA MAXI with the shopping cart between aisles. The supermarket layout is logical to me. I can find everything I need in ICA MAXI. I think that ICA MAXI has too many products for the size of the store. The interior design of ICA MAXI is visually appealing. There are enough directional signs (e.g. entrance, exit, toilets) in ICA MAXI. Q19 The directional signs (e.g. entrance, exit, toilets) give guidelines how to move in ICA MAXI. Q20 It is easy to follow directional signs (e.g. entrance, exit, toilets) displayed in ICA MAXI. Q21 Directional signs are visually appealing. Q22 Other types of communication such as brochures are visually appealing. Q23 I enjoy shopping in ICA MAXI. Q24 I am satisfied with the general impression of ICA MAXI. Q25 I like the environment of ICA MAXI. Q26 I would like to spend more time browsing in ICA MAXI. Q27 This is a place that I would avoid shopping in the future. Valid N (listwise) Source: SPSS outcome. Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14 Q15 Q16 Q17 Q18 N Mean 142 4,25 141 6,00 142 5,42 137 5,58 141 5,96 141 5,68 141 5,05 139 5,42 142 4,95 142 5,68 143 4,93 143 5,55 143 5,55 143 5,15 143 5,09 143 2,64 142 4,88 143 143 143 143 141 143 143 143 141 142 122 5,27 5,22 5,50 4,63 4,74 5,69 5,73 5,52 3,27 5,87 Std. Deviation 2,161 1,293 1,508 1,453 1,581 1,322 1,666 1,537 1,731 1,695 1,452 1,314 1,382 1,553 1,574 1,844 1,407 1,506 1,371 1,278 1,341 1,442 1,301 1,132 1,233 1,756 1,621

31

The mean is one of the most commonly used measure to examine the central tendency of the responses. In this case, the statement The lighting in ICA MAXI is appropriate has a mean of 6,0. It indicates, that the respondents rather agreed and had a positive opinion towards this question. On the other hand, I think that ICA MAXI has too many products for the size of the store has a mean of 2,64 which points out that the majority of the respondents rather disagreed with this statement. Standard deviation measures the spread of a set of observations. The larger the standard deviation was, the more spread out the observations. A low score means that the responses were concentrated, hence, the respondents had the same opinion about a statement and scored similar. For instance, I am satisfied with the general impression of ICA MAXI has the standard deviation of 1,132. On the other hand, the higher standard deviation means that the respondents had different opinion towards the same statement. For example, the statement I would prefer if ICA MAXI would play music, formed the standard deviation of 2,161. The respondents had different opinions (high score of the standard deviation) when it comes to the used colours (soothing (Q7), attractive (Q9), too bright (Q10)), the variety of available products (Q16) as well as time spent in the supermarket (Q26) and their patronage decision (Q27). On the other hand, the responses about the general impression of the store (Q24) and its environment (Q25) as well as ease following directional signs (Q20) were concentrated (low score of the standard deviation).

4.4

Analysing the significant differences

In order to analyse whether the significant differences exist between customers evaluation of various statements and factors, T-tests and One-way ANOVAs were conducted. 4.4.1 T-test

A T-test is used to compare the mean score on some continuous variable for two different groups of subject (Pallant, 2005). In order to conduct independent-samples T-test, two variables are needed: an independent variable (e.g. gender), and a dependent variable (e.g. ambient conditions). This test shows whether there is a statistically significant difference in the mean scores for two groups (e.g. whether males and females differ significantly in terms of evaluating ambient conditions). The T-test showed that there is no significant difference between gender and ambient conditions (sig. (2-tailed) value is 0,155), which is above 0,05, and means that significant difference between males and females while evaluating the questions about ambient conditions does not exist. What is more, the T-test also pointed out that there is no difference between gender and space/function (sig. (2-tailed) value is 0,389), which is above 0,05 and indicates that significant difference between males and females while evaluating the questions about space/function does not exist. The T-test between gender and signs, symbols and artifacts illustrated that there is no significant difference between gender and signs, symbols and artifacts (sig. (2-tailed) value is

32

0,166), which is above 0,05 and stands for no significant difference between males and females while evaluating the questions about signs, symbols and artifacts. However, the T-test confirmed that there is a significance difference between gender and behaviour (sig. (2-tailed) value is 0,000), which is below 0,05 and means that significant difference between males and females while evaluating the questions about customers attitudes and overall impression of ICA MAXI (the behaviour factor). The T-test between gender and the statement How often do you visit ICA MAXI showed that there is no significant difference (Sig. (2-tailored) value is 0,053) between females and males and their shopping frequency (Table A. 11, in the Appendix 1).
Table 6. T-tests for gender and ambient conditions, space/function, signs, symbols and artifacts, behaviour, and the statement How often do you visit ICA MAXI? Factor/statement Ambient conditions Space/function Signs, symbols and artifacts Behaviour How often do you visit ICA MAXI? Source: Own composition based on SPSS outcomes. Sig. ,717 ,852 ,063 ,071 ,212 Sig. (2-tailed) ,155 ,389 ,166 ,000 ,053 Significant difference Does not exist Does not exist Does not exist Exists Does not exist

Table 6 presents the results of T-tests when it comes to gender and ambient conditions, space/function, signs, symbols and artifacts, behaviour, and the statement How often do you visit ICA MAXI?. Moreover, Tables A.6-A.11 in the Appendix 1 present more detail information.
Table 7. T-tests for nationality and ambient conditions, space/function, signs, symbols and artifacts, behaviour, and the statement How often do you visit ICA MAXI? Factor Ambient conditions Space/function Signs, symbols and artifacts Behaviour How often do you visit ICA MAXI? Source: Own composition based on SPSS outcomes. Sig. ,209 ,004 ,246 ,152 ,001 Sig. (2-tailed) ,152 ,424 ,230 ,669 ,001 Significant difference Does not exist Does not exist Does not exist Does not exist Exists

The T-test for nationality and each factor (ambient conditions, space/function, signs, symbols and artifacts, and behaviour) has showed that there is no significant differences (Table 7) between Swedish and International respondents while evaluating the questions which belongs to above-mentioned factors. For detail information see Appendix 1 Table A. 12. When it comes to nationality (Table A. 13, in the Appendix 1), a significant difference exists between Swedish and International respondents and their shopping frequency (evaluated statement: How often do you visit ICA MAXI?). 4.4.2 One-way ANOVA

One-way ANOVA is used to measure the significance between one independent variable with three or more levels (e.g. age) and one dependent variable (e.g. ambient conditions).

33

One-way ANOVA tells if there is a significant difference in the mean scores in the dependent variable across different groups. Post-hoc test is usually used to find out where those differences are (Pallant, 2005). One-way ANOVA showed that there is no significant difference between different age groups and ambient conditions, since all significant values are above 0,05 (Table 8 and see Appendix 1, Table A. 14 for detail information).
Table 8. One-way ANOVA age and ambient conditions ANOVA Ambient conditions Between Groups Within Groups Total Source: SPSS outcome. Sum of Squares 344,005 12844,877 13188,882 df 3 123 126 Mean Square 114,668 104,430 F 1,098 Sig. ,353

However, a significant difference between different age groups and space/function exists (Table 9).
Table 9. One-way ANOVA age and space/function ANOVA Space/function Between Groups Within Groups Total Source: SPSS outcome. Sum of Squares 334,124 4558,073 4892,197 df 3 138 141 Mean Square 111,375 33,030 F 3,372 Sig. ,020

In the Appendix 1, in Table A. 15 can be seen that 30-39 age group and above 49 years old age group differ significantly in terms of their evaluation of space/function questions. Only three questions within this factor, such as It is easy to move in ICA MAXI with the shopping cart between aisles, I can find everything I need in ICA MAXI, and The interior design of ICA MAXI is visually appealing showed that different age groups evaluated differently those questions included in space/function (see Appendix 1, Table A. 16 and Table A. 17). Moreover, a significant difference also exists between different age groups and signs, symbols and artifacts (Table 10).
Table 10. One-way ANOVA age and signs, symbols and artifacts ANOVA Signs, symbols and artifacts Sum of Squares df Mean Square F Sig. Between Groups 482,798 3 160,933 5,517 ,001 Within Groups 3996,280 137 29,170 Total 4479,078 140 Source: SPSS outcome.

34

Table A. 18 in the Appendix 1 illustrates that a significant difference can be found between 18-29 age group and above 49 years old age group as well as 30-39 age group and above 49 years old age group. For example, the evaluation of the statement There are enough directional signs (e.g. entrance, exit, toilets) in ICA MAXI differs significantly within above-mentioned age groups. For detail information see Table A. 19 and Table A. 20 in the Appendix 1. Finally, one-way ANOVA showed that there is a significant difference between different age groups and behaviour (Table 11).
Table 11. One-way ANOVA age and behaviour ANOVA Behaviour Sum of Squares df Mean Square F Sig. Between Groups 565,319 3 188,440 7,967 ,000 Within Groups 3240,227 137 23,651 Total 3805,546 140 Source: SPSS outcome.

It can be seen (Appendix 1, Table A. 21) that only for the age group of 18-29 and above 49 years old as well as age group of 30-39 and above 49 years old there is no significant difference in terms of behaviour in ICA MAXI. For more details see Appendix 1 Table A. 22 and Table A. 23.
Table 12. One-way ANOVA disposable income and ambient conditions, space/function, signs, symbols and artifacts, and behaviour ANOVA Ambient conditions Between Groups Within Groups Total Between Groups Within Groups Total and Between Groups Within Groups Total Between Groups Within Groups Total Sum of Squares 673,718 11801,153 12474,871 147,668 4443,238 4590,906 350,426 4033,545 4383,971 261,183 3435,230 3696,413 df 4 119 123 4 134 138 4 133 137 4 133 137 Mean Square 168,430 99,169 36,917 33,158 87,607 30,327 65,296 25,829 F Sig. 1,698 ,155

Space/function

1,113

,353

Signs, artifacts Behaviour

symbols

2,889

,025

2,528

,044

Source: SPSS outcome.

The results of one-way ANOVA between monthly disposable income and all factors (ambient conditions, space/function, signs, symbols and artifacts, and behaviour) showed that there is a significant difference between monthly disposable income and signs, symbols and artifacts as well as behaviour (Table 12). For example, people who indicated that their monthly disposable income is 10.000-20.000 Swedish crowns had different responses in terms of the behaviour from people who chose option of 30.001-40.000 Swedish crowns. For more details, see Appendix 1 Table A. 24.

35

One-way ANOVA analysis was conducted as well in order to see which statements from signs, symbols and artifacts differ significantly. Table A. 25, in the Appendix 1, shows that the statement Directional signs are visually appealing has a value of 0,05 which indicates that there is a significant difference between this statement and monthly disposable income. However, the multiple comparisons table (Appendix 1, Table A. 26) does not show any asterisks in order to see where exactly the difference lies. Therefore, the authors decided to recode monthly disposable income from five into two groups. The first three income groups (< 10.000 kr; 10.000 20.000 kr; 20.001 30.000 kr) were connected and created the first group and the last two income groups (30.001 40.000 kr; > 40.000 kr) were connected and created the second group. Therefore, the T- test was conducted and showed that there is no significant difference between signs, symbols and artifacts and monthly disposable income (sig. (2-tailed) value is 0,861), see Table 13.
Table 13. T-test monthly disposable income and signs, symbols and artifacts Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,861 -,20753 95% Confidence Interval of the Difference Lower Upper 2,13817

F Equal variances assumed Equal variances not assumed Source: SPSS outcome. Signs, symbols and artifacts 1,265

Sig. ,263

t ,175

df 136

1,18616 -2,55323

48,671 ,188

,852

-,20753

1,10658 -2,43166

2,01660

One-way ANOVA showed that there is a significant difference between sentences I enjoy shopping at ICA MAXI and I like the environment of ICA MAXI (significant level below 0,05) and monthly disposable income (Table A. 27 in the Appendix 1). However, again the multiple comparisons table did not indicate any asterisks where those differences lie (Table A. 28 in the Appendix 1).
Table 14. T-test monthly disposable income and behaviour Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,053 ,016 2,10028 2,10028 95% Confidence Interval of the Difference Lower Upper 4,22480 3,79161

F Equal variances 3,573 assumed Equal variances not assumed Source: SPSS outcome. Behaviour

Sig.

df 136

,061 1,955

1,07431 -,02423 ,84732 ,40896

2,479 66,843

36

Due to this, again the re-coded monthly disposable income was used and the T- test carried out. The T-test showed that there is no significant difference between behaviour and monthly disposable income (sig. (2-tailed) value is 0,053), see Table 14. One-way ANOVA was also conducted to examine a significant difference between income and the statement How often do you visit ICA MAXI?. The results showed that the significant difference between various income groups and shopping frequency exists (Table 15).
Table 15. One-way ANOVA monthly disposable income and the statement How often do you visit ICA MAXI? ANOVA How often do you visit ICA MAXI? Sum of Squares df Mean Square F Sig. Between Groups 34,078 4 8,520 4,823 ,001 Within Groups 238,493 135 1,767 Total 272,571 139 Source: SPSS outcome.

For example, people whose monthly disposable income is less than 10.000 Swedish crowns and those who chose the option 20.001-30.000 Swedish crowns as well as people whose monthly disposable income is less than 10.000 Swedish crowns and those who have between 30.001-40.000 Swedish crowns, their shopping frequency differ significantly (see in Appendix 1, Table A. 29). What is more, there is a significant difference between various age groups and shopping frequency (Table 16). However, the multiple comparisons (Table A. 30 in the Appendix 1) does not show any asterisks, hence, it does not indicate where exactly the difference lies.
Table 16. One-way ANOVA age and the statement How often do you visit ICA MAXI? ANOVA How often do you visit ICA MAXI? Sum of Squares df Mean Square F Sig. Between Groups 16,066 3 5,355 2,890 ,038 Within Groups 257,599 139 1,853 Total 273,664 142 Source: SPSS outcome.

Due to that, the age was re-coded from four into two groups. The first two age groups (1829 years old; 30-39 years old) were connected and created the first group and two other age groups (40-49 years old; above 49 years old) were connected and created the second group. Therefore, the T- test was conducted and showed that there is no significant difference between the statement How often do you visit ICA MAXI and age groups (sig. (2tailed) value is 0,057), see Table 17.

37

Table 17. T-test age and the statement How often do you visit ICA MAXI? Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. Difference (2Mean Std. Error tailed) Difference Difference Lower Upper ,057 -,467 ,244 -,949 ,014

F Equal variances assumed Equal variances not assumed Source: SPSS outcome. How often do you visit ICA MAXI? ,524

Sig. ,470

t 1,919

df 141

104,925 1,995

,049

-,467

,234

-,932

-,003

Finally, since the respondents evaluated the sentence I would prefer if ICA MAXI would play music with the standard deviation of 2,161, it seemed interesting to investigate whether there is a significant difference in scoring this statement among different age groups. Table 18 shows that there is a significant difference. In more detail, the answers of 18-29 age group respondents and respondents above 49 years old age group differ significantly in terms of issue of playing music in ICA MAXI (see in Appendix 1, Table A. 31).
Table 18. One-way ANOVA age and the statement I would prefer if ICA MAXI would play music ANOVA I would prefer if ICA MAXI would play music. Sum of Squares df Mean Square F Sig. Between Groups 52,575 3 17,525 3,992 ,009 Within Groups 605,798 138 4,390 Total 658,373 141 Source: SPSS outcome.

4.5

Factor analysis

Factor analysis is used to reduce or summarize the number of factors or components. Ideally the overall sample size should be more than 150 responses and a ratio of at least five cases for each of the variables (Pallant, 2005). In this case, factor analysis was used to reduce a big number of related variables to a more manageable number. Therefore, the outcome of factor analysis could be used as well for another conducted analysis, namely multiple regression. Kaiser-Meyer-Olkin values are always scored between 0 and 1. The KMO measures the sampling adequacy tests whether the partial correlations among variables are small. For a satisfactory factor analysis it should be scored at least 0,6 (Pallant, 2005). In this case, the KMO equals 0,826, which stands for a satisfactory result (Table 19). The Bartletts test of sphericity measures whether the correlation matrix is an identity matrix, which would indicate that the factor model is inappropriate. The Bartletts test of sphericity should be statistically significant at p<0,05 (in the research p = 0,000). Therefore, factor analysis is appropriate.

38

Table 19. KMO and Bartletts Test KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. ,826 Bartlett's Test of Sphericity Approx. Chi-Square 2157,800 df 351 Sig. ,000 Source: SPSS outcome.

An eigenvalue rule or the Kaisers criterion is one of the most commonly used techniques. Table A. 32 in the Appendix 1 presents all the eigenvalues of the components (1-27). Due to the eigenvalue rule, only the eigenvalues of 1,0 or more should be retained (Table 20). Therefore, not all 27 components can be retained for further investigation. The eigenvalue of a factor shows the amount of the total variance explained by this factor (Pallant, 2005). For example, factor 1 has the highest score (8,861), therefore, this factor accounts for the most variance and explains 32,82 per cent of the total variance. Thus, seven components with the eigenvalue above 1,0 explain 70,09 per cent of the variance. This means that these seven factors should be retained for rotation.
Table 20. Eigenvalues Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % 1 8,861 32,820 32,820 8,861 32,820 32,820 2 2,748 10,176 42,996 2,748 10,176 42,996 3 1,873 6,939 49,935 1,873 6,939 49,935 4 1,651 6,115 56,050 1,651 6,115 56,050 5 1,493 5,529 61,579 1,493 5,529 61,579 6 1,254 4,644 66,223 1,254 4,644 66,223 7 1,044 3,867 70,089 1,044 3,867 70,089 Extraction Method: Principal Component Analysis. Source: SPSS outcome.

However, in order to make better judgement and choose the proper number of the factors, also the Scree Plot should be used. Based on Graph 3 (see Appendix 1), it can be seen that from the seventh factor on, the line is getting flat. This indicates that each successive factor is accounting for smaller amounts of the total variance. It is worth to mention that the authors conducted factor analysis for 5, 6 and 7 factors. Since the outcomes of factor analyses for 5 and 6 factors did not give clear solutions, the authors finally decided to base their analysis on 7 factors created by SPSS. As the number of the factors has been determined, they have to be rotated in order to start interpretation. The most commonly used method is the Varimax method. It allows minimizing the number of variables that have high loadings on each factor (Pallant, 2005). Table 21 presents the loadings of the different variables as well as in which factor it belongs.

39

Table 21. Rotated Component Matrix

Rotated Component Matrixa 1 ,845 ,821 ,791 ,740 ,694 ,652 ,598 2 Component 3 4 5 6 7

The lighting in ICA MAXI is-Inviting The lighting in ICA MAXI is-Comfortable Colours used in ICA MAXI are-Pleasant The lighting in ICA MAXI is-Appropriate Colours used in ICA MAXI are-Soothing Colours used in ICA MAXI are-Appropriate Colours used in ICA MAXI are-Attractive There are enough directional signs (e.g. entrance, exit, toilets) in ICA MAXI. The directional signs (e.g. entrance, exit, toilets) give guidelines how to move in ICA MAXI. It is easy to follow directional signs (e.g. entrance, exit, toilets) displayed in ICA MAXI. Directional signs are visually appealing. Other types of communication such as brochures are visually appealing. I am satisfied with the general impression of ICA MAXI. I enjoy shopping in ICA MAXI. I like the environment of ICA MAXI. I can find everything I need in ICA MAXI. This is a place that I would avoid shopping in the future. It is easy to move around ICA MAXI. It is easy to move in ICA MAXI with the shopping cart between aisles. The supermarket layout is logical to me. The noise level in ICA MAXI is acceptable for me. The interior design of ICA MAXI is visually appealing. I would prefer if ICA MAXI would play music. I think that ICA MAXI has too many products for the size of the store. I would like to spend more time browsing in ICA MAXI. Colours used in ICA MAXI are-Too bright The lighting in ICA MAXI is-Too dark Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. a. Rotation converged in 8 iterations. Source: SPSS outcome.

,854 ,815 ,798 ,744 ,585 ,768 ,716 ,673 ,646 ,619 ,830 ,779 ,626 ,553 ,519 ,438 ,831 ,617 ,776 ,731

Table 22 shows the comparison and the decision to retain the number of factors.
Table 22. Factors retained Component Label No. 1 lighting and colours 2 signs and symbols 3 customers behaviour in ICA MAXI 4 space/function and noise 5 design and music 6 products and browsing 7 bright colours and dark lightning Source: Own composition based on SPSS outcomes. Eigen value 8,861 2,748 1,873 1,651 1,493 1,254 1,044 Decision Retain Retain Retain Retain Retain Retain Retain

40

Factor 1 loaded variables that involved the features of the lightning as well as colours used in ICA MAXI. Factor 2 loaded variables that clustered around directional signs such as entrance, exit or toilets, displayed in the supermarket. Factor 3 included variables connected with customers opinion about shopping and the store itself. Factor 4 contained variables linked to space/function features and the noise level. Factor 5 loaded variables that clustered around the stores design and music. Factor 6 included variables that involved the product diversification and browsing in the supermarket. Finally, factor 7 consisted of variables connected with too bright colours and too dark lightning performed in the store.
Table 23. Cronbachs Alphas for all factors Factor Cronbach's Alpha 0,900 0,881 N of Items 7 5 5 4 2 2 2 Interpretation High reliability High reliability High reliability High reliability Low reliability Low reliability Low reliability % of the deviance in the score 90,0 88,1 83,2 77,4 18,7 42,5 45,2

Lightning and colours Signs and symbols Customers behaviour 0,832 in ICA MAXI Space/function and 0,774 noise Design and music 0,187 Products and browsing 0,425 Bright colour and dark 0,452 lightning Source: Own composition based on SPSS outcomes.

In order to measure the reliability of retained factors, the Cronbachs Alpha was used (Table 23). The scores of the Cronbachs Alpha for lightning and colours, signs and symbols, customers behaviour in ICA MAXI, space/function and noise factors formed high reliability which leads to acceptable scales (respectively 0,900; 0,881; 0,832; 0,774). It can be concluded that the higher the respondents scored the questions about lightning and colours, signs and symbols, customers behaviour in ICA MAXI, space/function and noise, the higher their intensions to stay or explore ICA MAXI. On the other hand, the scores of the Cronbachs Alpha for design and music, products and browsing and bright colours and dark lightning formed low reliability which leads to unacceptable scales (respectively 0,187; 0,425; 0,452). Moreover, when it comes to an error variance (random error), the factor 1 scored 0,19 error variance in the scores (0,9x0,9=0,81; 1,0-0,81=0,19). The rest factors scored respectively 0,22; 0,31; 0,40; 0,97; 0,82; 0,80. Therefore, it can be said that as the estimate of reliability increases, the fraction of a test score, attributable to error, will decrease. Reliability statistics for lightning and colours, signs and symbols, customers behaviour in ICA MAXI, space/function and noise show respectively 90,0 per cent, 88,1 per cent, 83,2 per cent, and 77,4 per cent of the deviance in these scores by combining respectively 7, 5, 5, and 4 items which will be considered as a true score variance or internally consistent reliable variance. Reliability statistics of design and music shows that only 18,7 per cent of the deviance in that score by combining 2 items which will be considered as a true score variance. Two variables which are included in this factor are scored close to 0,5. The statement The interior design of ICA MAXI is visually appealing scored 0,519 and the statement I

41

would prefer if ICA MAXI would play music scored below 0,5. Therefore, the
Cronbachs Alpha obtains a low score. Reliability statistics of products and browsing shows that only 42,5 per cent of the deviance in that score by combining 2 items which will be considered as a true score variance. Two variables which are included in this factor are scored above 0,6. The statement I think that ICA MAXI has too many products for the size of the store scored 0,831 and the statement I would like to spend more time browsing in ICA MAXI scored 0,617. Since the Cronbachs Alpha is sensitive to the number of items included in the scale, in this case, the low number of items could be the main reason why this indicator obtains a low score. Finally, reliability statistics of bright colours and dark lightning shows that only 45,2 per cent of the deviance in that score by combining 2 items which will be considered as a true score variance. Two variables which are included in this factor are scored above 0,7. The statement Colours used in ICA MAXI are too bright scored 0,776 and the statement The lighting in ICA MAXI is too dark scored 0,731. Since the Cronbachs Alpha is sensitive to the number of items included in the scale, in this case as well, the low number of items could be the main reason why this indicator obtains a low score. After conducting factor analysis, it can be concluded that there are seven important factors which are used in the analysis, namely: lighting and colours, signs and symbols, customers behaviour in ICA MAXI, space/function and noise, design and music, products and browsing, and bright colours and dark lightning. However, the Cronbachs Alphas for only the first four factors are scored above 0,6 which stands for a reliable scale. This indicates a strong internal consistency among the items included in a particular factor. Essentially, this can be interpret that respondents who selected a low score for one item tended to select low scores for the other items included in the factor, and conversely, the respondents who tented to select a high score for one item tended to select high scores for the others within the same factor. Hence, knowing the score for one item, it is possible to predict with some accuracy the possible scores for the other items within the same factor. The rest factors have a low reliability as the Cronbachs Alphas are scored respectively 0,187; 0,425; and 0,452 thus, they do not create a reliable scale. This means, that the ability to predict scores from one item is not possible. Therefore, only the first four factors will be retained for further analysis.

4.6

Multiple regression

Multiple regression analysis was conducted to see the effects between the servicescape variables and customers behaviour in ICA MAXI. It is usually used to investigate how much of the variance in dependent variable (e.g. customers behaviour in ICA MAXI) can be explained by independent variables (lighting and colours, signs and symbols, space/function and noise). Since only four factors were retained after conducting factor analysis, three of them were used as independent variables and one as the dependent (Table 24). This method also gives an indication of the relative contribution within independent variables (Pallant, 2005).

42

Table 24. Multiple regression variables entered Variables Entered/Removedb Model Variables Entered 1 space/function and noise, lighting and colours, signs and . symbols a. All requested variables entered. b. Dependent Variable: customers behaviour in ICA MAXI Source: SPSS outcome. Variables Removed Method Enter

4.6.1

Correlations and multicollinearity

The correlations between variables are provided in Table 25.

Table 25. Multiple regression - correlations Correlations customers behaviour in ICA MAXI customers behaviour in ICA MAXI lighting and colours signs and symbols space/function and noise Sig. (1-tailed) customers behaviour in . ICA MAXI lighting and colours signs and symbols space/function and noise N customers behaviour in ICA MAXI lighting and colours signs and symbols space/function and noise Source: SPSS outcome. Pearson Correlation 1,000 ,501 ,563 ,509 ,000 . ,000 ,000 141 129 139 141 lighting and colours ,501 1,000 ,429 ,380 ,000 ,000 . ,000 129 131 129 131 signs and symbols ,563 ,429 1,000 ,474 ,000 ,000 ,000 . 139 129 141 141 141 131 141 143 space/function and noise ,509 ,380 ,474 1,000 ,000 ,000 ,000

It can be seen that all independent variables show a relationship with the dependent variable. According to Pallant (2005) correlations suppose to be higher than 0,3. In this case, lightning and colours, signs and symbols, and space/function and noise correlate with customers behaviour in ICA MAXI (0,501; 0,563; and 0,509 respectively) and create a positive correlation, since all scores are between 0 and 1. Moreover, according to Cohen (1988) guidelines, all three factors create large relationships between dependent and independent variables, as they score above 0,5 (cited in Pallant, 2005). Furthermore, it can be seen that the most correlated factor, hence, the strongest relationship exists between customers behaviour in ICA MAXI and signs and symbols. What is more, possible multicollinearity between variables was checked. In this case, coefficients were taken into account, precisely Tolerance and VIF (Table 26). The value Tolerance is the measurement of how much of the variability of the specified independent is not explained by other independent variables in the model (Pallant, 2005). If this value is less than 0,10, then it points out that the multiple correlation with other variables is high and leads to a possible multicollinearity. In this case, the Tolerance value is above 0,7, which means that there is no high correlation with other variables in this model. Another

43

value, VIF, should be taken into consideration if it exceeds 10. It can be seen that all VIF values are slightly above 1, which do not exceed 10.
Table 26. Coefficient multiple regression Factor Tolerance Lightning and colours ,776 Signs and symbols ,703 Space/function and noise ,737 Source: Own composition based on SPSS outcome. VIF 1,289 1,423 1,357

4.6.2

Evaluating the regression model of customers behaviour in ICA MAXI

The model summary for the regression model is provided in Table 27. The model is based on the standard multiple regression where all of the three independent variables were entered simultaneously. The adjusted R square is the first step that should be inspected. Table 27 reveals results of R2 (0,446) and adjusted R2 (0,433). R square indicates how much of the variance in the dependent variable (customers behaviour in ICA MAXI) is explained by independent variables (lighting and colours, signs and symbols, and space/function and noise). This indicates, that this model explains 44,6 per cent of the variance in customers behaviour in ICA MAXI. However, according to Tabachnick and Fidell (2001, p.147) adjusted R square is more reliable measure because it provides a better estimate of the true population value (cited in Pallant, 2005, p. 151). Therefore, adjusted R square value shows that multiple regression model explains 43,3 per cent of variance in customers behaviour in ICA MAXI.
Table 27. Multiple regression R and R square Model Summaryb Change Statistics R Adjusted R Std. Error of R Square F Model R Square Square the Estimate Change Change df1 df2 1 ,668a ,446 ,433 ,78521 ,446 33,577 3 125 a. Predictors: (Constant), space/function and noise, Lighting and colours, Signs and symbols b. Dependent Variable: Customers behaviour in ICA MAXI Source: SPSS outcome. Sig. F Change ,000

To evaluate how this model fits for the analysis, it is useful to look at the ANOVA model (Table 28). In this case, the model fits very well, since the statistical significance is scored below 0,05. More precisely, the model is scoring 0,000.
Table 28. Multiple regression - ANOVA ANOVAb Model Sum of Squares df Mean Square F 1 Regression 62,106 3 20,702 33,577 Residual 77,069 125 ,617 Total 139,174 128 a. Predictors: (Constant), space/function and noise, Lighting and colours, Signs and symbols b. Dependent Variable: Customers behaviour in ICA MAXI Source: SPSS outcome.

Sig. ,000a

44

4.6.3

Evaluating contribution of the dependent variable in customers behaviour in ICA MAXI

Since it is interesting to compare the contribution of each independent variable to predict customers behaviour in ICA MAXI, beta values should be also taken into account (Table 29). It can be noticed that the largest beta coefficient (0,330) is for signs and symbols. This means, that this variable makes the strongest contribution when explaining customers behaviour in ICA MAXI. Based on Bitners (1992) study, signs, symbols and artifacts influence approach behaviour. However, since it is very complex dimension and includes various items, it is not effortlessly to examine the affect on the customers behaviour. The other beta coefficients are slightly lower; lightning and colours and space/function and noise have less contribution with the dependent variable.
Table 29. Multiple regression coefficients Beta Lightning and colours ,264 Signs and symbols ,330 Space/function and noise ,252

Sig. ,006 ,001 ,000

Part ,232 ,277 ,216

Part2 0,054 0,077 0,047

Contribution (%) 5,4 7,7 4,7

a. Dependent Variable: Customers behaviour in ICA MAXI Source: SPSS outcome.

For each variable, a significance value should be examined. This explains whether variables make a statistically significant unique contribution to the equation (Pallant, 2005). Since all independent variables have a significant value of less than 0,05, they are making a significant contribution of the dependent variable. In this case, all three independent variables have an unique and statistically significant contribution for customers behaviour in ICA MAXI. Another important information to look at is Part correlation coefficients column. When this number is squared, a measure of the contribution of a specific variable to the total R squared is obtained. Lighting and colours has a part correlation of 0,232 (Table 29). After squaring this number, the result of 0,054 is gained, indicating that lightning and colours explain slightly more than 5 per cent of customers behaviour in ICA MAXI. When it comes to signs and symbols, the squared value is 0,077, indicating a contribution of almost 8 per cent. Finally, space/function and noise squared value is 0,047, indicating a contribution of almost 5 per cent to the explanation of variance in customers behaviour in ICA MAXI.

Analysis

The fifth chapter presents the connection between the findings of the conducted research elaborated in the fourth chapter and the literature framework presented in the second chapter. When it comes to the music studies, various researchers showed that music played in the supermarket has either positive or negative influence. Based on Yalch and Spangenbergs (1990) study, customers tend to spend more time in the supermarket where background music is played than in the supermarket without music. Additionally, Linsen (1975) shows that customers like background music, as this means that the supermarket cares about

45

them. Moreover, Morrison et al. (2011) claim that music affects customers arousal by increasing their pleasure as well as behaviour in the store (e.g. money and time spent in the supermarket). However, the customers of ICA MAXI have a different attitude towards music played in this supermarket. Generally, respondents attitude towards background music in ICA MAXI (Q1) was rather neutral. The empirical study also showed that younger generation is more willing to hear background music in the store, whereas elder generation has more negative opinion towards this issue. Studies conducted upon the noise level showed, on the one hand, too much noise can be related with the negative stimulation such as irritability or decreased concentration. On the other hand, the silence can be seen as a problematic issue (Kryter, 1985). The customers of ICA MAXI evaluated the noise level in the supermarket as neutral. Therefore, the noise level in ICA MAXI is perceived to be displayed in an acceptable level for the customers. Taylor and Sucov (1974) find out that customers tend to be more attracted by the bright lights which results in analyzing and browsing products in the supermarket more than in the dark lights. Moreover, Areni and Kim (1993) state that the bright lightning does not influence time spent in the store. According to Birren (1969), bright lights are used in order to increase the impulse purchases in the store. In relation to the empirical studies, when it comes to customers attitudes towards lightning, the respondents agreed that the lightning in ICA MAXI is appropriate (Q2), comfortable (Q4), and not too dark (Q5). Since the respondents evaluated displayed lightning as not too dark, it can be said that they perceived the lightning as bright. This means that bright light may influence customers of ICA MAXI to spend more time while analyzing and browsing the products in the supermarket and increase the number of impulse purchases. Hence, it can be concluded that the customers of ICA MAXI are satisfied with the lightning displayed in the supermarket. Moreover, according to the customers of ICA MAXI, colours used in the supermarket are appropriate (Q6) and not too bright (Q10). Since the main colour displayed inside the supermarket as well as the logo of ICA MAXI itself are red (warm colour), based on Stones experiments (2003), warm colours stimulate customers in association with higher blood pressure, eyeblink frequency or respiratory rate. Additionally, red colour is linked with arousal (Cahoon, 1969). Therefore, customers who are aroused are stimulated and excited, that results in approach behaviour exploring the supermarket or spending more time on browsing the products. What is more, warm colour environments tend to draw customers attention more than the cool ones (Bellizzi et al., 1983). Additionally, according to Bitner (1992), signs, symbols and artifacts influence approach behaviour exploring the supermarket as well spending more time. However, according to Davis (1984) and Wener (1985) it is hard to examine the affect on customers behaviour in the store, as there are various items included in this factor. Based on the empirical study, ICA MAXI customers evaluated that it is easy to follow directional signs which are displayed in the store (Q20). According to Ballantine et al. (2010) large spaces encourage customers to explore the store as well as make them more comfortable. Moreover, Thang and Tan (2003) claim that ease of movement in the supermarket influences the store preference. In relation to this study, customers of ICA MAXI agreed that moving in the store (Q12) and with the shopping carts (Q13) is easy. It can be assumed that in terms of layout, customers of ICA MAXI feel comfortable while exploring the store.

46

According to the Mehrabian and Russell (1974) Stimulus-Organism-Response model, satisfaction is linked to pleasure which can be positively related to the approach-avoidance behavioural measures, in this case to the approach behaviour. Additionally, Sweeney and Wyber (2002) state that pleasure can result in higher shopping expenditures than the customer anticipated. Furthermore, it can be said that the customers feel happy, satisfied or relaxed which result in arousal, stimulation or excitement. Hence, based on the M-R model, pleasure and arousal are explaining the approach behaviour exploring the supermarket or spending more time. What is more, Baker et al. (1992) state that pleasure and arousal may influence on willingness to buy. The conducted study showed that the respondents will not avoid shopping in the store (Q27), as they enjoyed shopping in ICA MAXI (Q23) and their general impression of the supermarket (Q24) as well as its environment (Q25) were positive. Therefore, the respondents were satisfied while shopping in ICA MAXI. In addition, the study presented that the respondents would not like to spend more time browsing in the supermarket (Q26). What is more, the respondents disagreed with the statement that ICA MAXI has too many products (Q16). Moreover, the findings of an empirical study showed that when it comes to gender, females and males evaluated the perceived servicescape in terms of lightning, colours, music and noise in ICA MAXI in a similar way. Moreover, there was no difference found between males and females while evaluating the layout of ICA MAXI as well as its interior design. In relation to directional signs and other types of communication displayed in ICA MAXI, females and males responses did not vary as well. However, different opinions were found between females and males respondents in terms of satisfaction and enjoyment of an overall perceived servicescape in ICA MAXI. Furthermore, this study found that women and mens shopping frequency was alike. Based on the empirical findings, it can be said that Swedish and International respondents shop with different regularity. In addition, the research showed that nationality did not play an important role while evaluating the perceived servicescape in ICA MAXI. As far as age is concerned, the study indicated that different age groups scored the questions about lightning, colour, music and noise similarly as well as the statement How often do you visit ICA MAXI?. Nevertheless, the age groups had dissimilar attitudes on the layout and the interior design of ICA MAXI as well as towards directional signs and other types of communication, such as brochures, displayed in ICA MAXI. Moreover, an overall opinion about the servicescape in ICA MAXI varied among various age groups. Finally, the difference in opinions about background music in ICA MAXI and various age groups was interesting to investigate. The research showed that young respondents had different attitude towards music than elder customers of ICA MAXI. When taken into consideration monthly disposable income, the research showed that the difference between income and attitudes towards directional signs displayed in ICA MAXI as well as the general impression and satisfaction of shopping in ICA MAXI exists. Additionally, the respondents from different monthly disposable income groups visit ICA MAXI with different regularity. In order to summarize the number of variables, factor analysis was conducted. Since factor analysis does not distinguish between independent variables and a dependent variable, thus,

47

four main components such as lightning and colours, signs and symbols, customers behaviour in ICA MAXI, and space/function and noise were obtained. The higher the respondents scored questions in each of those factors, the higher their intensions to stay or explore ICA MAXI. The factor lightning and colours has the highest reliability which indicates a high degree to which each item correlates with a total score. The lightning and colours is followed by the signs and symbols component. In relation to the Bitners (1992) framework of servicescapes, lighting, noise, colours, signs and symbols as well as space/function conditions such as layout and equipment, are the factors that create an emotionally pleasant environment in the supermarket. This means that those factors influence customers mood, attitude or certain beliefs about the supermarket. They also can make customers feel happy, satisfied or relaxed in the supermarket which leads to the arousal stimulation or excitement. Moreover, environmental factors mentioned above explain the approach behaviour exploring the supermarket, spending more time on browsing the products which refers to the increase of number items bought as well as increase in sales for the marketers. Multiple regression was conducted based on four components created by factor analysis. Since in multiple regression it is necessary to have independent variables as well as a dependent variable, three factors created by factor analysis, namely lightning and colours, signs and symbols, and space/function and noise stand for the independent variables, whereas customers behaviour in the ICA for the dependent variable. Multiple regression showed that all independent variables (lightning and colours, signs and symbols, and space/function and noise) play an important role while understanding the customers behaviour in ICA MAXI variable. The largest beta coefficient stands for signs and symbols, which means that the questions included in this factor are the most crucial while explaining customers behaviour in ICA MAXI. Since beta coefficients is just slightly lower for other two factors, which are lightning and colours, and space/function and noise, it can be said that they have a strong contribution as well. Furthermore, it was found that the signs and symbols variable explains almost 8 per cent of customers behaviour in ICA MAXI. Two other components have a contribution of almost 5 per cent, slightly lower, however, those factors should be also taken into account. This means that lighting, noise, colours, signs and symbols as well as space/function conditions such as layout and equipment are the factors that have a highest contribution which leads to the approach behaviour in the supermarket attraction, exploring the supermarket, spending more and time and returning next time.

48

Conclusion

The sixth chapter concludes the thesis and presents the major findings as well as the authors contribution based on the research results. Thus, managerial implications, limitations and future research are discussed. The purpose of the conducted study was to find out what is the role of the perceived servicescape in ICA MAXI in Jnkping and how the perceived servicescape influences customers behaviour. The study represents the attempt to connect the theory based framework with the role of the perceived servicescape in the supermarket. The theoretical framework builds a wide range of literature about environmental psychology as well as customers behaviour to measure key servicescape variables. In addition to the theoretical contribution, the empirical research, and then testing the framework were made. In this regard, the study highlights that a number of factors (lightning and colours, signs and symbols, and space/function and noise) are linked to customers behaviour. In relation to the theoretical part, it can be said that factors such as lighting, noise, colours, signs and symbols as well as space/function conditions such as layout and equipment create emotionally pleasant environment in the supermarket. Therefore, those environmental factors influence customers mood, attitude or certain beliefs about the supermarket. This leads to a customers satisfaction or even happiness which refer to stimulation or excitement. When one is excited and aroused by the physical environment of the supermarket, one will explore the supermarket, spend more time on browsing the products, which refer to an increase of number items bought as well as an increase in sales for the marketers. Many studies within this area provide only theoretical bases of the servicescapes. However, this study also supplies an empirical evidence. Moreover, the authors confirmed perceived servicescape factors to be key determinants while understanding customers behaviour in the supermarket. Analysis via multiple regression revealed the effect between three factors (lightning and colours, signs and symbols, and space/function and noise) and customers behaviour in the supermarket. The findings of this paper support the Bitners (1992) framework of servicescapes as well as the Mehrabian and Russell (1974) Stimulus-Organism-Response model. The results indicate that the servicescape elements such as lightning and colours, signs and symbols, and space/function and noise are important in influencing customers pleasure and arousal in the supermarket. Simply saying, these factors induce favourable emotions such as satisfaction, happiness, enjoyment, relaxation, stimulation or excitement in the supermarket. Additionally, especially the signs and symbols factor was found to induce the highest level of pleasure and approach behaviour. Based on the Mahrabian and Russell (1974) model, favourable emotions lead to behaviour responses. In relation to the supermarket, when customers have favourable emotions, they tend to stay longer and explore the store, spend more money and probably return to the supermarket next time. Therefore, retailers should improve the signage as well as layout and ambient conditions to affect customers in their evaluation of the service, since the suitable servicescape induces more pleasure than the poor servicescape.

49

Many studies conducted have tended to focus on single elements of the servicescape such as music, lightning or layout. However, not that many studies have tried to analyze the influence of more than one or two elements of the servicescape. This research differs from previous by trying to use six servicescape variables and finding empirical support for them.

6.1

Managerial implications

The overall conclusion of this paper is that through careful and creative management of the servicescape, marketers and retailers might be able to achieve both internal and external marketing goals. This section will describe some specific implications. From a practical point of view, this research helps to discover environmental factors that creates physical environment in the supermarket and are used in determining customers impression of the store. Since those factors lead to the approach behaviour such as willingness to explore the supermarket, spend more time on browsing, which refer to intention to revisit the supermarket, an increased number of items bought as well as increase in sales for the marketers; understanding environmental factors allows both retailers and marketers to make improvements in these areas. Even this research did not make the attempt to determine what kind of colours customers prefer or what genre of music they would like to hear in the supermarket, the findings show extra caution and should be carefully considered when designing or redesigning the store. A clear implication of the theoretical framework is needed for multi functional cooperation when making decisions related to the physical environment. For instance, designing the supermarket might have an impact on marketing goals of the company such as customers attraction or satisfaction as well as operational goals such as cost reduction and efficiency. Nowadays, in more and more companies, environment management is performed by facility managers or facility planners. However, is some companies, environmental decisions are made without any attention to influence customers behaviour in the supermarket. However, it is necessary to meet customers needs and expectations by creating pleasant environment in the supermarket. Since the research was conducted among ICA MAXI customers in Jnkping, the management of ICA MAXI can also take an advantage of the findings. The study showed that the main factor that influences customers behaviour in the supermarket is signs and symbols. Therefore, the management should have in mind that the improvement of displayed signs in the store leads not only to a better communication about the place to its users, but also influences customers behaviour and, indirectly, an increase in sales. However, such factors as lightning and colours and space/function and noise cannot be omitted, as the stores environment consists of various factors that, especially combined, affect customers behaviour. Additionally, background music in ICA MAXI seems to be neuralgic issue as well. The managers should investigate this matter (whether to play music in the supermarket) more deeply, since younger customers have different opinion about playing music in ICA MAXI than the elder generation. Furthermore, the management of the supermarket has to be aware that customers needs and wants are changing over time, hence, studies examining their desires should be conducting continuously.

50

6.2

Limitations

A few limitations should be mentioned in relation to this research. Firstly, the data was collected only by conducting quantitative research method distributing questionnaires. Even this research method was seen as suitable one, to get a deeper insight on how the perceived servicescape influences customers behaviour in the supermarket, the qualitative research method may be conducted as well. Therefore, it would be useful to combine both research methods. The personal interviews or focus groups would give the authors deeper insights. However, since none of the authors of this study have conducted a qualitative research method before, it would be probably difficult to analyze and interpret the data as well as conduct interviews in terms of accuracy. Moreover, it is problematic to draw general conclusions of this research since the influence of the servicescape may be larger in the places where customers spend more time (e.g. hospitals, hotels). The more an individual spends time in the facility, the more influenced by the servicescape one is. What is more, this study found out that servicescapes elements such as lightning and colours, signs and symbols, and space/function and noise are important in influencing customers behaviour in the supermarket. However, the authors of this paper did not investigate in precise such attributes as what colours and genre of music can influence customers moods, beliefs and attitudes in a more positive way. Since the authors cannot generalise findings of this paper for all customers of ICA MAXI in Jnkping, Sweden, the sampling method can be also seen as a limitation.

6.3

Future research

When it comes to the future research, since this study identified important servicescape elements only for one precise supermarket in Jnkping, Sweden, future studies may focus on a broader number of supermarkets in the whole country. Furthermore, future studies may attempt to investigate all six (or more) servicescape variables in more detail while conducting a qualitative research method. Personal interviews might give more information on what colours customers prefer to see in the supermarket or what genre of music they would like to hear. The further research may also explore how do different environmental factors in the supermarket influence customers moods, attitudes or beliefs. There is a need for further research on single servicescape elements. Various studies have focused on the effect of music, colours or lightning. However, only few of them paid attention to the design or the signs, symbols and artifacts in the supermarket. What is more, since the majority of the environmental factors have emotional effects on customers, future researchers may try to investigate cognitive and physiological aspects of the servicescape. Since this study focused only on the customers perspective, the Bitners (1992) framework of servicescapes is also often use for explaining employees behaviour in the supermarket. Future researchers may try to investigate how social interactions with the employees influence customers behaviour in the supermarket.

51

The authors of this paper used the Mehrabian and Russell (1974) Stimulus-OrganismResponse model. However, the main focus was only on two emotional states - pleasure and arousal. The third emotional state which is dominance, was not investigated in this research because of the lack of the empirical support. However, future researchers could analyze the dominance dimension to find out whether it is appropriate to use it in servicescape studies.

52

References

Ang, H. S., & Leong, M. S. (1997). The mediating influence of pleasure and arousal on layout and signage effects. Journal of Retailing and Consumer Services, Vol. 4, No. 1, p. 13-24. Areni, Ch. S., & Kim, D. (1994). The influence of in-store lightning on consumers examination of merchandise in a wine store. International Journal of Research in Marketing, Vol. 11, p. 117-125. Babbie, E. (2007). The Practice of Social Research (11th ed.). USA: Thomson Wadsworth. Babin, B. J., Hardesty, D. M., & Suter, T. A. (2003). Colour and shopping intentions: The intervening effect of price fairness and perceived affect. Journal of Business Research, Vol. 56, p. 541-551. Baker, J., Levy, M., & Grewal, D. (1992). An Experimental Approach to Making Retail Store Environmental Decisions. Journal of Retailing, Vol. 68, No. 4, p. 445-460. Ballantine, P. W., Jack, R., & Parsons, A. G. (2010). Atmospherics cues and their effect on the hedonic retail experience. International Journal of Retail & Distribution Management, Vol. 38, No. 8, p. 641-653. Bellizzi, J. A., & Hite, R. E. (1992). Environmental Color, Consumer Feelings, and Purchase Likelihood. Psychology & Marketing, Vol. 9, No. 5, p. 347-363. Beri, G. C. (2008). Marketing Research (4th ed.). New Delhi: Tata McGraw-Hill. Bitner, J. M. (1992). Servicescapes: The Impact of Physical Surroundings on Customers and Employees. The Journal of Marketing, Vol. 56, No. 2, p. 57-71. Carter, E., & Porter, S. (2000). Validity and reliability. In D. Cormack (Ed.), Process in Nursing, (4th ed.). UK: Blackwell. Chang, H-J., Eckman, M., & Yan, R-N. (2011). Application of the Stimulus-OrganismResponse model to the retail environment: the role of hedonic motivation in impulse buying behavior. The International Review of Retail, Distribution and Consumer Research, Vol. 21, No. 3, p. 233-249. Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied Multiple Regression/Correlation Analysis for the Behavioural Sciences (3rd ed.). USA: Lawrence Erlbaum Associates. Converse, P. D., & Spencer, Ch. (1942). What Housewives Think of the Super-Market. The Journal of Marketing, p. 371-374. Countryman, C. C., & Jang, S. (2006). The effects of atmospheric elements on customer impression: the case of hotel lobbies. International Journal of Contemporary Hospitality Management, Vol. 18, No. 7, p. 534-545. Dataminor. (2010). Food Retail in Sweden (reference code 0181-2058) [Brochure]. Datamonitor.

53

Davis, E., Freedman, M., Lane, J., McCall, B., Nestoriak, N., & Park, T. (2009). Product Market Competition and Human Resource Practises in the Retail Food Sector. Industrial Relations, Vol. 48, No. 2, p. 350-371. Dennis, Ch., & Newman, A. (2005). Modeling the Effects of Mall Atmospherics on Shoppers Approach Behaviors. Brunel Business School Working Paper Series: Special Issue on Marketing, Vol. 2, p. 1-32. Donovan, R. J., & Rossiter, J. R. (1982). Store Atmosphere. An Environmental Psychology Approach. Journal of Retailing, Vol. 58, No. 1, p. 34-57. Donovan, R., Rossiter, J. R., Marcoolyn, G., & Nesdale, A. (1994). Store Atmosphere and Purchasing Behavior. Journal of Retailing, Vol. 70, No. 3, p. 283-294. Dunne, P., & Lusch, R. (2008). Retailing (6th Edition). USA: Thomson South-Western. Food Marketing Institute, Retrieved 2012-02-18, from www.fmi.org Foxall, G. R. (1997). The emotional texture of consumer environments: A systematic approach to atmospherics. Journal of Economic Psychology, Vol. 18, p. 505-523. Ghauri, P., & Grnhaug, K. (1995). Research Methods in Business Studies A Practical Guide (3rd ed.). New York. Prentice Hall. Grewal, D., & Baker, J. (1994). Do retail store environmental factors affect consumers price acceptability? An empirical examination. International Journal of Research in Marketing, Vol. 11, p. 107-115. Gullstrand, J., & Jrgensen, Ch. (2011). Local Price Competition: The Case of Swedish Food Retailers [Working Paper]. AgriFood Economics Centre. Retrieved 2012-02-19, from http://www.agrifood.se/Files/AgriFood_WP20113.pdf Hansen, P., & Deutscher, T. (1977). An Emperical Investigation of Attribute Importance in Retail Store Selection. Journal of Retailing, Vol. 53, p. 59-73. Harris, C. L., & Ezeh, C. (2008). Servicescape and loyalty intentions: an empirical investigation. European Journal of Marketing, Vol. 42, No. 3/4, p. 390-422. Herrington, J., & Capella, L. (1996). Effects of music in service environments: a field study. The Journal of services marketing, Vol. 10, No. 2, p. 26-41. IMAP. (2010). Food and Beverage Industry Report 2010 [Report]. IMAP. Jain, R., & Bagdare, S. (2011). Music and consumption experience: a review. International Journal of Retailing & Distribution Management, Vol. 39, No. 4, p. 289-302. Kearney, T., Kennedy, A., & Coughlan, J. (2007). Servicescapes: A Review of Contemporary Empirical Research. Sixteenth Annual Frontiers in Service Conference, San Francisco, CA, p. 58. Kenhove, P., & Desrumaux, P. (1997). The relationship between emotional states and approach or avoidance responses in a retail environment. The International Review of Retail, Distribution and Consumer Research, Vol. 7, No. 4, p. 351-368.

54

Klatcheva, V. D., & Weitz, B. (2006). When Should a Retailer Create an Exciting Store Environment? The Journal of Marketing, Vol. 70, No. 1, p. 107-118. Kotler, P. (1973). Atmospherics As a Marketing Tool. Journal of Retailing, Vol. 49, p. 48-64. Lin, Y. I. (2004). Evaluating a servicescape: the effect of cognition and emotion. International Journal of Hospitality Management, Vol. 23, p. 163-178. Malhotra, N. K., & Birks, D. F. (2007). Marketing Research. An Applied Approach (3rd ed.). New York: Prentice Hall. Mattila, A. S., & Wirtz, J. (2001). Congurency of scent and music as a driver of in-store evaluations and behavior. Journal of Retailing, Vol. 77, p. 273-289. Morrison, M., Gan, S., Dubelaar, Ch., & Oppewal, H. (2011). In-store music and aroma influences on shopper behaviour and satisfaction. Journal of Business Research, Vol. 64, p. 558564. Nguyen, N., & Leblanc, G. (2002). Contact Personnel, Physical Environment and the Perceived Corporate Image of Intangible Services by New Clients. International Journal of Service Industry Management, Vol. 13, No. 3, p. 242-262. Oakes, S. (2000). The influence of the musicscape within service environments. Journal of Services Marketing, Vol. 14, No. 7, p. 539-556. Pallant, J. (2005). SPSS survival manual (2nd ed.). New York: McGraw-Hill Education. Quartier, K., Christiaans, H., & Van Cleempoel, K. (2008). Retail design: lightning as an atmospheric tool, creating experiences which influence consumers mood and behaviour in commercial spaces. Proceedings of the Design Research Society Conference 2008. Sheffield, UK. Quartier, K., Vanrie, J., & Cleempoel, K. (n.d.). The mediating role of consumers perception of atmosphere on emotions and behavior. A study to analyze the impact of lightning in food retailing. Retrieved 2012-03-27, from http://www.retailology.be/Retail-ology/publications/Artikelen/2010/10/7_A_study_to_analyze_the_impact_of_lighting_in_ food_retailing_files/final2_1.pdf. Reimer, A., & Kuehn, R. (2005). The Impact of Servicescape on Quality Perception. European Journal of Marketing, Vol. 39, No. 7/8, p. 785-808. Saunders, M., Lewis, P., & Thornhill, A. (2009). Research Methods for Business Students (5th ed.). UK: Pitman Publishing. Soriano, M. Y., Foxall, G. R., & Pearson, G. J. (2001). Emotion and environment: A test of the behavioral perspective model in a Latin American context. Journal of Consumer Behavior, Vol. 2, No. 2, p. 138-154. Spangenberg, E. R, Sprott, D. E., Grohmann, B., & Tracy, D. L. (2006). Gender-congruent ambient scent influences on approach and avoidance behaviors in a retail store. Journal of Business Research, Vol. 59, p. 1281-1287. Sukamolson, S. (n.d.). Fundamentals of Quantitative research. Retrieved 2012-03-02, from http://www.culi.chula.ac.th/e-Journal/bod/Suphat%20Sukamolson.pdf

55

Sweeney, J. C., & Wyber, F. (2002). The role of cognitions and emotions in the musicapproach-avoidance behavior relationship. Journal of Services Marketing, Vol. 16, No. 1, p. 5169. Tai, S. H. C., & Fung, A. M. C. (1997). Application of an environmental psychology model to in-store buying behavior. The International Review of Retail, Distribution and Consumer Research, Vol. 7, No. 4, p. 311-337. Taylor, G. R. (2005). Quantitative Research Methods. In G. R. Taylor (Ed.), Integrating Quantitative and Qualitative Methods in Research, (2nd ed.). USA: University Press of America. Tombs, A., & McColl-Kennedy, J. (2003). Social-servicescape conceptual model. Marketing Theory, Vol. 3, No. 4, p. 447-475. USDA Foreign Agricultural Service. (2010). Retail Food Sector Report for Sweden and Finland (nr SW1008) [Report]. Stockholm: Foreign Agricultural Service. Wakefield, L. K., & Blodgett, G. J. (1996). The effect of the servicescape on customers behavioral intentions in leisure service settings. The Journal of Services Marketing, Vol. 10, No. 6, p. 45-61. Ward, P., Davies, B. J., & Kooijman, D. (2007). Olfaction and the retail environment: examining the influence of ambient scent. Science Business, Vol. 1, p. 295-316. Wysocki, B. (1979). Sight, Smell, Sound: Theyre All Arms in Retailers Arsenal. The Wall Street Journal, Nov. 17. Vaus, D. (2005). Research Design in Social Research. London: SAGE. Vida, I., Obadia, C., & Kunz, M. (2007). The Effects of Background Music on Consumer Responses in a High-end Supermarket. International Review Of Retail, Distribution and Consumer Research, Vol. 17, No. 5, p. 469-482. Yildirim, K., Akalin-Baskaya, A., & Hidayetoglu, M. L. (2007). Effects of indoor color on mood and cognitive performance. Building and Environment, Vol. 42, p. 3233-3240. Zeithaml, V. A., Bitner, M. J., & Gremler, D. D. (2009). Services Marketing. Integrating Customer Focus Across the Firm (5th ed.). Boston: McGraw Hill.

56

Appendices

APPENDIX 1 - TABLES
Table A. 1. Item-Total Statistics ambient conditions Item-Total Statistics Scale Mean if Scale Variance if Item Deleted Item Deleted I would prefer if ICA 54,18 93,054 MAXI would play music. The lighting in ICA MAXI 52,47 88,648 is-Appropriate The lighting in ICA MAXI 53,07 83,463 is-Inviting The lighting in ICA MAXI 52,85 84,446 is-Comfortable The lighting in ICA MAXI 52,54 96,345 is-Too dark Colours used in ICA MAXI 52,71 88,891 are-Appropriate Colours used in ICA MAXI 53,48 81,521 are-Soothing Colours used in ICA MAXI 53,06 81,901 are-Pleasant Colours used in ICA MAXI 53,54 81,902 are-Attractive Colours used in ICA MAXI 52,80 96,990 are-Too bright The noise level in ICA 53,47 93,251 MAXI is acceptable for me. Source: SPSS outcome.

Corrected ItemTotal Correlation ,163 ,576 ,681 ,677 ,177 ,597 ,668 ,726 ,634 ,142 ,341

Cronbach's Alpha if Item Deleted ,829 ,778 ,764 ,766 ,815 ,777 ,763 ,759 ,767 ,820 ,798

Table A. 2. Item-Total Statistics space/function Item-Total Statistics Scale Mean if Scale Variance Corrected Item- Cronbach's Alpha Item Deleted if Item Deleted Total Correlation if Item Deleted It is easy to move around ICA 23,33 24,351 ,662 ,626 MAXI. It is easy to move in ICA MAXI with the shopping cart between 23,33 25,571 ,514 ,664 aisles. The supermarket layout is logical 23,74 23,329 ,595 ,635 to me. I can find everything I need in 23,77 24,062 ,537 ,654 ICA MAXI. I think that ICA MAXI has too many products for the size of 26,25 29,496 ,089 ,805 the store. The interior design of ICA 24,01 25,879 ,478 ,674 MAXI is visually appealing. Source: SPSS outcome.

57

Table A. 3. Item-Total Statistics signs, symbols and artifacts Item-Total Statistics Scale Variance Scale Mean if if Item Corrected ItemItem Deleted Deleted Total Correlation There are enough directional signs (e.g. entrance, exit, toilets) in ICA 20,10 19,390 ,796 MAXI. The directional signs (e.g. entrance, exit, toilets) give guidelines how to 20,14 20,780 ,767 move in ICA MAXI. It is easy to follow directional signs (e.g. entrance, exit, toilets) displayed 19,87 21,826 ,727 in ICA MAXI. Directional signs are visually 20,75 21,017 ,756 appealing. Other types of communication such 20,64 22,418 ,549 as brochures are visually appealing. Source: SPSS outcome. Table A. 4. Item-Total Statistics behaviour Item-Total Statistics Scale Mean if Scale Variance if Item Deleted Item Deleted I enjoy shopping in ICA 20,38 16,879 MAXI. I am satisfied with the general impression of ICA 20,33 18,264 MAXI. I like the environment of 20,55 17,521 ICA MAXI. I would like to spend more 22,79 18,997 time browsing in ICA MAXI. This is a place that I would 20,19 20,127 avoid shopping in the future. Source: SPSS outcome. Table A. 5. Item-Total Statistics all factors Item-Total Statistics Scale Mean if Item Deleted I would prefer if ICA MAXI would play music. The lighting in ICA MAXI isAppropriate The lighting in ICA MAXI is-Inviting The lighting in ICA MAXI isComfortable The lighting in ICA MAXI is-Too dark Colours used in ICA MAXI areAppropriate Colours used in ICA MAXI areSoothing Colours used in ICA MAXI arePleasant Colours used in ICA MAXI areAttractive 134,12 132,40 133,01 132,73 132,45 132,65 133,43 132,98 133,48 Scale Variance if Item Corrected ItemDeleted Total Correlation 430,241 415,664 406,884 410,397 429,522 412,709 403,470 402,049 401,806 ,095 ,484 ,571 ,561 ,168 ,565 ,567 ,646 ,566

Cronbach's Alpha if Item Deleted ,835 ,843 ,853 ,846 ,895

Corrected ItemTotal Correlation ,802 ,788 ,788 ,333 ,302

Cronbach's Alpha if Item Deleted ,659 ,679 ,670 ,831 ,830

Cronbach's Alpha if Item Deleted ,903 ,891 ,889 ,890 ,898 ,890 ,889 ,888 ,889

58

Colours used in ICA MAXI are-Too bright The noise level in ICA MAXI is acceptable for me. It is easy to move around ICA MAXI. It is easy to move in ICA MAXI with the shopping cart between aisles. The supermarket layout is logical to me. I can find everything I need in ICA MAXI. I think that ICA MAXI has too many products for the size of the store. The interior design of ICA MAXI is visually appealing. There are enough directional signs (e.g. entrance, exit, toilets) in ICA MAXI. The directional signs (e.g. entrance, exit, toilets) give guidelines how to move in ICA MAXI. It is easy to follow directional signs (e.g. entrance, exit, toilets) displayed in ICA MAXI. Directional signs are visually appealing. Other types of communication such as brochures are visually appealing. I enjoy shopping in ICA MAXI. I am satisfied with the general impression of ICA MAXI. I like the environment of ICA MAXI. I would like to spend more time browsing in ICA MAXI. This is a place that I would avoid shopping in the future. Source: SPSS outcome.

132,70 133,39 132,81 132,89 133,20 133,26 135,67 133,50 133,06 133,09 132,85 133,73 133,61 132,66 132,60 132,80 135,00 132,44

431,284 414,421 413,196 414,450 412,842 409,203 434,173 404,913 405,493 411,984 411,069 410,182 412,619 401,481 410,507 405,735 410,645 420,348

,130 ,460 ,535 ,464 ,455 ,511 ,073 ,626 ,606 ,544 ,589 ,586 ,496 ,772 ,707 ,741 ,415 ,316

,900 ,892 ,890 ,892 ,892 ,891 ,902 ,888 ,889 ,890 ,889 ,889 ,891 ,886 ,888 ,887 ,893 ,895

Table A. 6. T-test gender and ambient conditions Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,155 -2,59179 95% Confidence Interval of the Difference Lower Upper ,99232

F Ambient Equal conditions variances assumed Equal variances not assumed Source: SPSS outcome. ,132

Sig. ,717

t 1,431

df 125

1,81096 -6,17590

123,316 1,431

,155

-2,59179

1,81152 -6,17749

,99391

59

Table A. 7. T-test gender and space/function Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,389 -,85612 ,99039 95% Confidence Interval of the Difference Lower Upper 1,10193 2,81417 1,09965 2,81189

F Space/function Equal variances assumed Equal variances not assumed Source: SPSS outcome.

Sig.

t ,864

df 140

,035 ,852

139,545 ,865

,388

-,85612

,98921

Table A. 8. T-test gender and signs, symbols and artifacts Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,166 -1,32252 95% Confidence Interval of the Difference Lower Upper ,55602

F Equal variances assumed Equal variances not assumed Source: SPSS outcome. Signs, symbols and artifacts 3,505

Sig. ,063

t 1,392

df 139

,95011 -3,20106

127,938 1,381

,170

-1,32252

,95776 -3,21762

,57258

Table A. 9. T-test gender and behaviour Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,000 -4,45926 ,79721 95% Confidence Interval of the Difference Lower Upper 6,03548 2,88303 6,05523 2,86329

Sig.

df

Behaviour Equal variances 3,304 ,071 139 5,594 assumed Equal variances 125,545 not 5,530 assumed Source: SPSS outcome. ,

,000

-4,45926

,80644

60

Table A. 10. T-test gender and behaviour (detail information) Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. Difference (2Mean Std. Error tailed) Difference Difference Lower Upper ,000 -1,012 ,201 -1,410 -,613

F I enjoy Equal shopping in variances ICA MAXI. assumed Equal variances not assumed I am satisfied Equal with the variances general assumed impression of Equal ICA MAXI. variances not assumed I like the Equal environment variances of ICA assumed MAXI. Equal variances not assumed I would like to Equal spend more variances time browsing assumed in ICA Equal MAXI. variances not assumed This is a place Equal that I would variances avoid assumed shopping in Equal the future. variances not assumed Source: SPSS outcome.

Sig.

t 5,023

df 141

9,350 ,003

122,573 4,951 20,939 ,000 5,127 141

,000

-1,012

,204

-1,416

-,607

,000

-,895

,175 -1,240

-,550

112,370 5,024 2,250 ,136 3,560 141

,000

-,895

,178 -1,248

-,542

,001

-,706

,198 -1,098

-,314

129,227 3,524 ,267 ,606 2,145 139

,001

-,706

,200

-1,103

-,310

,034

-,627

,292 -1,205

-,049

136,508 2,141 16,004 ,000 4,291 140

,034

-,627

,293 -1,207

-,048

,000

-1,103

,257

-1,611

-,595

117,349 4,211

,000

-1,103

,262 -1,622

-,584

61

Table A. 11. T-test gender and the statement How often do you visit ICA MAXI? Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,053 ,449 ,230 95% Confidence Interval of the Difference Lower Upper -,006 ,904

F Equal variances assumed Equal variances not assumed Source: SPSS outcome. How often do you visit ICA MAXI? 1,571

Sig.

df 141

,212 1,952

1,943 136,062

,054

,449

,231

-,008

,907

Table A. 12. T-test nationality and all factors Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means 95% Confidence Interval of the Sig. Difference (2Mean Std. Error tailed) Difference Difference Lower Upper ,152 2,85108 1,97682 6,76375 1,06160

F Sig. t df Equal variances 1,594 ,209 1,442 124 assumed Equal variances 1,513 78,707 not assumed Space/function Equal variances ,004 ,950 ,810 137 assumed Equal variances ,804 71,096 not assumed Signs, symbols Equal and artifacts variances ,246 ,621 1,206 136 assumed Equal variances 1,216 70,927 not assumed Behaviour Equal variances ,152 ,697 ,428 136 assumed Equal variances ,427 72,155 not assumed Source: SPSS outcome. Ambient conditions

,134

2,85108

1,88437 -,89988 6,60204 3,08228 1,29087 3,11655 1,32514

,419

,89571

1,10576

,424

,89571

1,11382

,230

1,29604

1,07508 -,83000 3,42208

,228

1,29604

1,06570 -,82895 3,42102 2,37951 1,53257 2,39929 1,55236

,669

,42347

,98912

,670

,42347

,99119

62

Table A. 13. T-test nationality and the statement How often do you visit ICA MAXI? Independent Samples Test Levene's Test for Equality of Variances

t-test for Equality of Means Sig. (2Mean Std. Error tailed) Difference Difference ,000 1,030 ,244 95% Confidence Interval of the Difference Lower Upper ,548 1,512

F Equal variances assumed Equal variances not assumed Source: SPSS outcome How often do you visit ICA MAXI? 12,601

Sig.

df 138

,001 4,226

3,647 55,097

,001

1,030

,282

,464

1,596

Table A. 14. One-way ANOVA age and ambient conditions (detail information) Multiple Comparisons Ambient conditions Tukey HSD 95% Confidence Interval (I) Age (J) Age Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound 18-29 30-39 4,44928 2,48959 ,284 -2,0349 10,9334 40-49 ,66667 3,11299 ,997 -7,4411 8,7744 > 49 1,70238 2,32105 ,883 -4,3428 7,7476 30-39 18-29 -4,44928 2,48959 ,284 -10,9334 2,0349 40-49 -3,78261 3,54591 ,710 -13,0179 5,4527 > 49 -2,74689 2,87577 ,775 -10,2368 4,7431 40-49 18-29 -,66667 3,11299 ,997 -8,7744 7,4411 30-39 3,78261 3,54591 ,710 -5,4527 13,0179 > 49 1,03571 3,42968 ,990 -7,8969 9,9683 > 49 18-29 -1,70238 2,32105 ,883 -7,7476 4,3428 30-39 2,74689 2,87577 ,775 -4,7431 10,2368 40-49 -1,03571 3,42968 ,990 -9,9683 7,8969 Source: SPSS outcome. Table A. 15. One-way ANOVA age and space/function (detail information) Multiple Comparisons Space/function Tukey HSD (J) Age Mean Difference (I-J) Std. Error 30-39 1,39746 1,31007 40-49 1,18613 1,56073 > 49 -2,92393 1,24838 30-39 18-29 -1,39746 1,31007 40-49 -,21133 1,77939 > 49 -4,32139* 1,51287 40-49 18-29 -1,18613 1,56073 30-39 ,21133 1,77939 > 49 -4,11006 1,73447 > 49 18-29 2,92393 1,24838 30-39 4,32139* 1,51287 40-49 4,11006 1,73447 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome. (I) Age 18-29 Sig. ,710 ,872 ,094 ,710 ,999 ,025 ,872 ,999 ,088 ,094 ,025 ,088 95% Confidence Interval Lower Bound Upper Bound -2,0095 4,8045 -2,8727 5,2450 -6,1705 ,3226 -4,8045 2,0095 -4,8388 4,4162 -8,2558 -,3870 -5,2450 2,8727 -4,4162 4,8388 -8,6207 ,4006 -,3226 6,1705 ,3870 8,2558 -,4006 8,6207

63

Table A. 16. One-way ANOVA age and space/function (detail information) ANOVA Sum of Squares It is easy to move around ICA MAXI. Between Groups 10,310 Within Groups 235,047 Total 245,357 It is easy to move in ICA MAXI with Between Groups 18,693 the shopping cart between aisles. Within Groups 252,664 Total 271,357 The supermarket layout is logical to me. Between Groups 16,557 Within Groups 326,059 Total 342,615 I can find everything I need in ICA Between Groups 24,073 MAXI. Within Groups 327,745 Total 351,818 I think that ICA MAXI has too many Between Groups 9,027 products for the size of the store. Within Groups 473,784 Total 482,811 The interior design of ICA MAXI is Between Groups 20,103 visually appealing. Within Groups 258,861 Total 278,965 Source: SPSS outcome. Table A. 17. One-way ANOVA age and space/function (detail information) Multiple Comparisons Tukey HSD (I) Dependent Variable Age It is easy to move around ICA 18-29 MAXI. 30-39 (J) Mean Difference Age (I-J) 30-39 -,026 40-49 ,279 > 49 -,576 18-29 ,026 40-49 ,305 > 49 -,551 18-29 -,279 30-39 -,305 > 49 -,856 18-29 ,576 30-39 ,551 40-49 ,856 30-39 -,431 40-49 -,294 > 49 -,926* 18-29 ,431 40-49 ,137 > 49 -,495 18-29 ,294 30-39 -,137 > 49 -,632 18-29 ,926* 30-39 ,495 40-49 ,632 30-39 ,014 Std. Error ,296 ,353 ,282 ,296 ,403 ,342 ,353 ,403 ,392 ,282 ,342 ,392 ,307 ,366 ,292 ,307 ,417 ,355 ,366 ,417 ,407 ,292 ,355 ,407 ,348

df 3 139 142 3 139 142 3 139 142 3 139 142 3 139 142 3 138 141

Mean Square F Sig. 3,437 2,032 ,112 1,691 6,231 3,428 ,019 1,818 5,519 2,353 ,075 2,346 8,024 3,403 ,020 2,358 3,009 3,409 ,883 ,452

6,701 3,572 ,016 1,876

40-49

> 49

It is easy to move in ICA MAXI 18-29 with the shopping cart between aisles. 30-39

40-49

> 49

The supermarket layout is logical 18-29

Sig. 1,000 ,858 ,177 1,000 ,873 ,377 ,858 ,873 ,134 ,177 ,377 ,134 ,497 ,852 ,010 ,497 ,988 ,505 ,852 ,988 ,409 ,010 ,505 ,409 1,000

95% Confidence Interval Lower Upper Bound Bound -,79 ,74 -,64 1,20 -1,31 ,16 -,74 ,79 -,74 1,35 -1,44 ,34 -1,20 ,64 -1,35 ,74 -1,88 ,16 -,16 1,31 -,34 1,44 -,16 1,88 -1,23 ,37 -1,24 ,66 -1,69 -,17 -,37 1,23 -,95 1,22 -1,42 ,43 -,66 1,24 -1,22 ,95 -1,69 ,43 ,17 1,69 -,43 1,42 -,43 1,69 -,89 ,92

64

40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 I can find everything I need in 18-29 30-39 ICA MAXI. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 I think that ICA MAXI has too 18-29 30-39 many products for the size of 40-49 the store. > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 The interior design of ICA 18-29 30-39 MAXI is visually appealing. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome.

to me.

,809 -,419 -,014 ,795 -,432 -,809 -,795 -1,228* ,419 ,432 1,228* ,857 ,868 -,081 -,857 ,011 -,938 -,868 -,011 -,949 ,081 ,938 ,949 ,395 -,147 -,382 -,395 -,542 -,778 ,147 ,542 -,235 ,382 ,778 ,235 ,562 -,356 -,566 -,562 -,917 -1,128* ,356 ,917 -,211 ,566 1,128* ,211

,415 ,213 ,332 ,589 ,348 1,000 ,474 ,340 ,403 ,707 ,415 ,213 ,474 ,340 ,462 ,043 ,332 ,589 ,403 ,707 ,462 ,043 ,349 ,072 ,416 ,164 ,333 ,995 ,349 ,072 ,475 1,000 ,404 ,098 ,416 ,164 ,475 1,000 ,463 ,176 ,333 ,995 ,404 ,098 ,463 ,176 ,420 ,783 ,501 ,991 ,400 ,775 ,420 ,783 ,572 ,778 ,486 ,382 ,501 ,991 ,572 ,778 ,557 ,975 ,400 ,775 ,486 ,382 ,557 ,975 ,312 ,278 ,372 ,775 ,298 ,231 ,312 ,278 ,424 ,139 ,361 ,011 ,372 ,775 ,424 ,139 ,413 ,957 ,298 ,231 ,361 ,011 ,413 ,957

-,27 -1,28 -,92 -,44 -1,48 -1,89 -2,03 -2,43 -,44 -,62 ,03 -,05 -,22 -,95 -1,77 -1,23 -1,99 -1,95 -1,25 -2,15 -,78 -,11 -,26 -,70 -1,45 -1,42 -1,49 -2,03 -2,04 -1,15 -,94 -1,68 -,66 -,49 -1,21 -,25 -1,32 -1,34 -1,37 -2,02 -2,07 -,61 -,19 -1,29 -,21 ,19 -,86

1,89 ,44 ,89 2,03 ,62 ,27 ,44 -,03 1,28 1,48 2,43 1,77 1,95 ,78 ,05 1,25 ,11 ,22 1,23 ,26 ,95 1,99 2,15 1,49 1,15 ,66 ,70 ,94 ,49 1,45 2,03 1,21 1,42 2,04 1,68 1,37 ,61 ,21 ,25 ,19 -,19 1,32 2,02 ,86 1,34 2,07 1,29

65

Table A. 18. One-way ANOVA age and signs, symbols and artifacts (detail information) Multiple Comparisons Signs, symbols and artifacts Tukey HSD 95% Confidence Interval (I) Age (J) Age Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound 18-29 30-39 1,84848 1,23383 ,441 -1,3605 5,0575 40-49 -,56328 1,46896 ,981 -4,3838 3,2573 > 49 -3,69990* 1,17598 ,011 -6,7584 -,6414 30-39 18-29 -1,84848 1,23383 ,441 -5,0575 1,3605 40-49 -2,41176 1,67220 ,475 -6,7609 1,9374 > 49 -5,54839* 1,42174 ,001 -9,2461 -1,8507 40-49 18-29 ,56328 1,46896 ,981 -3,2573 4,3838 30-39 2,41176 1,67220 ,475 -1,9374 6,7609 > 49 -3,13662 1,62998 ,223 -7,3760 1,1027 * > 49 18-29 3,69990 1,17598 ,011 ,6414 6,7584 30-39 5,54839* 1,42174 ,001 1,8507 9,2461 40-49 3,13662 1,62998 ,223 -1,1027 7,3760 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome Table A. 19. One-way ANOVA age and signs, symbols and artifacts (detail information) ANOVA Sum of Squares df Mean Square There are enough directional signs (e.g. Between Groups 29,776 3 9,925 entrance, exit, toilets) in ICA MAXI. Within Groups 292,126 139 2,102 Total 321,902 142 The directional signs (e.g. entrance, Between Groups 23,898 3 7,966 exit, toilets) give guidelines how to Within Groups 242,941 139 1,748 move in ICA MAXI. Total 266,839 142 It is easy to follow directional signs Between Groups 13,726 3 4,575 (e.g. entrance, exit, toilets) displayed in Within Groups 218,022 139 1,569 ICA MAXI. Total 231,748 142 Directional signs are visually appealing. Between Groups 18,391 3 6,130 Within Groups 236,965 139 1,705 Total 255,357 142 Other types of communication such as Between Groups 21,282 3 7,094 brochures are visually appealing. Within Groups 270,008 137 1,971 Total 291,291 140 Source: SPSS outcome. Table A. 20. One-way ANOVA age and signs, symbols and artifacts (detail information) Multiple Comparisons Tukey HSD Mean Difference (I- Std. (J) Age J) Error Sig. 30-39 ,325 ,330 ,758 40-49 -,074 ,393 ,998 > 49 -,994* ,314 ,010 18-29 -,325 ,330 ,758 40-49 -,399 ,449 ,811 > 49 -1,319* ,382 ,004 18-29 ,074 ,393 ,998 30-39 ,399 ,449 ,811 > 49 -,920 ,438 ,157 95% Confidence Interval Lower Upper Bound Bound -,53 1,18 -1,10 ,95 -1,81 -,18 -1,18 ,53 -1,57 ,77 -2,31 -,33 -,95 1,10 -,77 1,57 -2,06 ,22

F Sig. 4,723 ,004

4,558 ,004

2,917 ,036

3,596 ,015

3,599 ,015

Dependent Variable (I) Age There are enough directional signs 18-29 (e.g. entrance, exit, toilets) in ICA MAXI. 30-39

40-49

66

18-29 30-39 40-49 The directional signs (e.g. entrance, 18-29 30-39 exit, toilets) give guidelines how to 40-49 move in ICA MAXI. > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 It is easy to follow directional signs 18-29 30-39 (e.g. entrance, exit, toilets) displayed 40-49 in ICA MAXI. > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 Directional signs are visually 18-29 30-39 appealing. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 Other types of communication such 18-29 30-39 as brochures are visually appealing. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome.

> 49

,994* 1,319* ,920 ,000 -,059 -1,000* ,000 -,059 -1,000* ,059 ,059 -,941 1,000* 1,000* ,941 ,434 ,059 -,529 -,434 -,375 -,963* -,059 ,375 -,588 ,529 ,963* ,588 ,566 -,103 -,558 -,566 -,669 -1,124* ,103 ,669 -,455 ,558 1,124* ,455 ,488 -,422 -,654 -,488 -,911 -1,142* ,422 ,911 -,231 ,654 1,142* ,231

,314 ,010 ,382 ,004 ,438 ,157 ,301 1,000 ,358 ,998 ,287 ,004 ,301 1,000 ,409 ,999 ,348 ,024 ,358 ,998 ,409 ,999 ,399 ,090 ,287 ,004 ,348 ,024 ,399 ,090 ,285 ,427 ,340 ,998 ,271 ,212 ,285 ,427 ,388 ,769 ,330 ,021 ,340 ,998 ,388 ,769 ,378 ,407 ,271 ,212 ,330 ,021 ,378 ,407 ,297 ,231 ,354 ,991 ,283 ,203 ,297 ,231 ,404 ,352 ,344 ,007 ,354 ,991 ,404 ,352 ,394 ,656 ,283 ,203 ,344 ,007 ,394 ,656 ,321 ,427 ,382 ,686 ,306 ,146 ,321 ,427 ,435 ,160 ,370 ,013 ,382 ,686 ,435 ,160 ,424 ,947 ,306 ,146 ,370 ,013 ,424 ,947

,18 ,33 -,22 -,78 -,99 -1,75 -,78 -1,12 -1,90 -,87 -1,01 -1,98 ,25 ,10 -,10 -,31 -,82 -1,24 -1,17 -1,38 -1,82 -,94 -,63 -1,57 -,18 ,11 -,39 -,21 -1,02 -1,29 -1,34 -1,72 -2,02 -,82 -,38 -1,48 -,18 ,23 -,57 -,35 -1,42 -1,45 -1,32 -2,04 -2,10 -,57 -,22 -1,33 -,14 ,18 -,87

1,81 2,31 2,06 ,78 ,87 -,25 ,78 1,01 -,10 ,99 1,12 ,10 1,75 1,90 1,98 1,17 ,94 ,18 ,31 ,63 -,11 ,82 1,38 ,39 1,24 1,82 1,57 1,34 ,82 ,18 ,21 ,38 -,23 1,02 1,72 ,57 1,29 2,02 1,48 1,32 ,57 ,14 ,35 ,22 -,18 1,42 2,04 ,87 1,45 2,10 1,33

67

Table A. 21. One-way ANOVA age and behaviour (more details) Multiple Comparisons Behaviour Tukey HSD (J) Age Mean Difference (I-J) Std. Error 30-39 4,08402* 1,10859 40-49 4,51975* 1,32070 > 49 -,19005 1,06835 30-39 18-29 -4,08402* 1,10859 40-49 ,43573 1,50573 > 49 -4,27407* 1,29010 40-49 18-29 -4,51975* 1,32070 30-39 -,43573 1,50573 > 49 -4,70980* 1,47636 > 49 18-29 ,19005 1,06835 30-39 4,27407* 1,29010 40-49 4,70980* 1,47636 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome. (I) Age 18-29 Sig. ,002 ,005 ,998 ,002 ,992 ,006 ,005 ,992 ,009 ,998 ,006 ,009 95% Confidence Interval Lower Bound Upper Bound 1,2007 6,9673 1,0848 7,9547 -2,9687 2,5886 -6,9673 -1,2007 -3,4804 4,3519 -7,6294 -,9187 -7,9547 -1,0848 -4,3519 3,4804 -8,5496 -,8700 -2,5886 2,9687 ,9187 7,6294 ,8700 8,5496

Table A. 22. One-way ANOVA age and behaviour (more details) ANOVA Sum of Squares I enjoy shopping in ICA MAXI. Between Groups 37,204 Within Groups 203,258 Total 240,462 I am satisfied with the general Between Groups 39,251 impression of ICA MAXI. Within Groups 142,651 Total 181,902 I like the environment of ICA MAXI. Between Groups 20,776 Within Groups 194,931 Total 215,706 I would like to spend more time Between Groups 28,897 browsing in ICA MAXI. Within Groups 402,862 Total 431,759 This is a place that I would avoid Between Groups 21,074 shopping in the future. Within Groups 349,384 Total 370,458 Source: SPSS outcome. Table A. 23. One-way ANOVA age and behaviour (more details) Multiple Comparisons Tukey HSD Mean Difference (I-J) ,986* ,912* -,346 -,986* -,074 -1,332* -,912* ,074 -1,258*

df Mean Square F Sig. 3 12,401 8,481 ,000 139 1,462 142 3 13,084 12,749 ,000 139 1,026 142 3 6,925 4,938 ,003 139 1,402 142 3 9,632 3,276 ,023 137 2,941 140 3 7,025 2,775 ,044 138 2,532 141

Dependent Variable (I) Age (J) Age I enjoy shopping in ICA 18-29 30-39 MAXI. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49

Std. Error Sig. ,275 ,003 ,328 ,031 ,262 ,551 ,275 ,003 ,374 ,997 ,318 ,000 ,328 ,031 ,374 ,997 ,365 ,004

95% Confidence Interval Lower Upper Bound Bound ,27 1,70 ,06 1,76 -1,03 ,34 -1,70 -,27 -1,05 ,90 -2,16 -,50 -1,76 -,06 -,90 1,05 -2,21 -,31

68

18-29 30-39 40-49 I am satisfied with the general 18-29 30-39 impression of ICA MAXI. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 I like the environment of ICA 18-29 30-39 MAXI. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 I would like to spend more 18-29 30-39 time browsing in ICA MAXI. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 This is a place that I would 18-29 30-39 avoid shopping in the future. 40-49 > 49 30-39 18-29 40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome.

> 49

,346 1,332* 1,258* ,897* 1,147* -,320 -,897* ,251 -1,216* -1,147* -,251 -1,467* ,320 1,216* 1,467* ,581 ,676 -,415 -,581 ,096 -,995* -,676 -,096 -1,091* ,415 ,995* 1,091* 1,124* ,599 ,750 -1,124* -,525 -,374 -,599 ,525 ,151 -,750 ,374 -,151 ,473 1,162* ,036 -,473 ,688 -,437 -1,162* -,688 -1,125 -,036 ,437 1,125

,262 ,551 ,318 ,000 ,365 ,004 ,230 ,001 ,275 ,000 ,220 ,467 ,230 ,001 ,314 ,855 ,267 ,000 ,275 ,000 ,314 ,855 ,306 ,000 ,220 ,467 ,267 ,000 ,306 ,000 ,269 ,141 ,321 ,156 ,257 ,373 ,269 ,141 ,367 ,994 ,312 ,009 ,321 ,156 ,367 ,994 ,357 ,014 ,257 ,373 ,312 ,009 ,357 ,014 ,391 ,024 ,466 ,574 ,377 ,197 ,391 ,024 ,531 ,756 ,455 ,844 ,466 ,574 ,531 ,756 ,521 ,991 ,377 ,197 ,455 ,844 ,521 ,991 ,362 ,560 ,431 ,039 ,349 1,000 ,362 ,560 ,493 ,503 ,422 ,729 ,431 ,039 ,493 ,503 ,483 ,096 ,349 1,000 ,422 ,729 ,483 ,096

-,34 ,50 ,31 ,30 ,43 -,89 -1,50 -,57 -1,91 -1,86 -1,07 -2,26 -,25 ,52 ,67 -,12 -,16 -1,08 -1,28 -,86 -1,81 -1,51 -1,05 -2,02 -,25 ,18 ,16 ,11 -,61 -,23 -2,14 -1,91 -1,56 -1,81 -,86 -1,20 -1,73 -,81 -1,50 -,47 ,04 -,87 -1,41 -,59 -1,53 -2,28 -1,97 -2,38 -,94 -,66 -,13

1,03 2,16 2,21 1,50 1,86 ,25 -,30 1,07 -,52 -,43 ,57 -,67 ,89 1,91 2,26 1,28 1,51 ,25 ,12 1,05 -,18 ,16 ,86 -,16 1,08 1,81 2,02 2,14 1,81 1,73 -,11 ,86 ,81 ,61 1,91 1,50 ,23 1,56 1,20 1,41 2,28 ,94 ,47 1,97 ,66 -,04 ,59 ,13 ,87 1,53 2,38

69

Table A. 24. One-way ANOVA monthly disposable income and signs, symbols and artifacts and behaviour Multiple Comparisons Tukey HSD 95% Confidence Interval (I) Monthly (J) Monthly Mean Dependent disposable disposable Difference (IStd. Lower Upper Variable income income J) Error Sig. Bound Bound Signs, symbols < 10.00 kr 10.000-20.000 kr -3,45000 1,29987 ,067 -7,0449 ,1449 and artifacts 20.001-30.000 kr -,31290 1,25891 ,999 -3,7945 3,1687 30.001-40.000 kr -,38000 1,34894 ,999 -4,1106 3,3506 > 40.000 kr -6,20000 2,86154 ,199 -14,1138 1,7138 10.000-20.000 kr < 10.000 kr 3,45000 1,29987 ,067 -,1449 7,0449 20.001-30.000 kr 3,13710 1,43577 ,192 -,8336 7,1078 30.001-40.000 kr 3,07000 1,51533 ,259 -1,1207 7,2607 > 40.000 kr -2,75000 2,94363 ,883 -10,8908 5,3908 20.001-30.000 kr < 10.000 kr ,31290 1,25891 ,999 -3,1687 3,7945 10.000-20.000 kr -3,13710 1,43577 ,192 -7,1078 ,8336 30.001-40.000 kr -,06710 1,48034 1,000 -4,1611 4,0269 > 40.000 kr -5,88710 2,92577 ,266 -13,9785 2,2043 30.001-40.000 kr < 10.000 kr ,38000 1,34894 ,999 -3,3506 4,1106 10.000-20.000 kr -3,07000 1,51533 ,259 -7,2607 1,1207 20.001-30.000 kr ,06710 1,48034 1,000 -4,0269 4,1611 > 40.000 kr -5,82000 2,96563 ,290 -14,0216 2,3816 > 40.000 kr < 10.000 kr 6,20000 2,86154 ,199 -1,7138 14,1138 10.000-20.000 kr 2,75000 2,94363 ,883 -5,3908 10,8908 20.001-30.000 kr 5,88710 2,92577 ,266 -2,2043 13,9785 30.001-40.000 kr 5,82000 2,96563 ,290 -2,3816 14,0216 Behaviour < 10.000 kr 10.000-20.000 kr -1,29000 1,19960 ,819 -4,6076 2,0276 20.001-30.000 kr 1,46000 1,16179 ,718 -1,7530 4,6730 30.001-40.000 kr 2,70000 1,24488 ,198 -,7428 6,1428 > 40.000 kr -1,04000 2,64079 ,995 -8,3433 6,2633 10.000-20.000 kr < 10.000 kr 1,29000 1,19960 ,819 -2,0276 4,6076 20.001-30.000 kr 2,75000 1,32501 ,237 -,9144 6,4144 30.001-40.000 kr 3,99000* 1,39843 ,040 ,1226 7,8574 > 40.000 kr ,25000 2,71655 1,000 -7,2628 7,7628 20.001-30.000 kr < 10.000 kr -1,46000 1,16179 ,718 -4,6730 1,7530 10.000-20.000 kr -2,75000 1,32501 ,237 -6,4144 ,9144 30.001-40.000 kr 1,24000 1,36614 ,894 -2,5381 5,0181 > 40.000 kr -2,50000 2,70007 ,886 -9,9672 4,9672 30.001-40.000 kr < 10.000 kr -2,70000 1,24488 ,198 -6,1428 ,7428 10.000-20.000 kr -3,99000* 1,39843 ,040 -7,8574 -,1226 20.001-30.000 kr -1,24000 1,36614 ,894 -5,0181 2,5381 > 40.000 kr -3,74000 2,73685 ,650 -11,3089 3,8289 > 40.000 kr < 10.000 kr 1,04000 2,64079 ,995 -6,2633 8,3433 10.000-20.000 kr -,25000 2,71655 1,000 -7,7628 7,2628 20.001-30.000 kr 2,50000 2,70007 ,886 -4,9672 9,9672 30.001-40.000 kr 3,74000 2,73685 ,650 -3,8289 11,3089 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome.

70

Table A. 25. One-way ANOVA monthly disposable income and signs, symbols and artifacts (detail information) ANOVA Sum of Squares df Mean Square F Sig. There are enough directional signs Between Groups 8,820 4 2,205 ,961 ,431 (e.g. entrance, exit, toilets) in ICA Within Groups 309,922 135 2,296 MAXI. Total 318,743 139 The directional signs (e.g. entrance, Between Groups 7,498 4 1,875 ,988 ,416 exit, toilets) give guidelines how to Within Groups 256,073 135 1,897 move in ICA MAXI. Total 263,571 139 It is easy to follow directional signs Between Groups 11,699 4 2,925 1,834 ,126 (e.g. entrance, exit, toilets) displayed in Within Groups 215,294 135 1,595 ICA MAXI. Total 226,993 139 Directional signs are visually Between Groups 16,814 4 4,203 2,445 ,050 appealing. Within Groups 232,122 135 1,719 Total 248,936 139 Other types of communication such Between Groups 19,327 4 4,832 2,424 ,051 as brochures are visually appealing. Within Groups 265,079 133 1,993 Total 284,406 137 Source: SPSS outcome. Table A. 26. One-way ANOVA monthly disposable income and signs, symbols and artifacts (detail information) Multiple Comparisons Tukey HSD 95% Confidence Interval (I) Monthly (J) Monthly Mean Dependent disposable disposable Difference (I- Std. Lower Upper Variable income income J) Error Sig. Bound Bound Directional signs < 10.000 kr 10.000-20.000 kr -,579 ,305 ,324 -1,42 ,26 are visually 20.001-30.000 kr ,103 ,299 ,997 -,72 ,93 appealing. 30.001-40.000 kr ,050 ,320 1,000 -,83 ,94 > 40.000 kr -1,510 ,681 ,180 -3,39 ,37 10.000-20.000 < 10.000 kr ,579 ,305 ,324 -,26 1,42 kr 20.001-30.000 kr ,682 ,339 ,265 -,25 1,62 30.001-40.000 kr ,629 ,358 ,403 -,36 1,62 > 40.000 kr -,931 ,699 ,672 -2,86 1,00 20.001-30.000 < 10.000 kr -,103 ,299 ,997 -,93 ,72 kr 10.000-20.000 kr -,682 ,339 ,265 -1,62 ,25 30.001-40.000 kr -,053 ,352 1,000 -1,03 ,92 > 40.000 kr -1,613 ,697 ,146 -3,54 ,31 30.001-40.000 < 10.000 kr -,050 ,320 1,000 -,94 ,83 kr 10.000-20.000 kr -,629 ,358 ,403 -1,62 ,36 20.001-30.000 kr ,053 ,352 1,000 -,92 1,03 > 40.000 kr -1,560 ,706 ,183 -3,51 ,39 > 40.000 kr < 10.000 kr 1,510 ,681 ,180 -,37 3,39 10.000-20.000 kr ,931 ,699 ,672 -1,00 2,86 20.001-30.000 kr 1,613 ,697 ,146 -,31 3,54 30.001-40.000 kr 1,560 ,706 ,183 -,39 3,51 Source: SPSS outcome.

71

Table A. 27. One-way ANOVA monthly disposable income and behaviour ANOVA I enjoy shopping in ICA MAXI. Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Between Groups Within Groups Total Sum of Squares 20,013 215,208 235,221 10,971 166,021 176,993 14,121 194,850 208,971 21,871 387,608 409,478 14,580 329,578 344,158 df Mean Square F Sig. 4 5,003 3,139 ,017 135 1,594 139 4 2,743 2,230 ,069 135 1,230 139 4 3,530 2,446 ,049 135 1,443 139 4 5,468 1,876 ,118 133 2,914 137 4 3,645 1,482 ,211 134 2,460 138

I am satisfied with the general impression of ICA MAXI. I like the environment of ICA MAXI. I would like to spend more time browsing in ICA MAXI. This is a place that I would avoid shopping in the future. Source: SPSS outcome.

.Table A. 28. One-way ANOVA monthly disposable income and signs, symbols and artifacts Multiple Comparisons Tukey HSD (I) Monthly Dependent disposable Variable income I enjoy shopping < 10.000 kr in ICA MAXI. (J) Monthly Mean disposable Difference (I- Std. income J) Error 10.000-20.000 kr -,584 ,294 20.001-30.000 kr ,137 ,288 30.001-40.000 kr ,308 ,308 > 40.000 kr -1,412 ,656 < 10.000 kr ,584 ,294 20.001-30.000 kr ,721 ,326 30.001-40.000 kr ,892 ,345 > 40.000 kr -,828 ,673 < 10.000 kr -,137 ,288 10.000-20.000 kr -,721 ,326 30.001-40.000 kr ,172 ,339 > 40.000 kr -1,548 ,671 < 10.000 kr -,308 ,308 10.000-20.000 kr -,892 ,345 20.001-30.000 kr -,172 ,339 > 40.000 kr -1,720 ,680 < 10.000 kr 1,412 ,656 10.000-20.000 kr ,828 ,673 20.001-30.000 kr 1,548 ,671 30.001-40.000 kr 1,720 ,680 10.000-20.000 kr -,588 ,279 20.001-30.000 kr ,121 ,274 30.001-40.000 kr ,252 ,293 > 40.000 kr -,838 ,624 < 10.000 kr ,588 ,279 20.001-30.000 kr ,710 ,310 30.001-40.000 kr ,840 ,328 95% Confidence Interval Lower Upper Bound Bound -1,40 ,23 -,66 ,93 -,54 1,16 -3,22 ,40 -,23 1,40 -,18 1,62 -,06 1,85 -2,69 1,03 -,93 ,66 -1,62 ,18 -,77 1,11 -3,40 ,31 -1,16 ,54 -1,85 ,06 -1,11 ,77 -3,60 ,16 -,40 3,22 -1,03 2,69 -,31 3,40 -,16 3,60 -1,36 ,18 -,64 ,88 -,56 1,06 -2,56 ,89 -,18 1,36 -,15 1,57 -,07 1,75

10.000-20.000 kr

20.001-30.000 kr

30.001-40.000 kr

> 40.000 kr

I like the < 10.000 kr environment of ICA MAXI. 10.000-20.000 kr

Sig. ,277 ,989 ,855 ,204 ,277 ,182 ,078 ,735 ,989 ,182 ,987 ,149 ,855 ,078 ,987 ,090 ,204 ,735 ,149 ,090 ,224 ,992 ,911 ,664 ,224 ,156 ,083

72

20.001-30.000 kr

30.001-40.000 kr

> 40.000 kr

> 40.000 kr < 10.000 kr 10.000-20.000 kr 30.001-40.000 kr > 40.000 kr < 10.000 kr 10.000-20.000 kr 20.001-30.000 kr > 40.000 kr < 10.000 kr 10.000-20.000 kr 20.001-30.000 kr 30.001-40.000 kr

-,250 -,121 -,710 ,130 -,960 -,252 -,840 -,130 -1,090 ,838 ,250 ,960 1,090

,641 ,274 ,310 ,323 ,638 ,293 ,328 ,323 ,647 ,624 ,641 ,638 ,647

,995 ,992 ,156 ,994 ,562 ,911 ,083 ,994 ,447 ,664 ,995 ,562 ,447

-2,02 -,88 -1,57 -,76 -2,72 -1,06 -1,75 -1,02 -2,88 -,89 -1,52 -,81 -,70

1,52 ,64 ,15 1,02 ,81 ,56 ,07 ,76 ,70 2,56 2,02 2,72 2,88

Source: SPSS outcome. Table A. 29. One-way ANOVA monthly disposable income and the statement How often do you visit ICA MAXI? (detail information) Multiple Comparisons How often do you visit ICA MAXI? Tukey HSD 95% Confidence Interval (I) Monthly (J) Monthly Mean Difference Std. Lower Upper disposable income disposable income (I-J) Error Sig. Bound Bound < 10.000 kr 10.000-20.000 kr -,747 ,309 ,117 -1,60 ,11 20.001-30.000 kr -1,075* ,303 ,005 -1,91 -,24 30.001-40.000 kr -1,013* ,325 ,018 -1,91 -,12 > 40.000 kr -1,583 ,690 ,153 -3,49 ,32 10.000-20.000 kr < 10.000 kr ,747 ,309 ,117 -,11 1,60 20.001-30.000 kr -,328 ,343 ,874 -1,28 ,62 30.001-40.000 kr -,266 ,363 ,948 -1,27 ,74 > 40.000 kr -,836 ,709 ,763 -2,80 1,12 20.001-30.000 kr < 10.000 kr 1,075* ,303 ,005 ,24 1,91 10.000-20.000 kr ,328 ,343 ,874 -,62 1,28 30.001-40.000 kr ,062 ,357 1,000 -,93 1,05 > 40.000 kr -,508 ,706 ,952 -2,46 1,44 30.001-40.000 kr < 10.000 kr 1,013* ,325 ,018 ,12 1,91 10.000-20.000 kr ,266 ,363 ,948 -,74 1,27 20.001-30.000 kr -,062 ,357 1,000 -1,05 ,93 > 40.000 kr -,570 ,716 ,931 -2,55 1,41 > 40.000 kr < 10.000 kr 1,583 ,690 ,153 -,32 3,49 10.000-20.000 kr ,836 ,709 ,763 -1,12 2,80 20.001-30.000 kr ,508 ,706 ,952 -1,44 2,46 30.001-40.000 kr ,570 ,716 ,931 -1,41 2,55 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome Table A. 30. One-way ANOVA age and the statement How often do you visit ICA MAXI? (detail information) Multiple Comparisons How often do you visit ICA MAXI? Tukey HSD 95% Confidence Interval (I) Age (J) Age Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound 18-29 30-39 -,637 ,310 ,173 -1,44 ,17 40-49 -,868 ,369 ,092 -1,83 ,09 > 49 -,528 ,295 ,283 -1,30 ,24 30-39 18-29 ,637 ,310 ,173 -,17 1,44

73

40-49 > 49 40-49 18-29 30-39 > 49 > 49 18-29 30-39 40-49 Source: SPSS outcome.

-,231 ,109 ,868 ,231 ,340 ,528 -,109 -,340

,421 ,358 ,369 ,421 ,411 ,295 ,358 ,411

,947 ,990 ,092 ,947 ,842 ,283 ,990 ,842

-1,33 -,82 -,09 -,87 -,73 -,24 -1,04 -1,41

,87 1,04 1,83 1,33 1,41 1,30 ,82 ,73

Table A. 31. One-way ANOVA age and the statement I would prefer if ICA MAXI would play music (detail information) Multiple Comparisons I would prefer if ICA MAXI would play music. Tukey HSD 95% Confidence Interval (I) Age (J) Age Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound 18-29 30-39 ,088 ,483 ,998 -1,17 1,34 40-49 ,000 ,568 1,000 -1,48 1,48 > 49 1,491* ,454 ,007 ,31 2,67 30-39 18-29 -,088 ,483 ,998 -1,34 1,17 40-49 -,088 ,654 ,999 -1,79 1,61 > 49 1,403 ,557 ,061 -,05 2,85 40-49 18-29 ,000 ,568 1,000 -1,48 1,48 30-39 ,088 ,654 ,999 -1,61 1,79 > 49 1,491 ,632 ,090 -,15 3,14 > 49 18-29 -1,491* ,454 ,007 -2,67 -,31 30-39 -1,403 ,557 ,061 -2,85 ,05 40-49 -1,491 ,632 ,090 -3,14 ,15 *. The mean difference is significant at the 0.05 level. Source: SPSS outcome. Table A. 32. Eigenvalues Factor analysis Total Variance Explained Component 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Total 8,861 2,748 1,873 1,651 1,493 1,254 1,044 ,885 ,834 ,727 ,691 ,625 ,554 ,501 ,473 ,442 ,378 ,311 ,295 ,255 ,238 Initial Eigenvalues Extraction Sums of Squared Loadings % of Variance Cumulative % Total % of Variance Cumulative % 32,820 32,820 8,861 32,820 32,820 10,176 42,996 2,748 10,176 42,996 6,939 49,935 1,873 6,939 49,935 6,115 56,050 1,651 6,115 56,050 5,529 61,579 1,493 5,529 61,579 4,644 66,223 1,254 4,644 66,223 3,867 70,089 1,044 3,867 70,089 3,277 73,366 3,087 76,454 2,692 79,145 2,559 81,705 2,313 84,018 2,054 86,071 1,855 87,926 1,753 89,679 1,639 91,318 1,400 92,718 1,152 93,871 1,091 94,962 ,943 95,905 ,882 96,786

74

22 ,229 ,848 97,635 23 ,182 ,673 98,308 24 ,167 ,618 98,925 25 ,122 ,450 99,376 26 ,093 ,346 99,722 27 ,075 ,278 100,000 Extraction Method: Principal Component Analysis. Source: SPSS outcome. Graph 3. Scree Plot

Source: SPSS outcome.

APPENDIX 2 QUESTIONNAIRES 1. IN SWEDISH (A PRINTED VERSION ) Vi r tv magisterstudenter frn Handelshgskolan i Jnkping som vill underska hur olika faktorer kan pverka konsumenter i livsmedelsbutiker. Underskningen r endast fr kunder hos ICA MAXI i Jnkping, Sverige. Denna underskning tar ca 5-6 minuter att fylla i och resultatet r vldigt anvndbart fr vr magisteruppsats. Alla som deltar kommer frbli anonyma och den information som vi samlar in kommer endast anvndas som underlag fr vr underskning.

75

1. Hur ofta besker du ICA MAXI? o Mindre n en gng i mnaden o En gng i mnaden o 2-3 ggr/ mnad o En gng i veckan o 2-3 ggr/ vecka o Varje dag 2. Jag skulle tycka det vore trevligt om ICA MAXI spelade musik i butiken. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 3. Ljussttningen i ICA MAXI r: Instmmer inte alls Passande Inbjudande Behaglig Fr mrk Instmmer helt

4. Frgskalan som anvnds i ICA MAXI r: Instmmer inte alls Passande Lugnande Trevlig Attraktiv Fr ljus

Instmmer helt

5. Ljudnivn i ICA MAXI r behaglig fr mig. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 6. Det r enkelt att rra sig genom butiken. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 7. Det r enkelt att rra sig genom gngarna med en kundvagn. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt

76

8. Butikens uppbyggnad r logisk fr mig. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 9. Jag kan hitta allt jag sker i butiken. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 10. Jag tycker att ICA MAXI har fr mnga produkter med tanke p butikens storlek. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 11. Inredningen i butiken r visuellt attraktiv. Instmmer alls inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer helt

12. Har du ngra kommentarer om ICA MAXIs planlsning?

13. Jag tycker det r tillrckligt med anvisningsskyltar (t.ex., ingng, utgng, toaletter) inne i butiken. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 14. Anvisningsskyltarna (t.ex. ingng, utgng, toaletter) visar tydligt hur jag ska hitta inne i butiken. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 15. Det r enkelt att flja skyltanvisningarna uppsatta i ICA MAXI. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 16. Anvisningskyltarna r visuellt tilltalande. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt

77

17. Andra typer av kommunikation, ssom broschyrer r visuellt tilltalande. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 18. Jag gillar att handla p ICA MAXI. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 19. Jag r tillfreds med det generella intrycket av ICA MAXI. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 20. Jag gillar miljn i ICA MAXI. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 21. Jag skulle vilja spendera mer tid till att strosa runt i butiken. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt 22. Detta r en butik jag skulle vilja undvika att handla hos i framtiden. Instmmer inte nccccccc Nccccccc nccccccc nccccccc nccccccc Instmmer alls helt lder o o o o

18-29 30-39 40-49 > 49

Kn o Man o Kvinna Nationalitet o Svensk o Internationell Yrke Heltid Deltid Student

78

 Pensionr Arbetsls Genomsnittliga ln per mnad o < 10.000 kr o 10.000 20.000 kr o 20.001 30.000 kr o 30.001 40.000 kr o > 40.000 kr Tack fr att du tog dig tid att fylla i underskningen. 2. IN ENGLISH (AN E -VERSION) We are two Master students from Jnkping International Business School who want to investigate factors that can affect customers in the supermarket. The survey is carried out only among customers of ICA MAXI in Jnkping, Sweden. This survey will take approximately 5-6 minutes of your time but the results will be very useful for our Master Thesis. All respondents will remain anonymous and the data collected will only be used in the purpose of this research! If you have any questions regarding the survey, do not hesitate to contact Maria Maciejewska (mama1187@student.hj.se). 1. How often do you visit ICA MAXI? o Never o Less than Once a Month o Once a Month o 2-3 Times a Month o Once a Week o 2-3 Times a Week o Daily 2. I would prefer if ICA MAXI would play music. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 3. The lightning in ICA MAXI is: Strongly disagree Appropriate Inviting Comfortable Too dark Strongly agree

79

4. Colours used in ICA MAXI are: Strongly disagree Appropriate Soothing Pleasant Attractive Too bright

Strongly agree

5. The noise level in ICA MAXI is acceptable for me. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 6. It is easy to move around ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 7. It is easy to move in ICA MAXI with the shopping cart between aisles. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 8. The supermarket layout is logical to me. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 9. I can find everything I need in ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 10. I think that ICA MAXI has too many products for the size of the store. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 11. The interior design of ICA MAXI is visually appealing. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 12. Do you have any comments about the layout of ICA MAXI?

13. There are enough directional signs (e.g. entrance, exit, toilets) in ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree

80

14. The directional signs (e.g. entrance, exit, toilets) give guidelines how to move in ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 15. It is easy to follow directional signs (e.g. entrance, exit, toilets) displayed in ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 16. Directional signs are visually appealing. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 17. Other types of communication, such as brochures, are visually appealing. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 18. I enjoy shopping in ICA MXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 19. I am satisfied with the general impression of ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 20. I like the environment of ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 21. I would like to spend more time browsing in ICA MAXI. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree 22. This is a place that I would avoid shopping in the future. Strongly disagree nccccccc Nccccccc nccccccc nccccccc nccccccc Strongly agree Age o o o o

18-29 30-39 40-49 > 49

Gender o Male o Female

81

Nationality o Swedish o International Employment status Full time worker Part time worker Student Retired Unemployed Monthly disposable income o < 10.000 kr o 10.000 20.000 kr o 20.001 30.000 kr o 30.001 40.000 kr o > 40.000 kr Thank you for taking the time to complete this survey.

82