Faculty of Economics and Business Administration

Food-Related Lifestyle &

‘Functional Food’ & Cultural Background

“Food is an integral part of our society and necessary for survival„

Course:

Thesis Master Marketing

Student:

Suzanne van der Zouwen

Student number:

1417703

Supervisor:

Dr. L. Paas

Summary

Lately, new kinds of food, so-called functional food, have been developed and launched in the market place. They provide a novel approach to the idea of healthy eating by linking a single product component to a certain health effect in a single product. This research focuses on evaluating the consumption of functional food in respect to citizens’ lifestyle as well as cultural background. The central problem statement of this study is: What is the relationship between cultural background, lifestyle, and functional food consumption in the Netherlands?

Lifestyle in this research is specifically linked to food using the food-related lifestyle instrument developed by Grunert & Brunso (1993). Citizens are segmented in one of the following food-related lifestyle segments (1) the Careless food consumer segment, (2) the Uninvolved food consumer segment, (3) the Rational food consumer segment, (4) the Conservative food consumer segment, and (5) the Adventurous food consumer segment. Citizens in the Netherlands with a Turkish background are interviewed and compared with citizens with a Dutch background in order to reveal cultural differences in lifestyle choice and functional food consumption.

Hypotheses regarding the relationship between the three concepts were based on a literature study and empirically investigated. The assumption that membership of a particular food-related lifestyle segment relates to the consumption of functional food was not confirmed in this research. The expectation that citizens’ cultural background influences membership of a particular segment was confirmed; however this was found in other segments than initially expected on the basis of literature. The fact that the respondents of this research are university students is a possible explanation of this outcome. Students with a Turkish background are probably not representative of the Turkish population in the Netherlands.

Vrije Universiteit Amsterdam

2

2006

Whether a relation exists between cultural background and the consumption of functional food could not be funded theoretically, because literature did not offer substantial information. The empirical research revealed –surprisingly- that respondents with a Turkish cultural background consumed significantly more functional food than those with a Dutch cultural background. It is reasonable to assume that respondents with a Turkish background in this study belong to the part of the Turkish population in the Netherlands which is relatively modern, integrated and keen on development.

The sample used for this study was taken from students, studying at the Vrije Universiteit in the period between May 5 and 20, 2006. Respondents were split in two groups based on their place of birth and that of their parents. The survey consisted of 77 questions, including 69 questions from the FoodRelated Lifestyle instrument developed by Grunert & Brunso (1993), plus questions related to the consumption of functional food and for profiling respondents’ cultural background. Several statistical tests were carried out to evaluate the relation between different explanatory variables, including demographics, on the consumption of functional food.

Given the limits of this research, the results indicate that segmenting consumers into a particular foodrelated lifestyle does not significantly affect the consumption of functional food. Cultural background, however, has proven to influence/impact the segmentation of consumers into a particular food-related lifestyle. In addition, cultural background does influence the consumption of functional food. The outcomes of this study suggest that citizens with a Turkish background are a relevant target group. Strategies to promote functional food, should not only be focused on health aspects of food, but also include cultural values.

Vrije Universiteit Amsterdam

3

2006

Acknowledgements

This thesis is the academic paper meant to complete my Master Marketing at the Vrije Universiteit in Amsterdam. I started this study after finishing my Bachelors degree. Recently I followed the course cross-cultural marketing, which inspired me for this particular subject. Food and health issues have my interest since I like to feel fit and healthy and I believe that a healthy mind lives in a healthy body.

I would like to thank my supervisor Dr. L. Paas for his guidance throughout this project. I especially thank him for his specific remarks on every part I handed in. My sincere appreciation goes to Prof Dr J. de Savornin Lohman for her honest and professional guidance and especially for encouraging me to finish this thesis.

My dearest thanks go to my parents, for their support not only during the completion of my thesis but over the duration of my entire study. Finally, I wish my most loving thanks for my boyfriend for his words of encouragement and for coping with me during this stressful period the last couple of months.

Amsterdam, June 2006

Suzanne van der Zouwen

Vrije Universiteit Amsterdam

4

2006

Contents

Summary Acknowledgements

CHAPTER 1

8

Introduction The consumption of functional food, of consumers belonging to different sub-cultures in the Netherlands. 1.1 ‘Functional Food’: A Booming Market 1.1.1 Defining Functional Food 1.2 Research Motivation: Specific Knowledge of Target Groups

8 9 9

(lifestyle & culture) 1.2.1 Lifestyle & Food

10

1.2.2 Cultural Background & Food

11

1.2.3 Aims of the Study

11

1.2.4 Motivation for Studying Citizens with a Turkish Background

12

1.3 Problem Statement

CHAPTER 2

13

1.3.1 Sub-questions

13

1.3.2 Academic Relevance

13

1.3.3 Managerial Relevance

14

1.3.4 Limitations of this Study

15

1.3.5 Overview of this Study

15

Discussion of the Relevant Literature

16

2.1 Lifestyle segmentation

16

2.1.1 Lifestyle Approaches in Marketing Research

17

2.1.2 Criticism on Lifestyle Research

18

2.1.3 Food-related Lifestyle Aspects

19

2.1.4 Food-related Lifestyle Segments

21

2.2 Culture & Food Related Lifestyle

26

2.2.1 Turkish Food Consumers

27

2.2.2 Turkish & Functional Food

31

2.3 Conceptual Framework

Vrije Universiteit Amsterdam

33

5

2006

CHAPTER 3

CHAPTER 4

Methods and Means of Data Collection

35

3.1 Sample

35

3.2 Questionnaire

36

3.3 Data Analysis

37

3.3.1 Testing Hypotheses 1 & 2& 3 & 4 & 5

37

3.3.2 Testing Hypotheses 6 & 7

39

3.3.3 Investigating the Exploratory Research Question

40

3.3.4 Partial Correlations

41

3.3.5 Outline of the Questionnaire

42

Analysis of the Collected Data

43

4.1 Pre-Analysis

43

4.1.1 Socio-Demographics

44

4.1.2 Analysis of Scale Reliability

44

4.2 Clustering Respondents

44

4.3 Segments & Functional Food Consumption

47

4.4 Cultural Background & Segments

48

4.4.1 Hypothesis 6

49

4.4.2 Hypothesis 7

49

4.4.3 Other Differences

50

4.5 Exploratory Research Question 4.5.1 Regression for Consumption of Functional Food and Cultural

50 51

Background 51

4.6 Partial Correlations 4.6.1 Effects of Demographics

52

CHAPTER 5 Conclusions & Recommendations & Areas for Future Research 5.1 Answering the Problem Statement

54 54

5.1.1 Expectations and Conclusions of the first Sub-question

54

5.1.2 Expectations and Conclusion of the second Sub-question

55

5.1.3 Expectations and Conclusions of the third Sub-question

56

5.1.4 Answering the Problem Statement

57

5.2 Theoretical Contribution

57

5.3 Managerial Implications

58

5.4 Limitations

59

5.5 Suggestions for Future Research

60

Vrije Universiteit Amsterdam

6

2006

REFERENCES

APPENDICES Appendix 1: Questionnaire Appendix 2: Kolmogorov-Smirnov Test Results Appendix 3: Outliers Appendix 4: Reliability Statistics Appendix 5: Final Cluster Centres Appendix 6: Aspects of Segments Appendix 7: Post Hoc Test Results Appendix 8: Binomial Test Results Appendix 9: Result of Mann-Whitney Test Appendix 10: Correlations Appendix 11: Cultural Background & Functional Food Consumption moderated for Demographics Appendix 12: Cultural Background & Segments moderated for Demographics Appendix 13: Segments & Functional Food Consumption moderated for Demographic Appendix 14: Household sizes split according to Cultural Background

Vrije Universiteit Amsterdam

7

2006

1

Introduction

This chapter contains a general introduction together with the background to why the problem area is relevant and interesting to look into.

“An apple a day keeps the doctor away”

1.1

‘Functional Food’: A Booming Market

Interest in functional food has been driven by a rapid development of scientific knowledge of the importance of a healthy diet, technical advances in the food industry, increased consumer demands for health promoting food products as well as predictions for an ageing population and increasing health costs (Biström & Nordström, 2002). As a result food industry companies have rather high expectations in food products that meet the consumers’ demand for a healthy lifestyle. Functional food may be used in the context of a healthy lifestyle or as a means to compensate for an unhealthy lifestyle. As a consequence of this increasing interest in improving or maintaining health in a proactive and convenient approach, consumers have become more concerned about nutrition, health and of the quality of the food they eat (Gil et al., 2000). In this context functional food plays a specific role, because these foods are not intended only to satisfy hunger and provide humans with necessary nutrients, but also to prevent nutrition related diseases and increase the physical and mental well-being of consumers (Menrad, 2003).

In the next few decades a range of newly developed functional foods will be introduced, together with media messages and advertisements on the need to optimise nutrition, health and quality of life (Wrick, 1995). Food manufacturing companies are developing products that will help control weight, improve general health, prevent ageing, and lower the risk of diseases. According to the available future market estimations, Menrad (2003) assumes that functional food will increase its market volume in the coming years considerably. In addition, increased consumer understanding of the

Vrije Universiteit Amsterdam

8

2006

relationship between diet and health is driving demand for added value food products that provide a health benefit beyond basic nutrition. The global functional foods market is experiencing fast growth and is driven by factors like greater consumer interest in the diet and health relationship, ageing populations and the associated rising health care costs, and advances in food technology (Hasler, 1998). The risk of common diseases such as cancer, coronary heart disease, diabetes and osteoporosis could be considerably decreased by the food that we eat. Thus, the food industry is faced with the big challenge to develop and promote healthy and tasty food products. A growing number of consumers prefer health enhancing food products that enhance their quality of life, improve mental and physical well being, help decrease disease and promote long life (Sloan, 1999).

1.1.1

Defining Functional Food

Strictly speaking, all food is functional, in that it provides energy and nutrients necessary for survival. But the term ‘functional food’ used today can be defined as foods which promise health benefits above and beyond basic nutritional value. Examples include the cholesterol-reducing spreads like Benecol and Becel proactive, milk enriched with calcium, dairy products containing live lactic acid bacteria (probiotics), e.g. Actimel, Activia and Yakult, fortified breakfast cereals, drinks with added vitamins and minerals e.g. Aquarius and Carpe Diem, and bread with extra fibres (inuline) like Flora Vitaalbrood and Blue Band. Using this definition and taking these kinds of products in mind, the traditional apple at which is referred to at the start of this paragraph is not meant as a ‘functional food’ in this research.

1.2

Research Motivation: Specific Knowledge of Target Groups (Lifestyle and Culture)

So far the available (research) literature suggests a response from the food industry on changing consumers demand. However, the probability of changing consumer behaviour is just as relevant, both from the perspective of food industry as well as from the perspective of the government. The aim of the food industry is primarily to expand its market share, while government policies are directed

Vrije Universiteit Amsterdam

9

2006

toward improving the health of its citizens. From both perspectives the need for specific information of target groups is obvious.

1.2.1

Lifestyle & Food

In order to differentiate between consumers, the concept of lifestyle will be used in this study. Studying consumers, lifestyle instruments have been used to become aware of trends in order to analyse differences and similarities across markets in search of global or cross-national segments. Most lifestyle studies are used to classify consumers on specific dimensions, which lead to particular lifestyle segments. Well-known lifestyle instruments like RISC, CCA and VALS mainly work in this way. It can be assumed that specific lifestyles relate to a particular kind of food consumption behaviour, as a person’s lifestyle or way of life and health related behaviour are closely interrelated. In addition, food serves a much broader role than just satisfying hunger needs and providing a sense of satiety (Saguy & Moskowitz, 1999). Numerous demographic, economic, social, and cultural changes in consumer lifestyles continue to have a significant impact on attitudes towards food products. Insight in different lifestyles makes it also possible to fine-tune the communication to the public (advertising or information service); thus improving its effectiveness. Therefore it is interesting to investigate the relationship between lifestyle and the consumption of functional foods.

In this study, a domain specific lifestyle segmentation technique is used as a way of obtaining product specific data about consumers. Grunert and Brunsø (1993) developed the food-related life style instrument, which distinguishes a new definition of lifestyle and an outline of a model for lifestyle in the context of food and drink. It is rooted in the cognitive approach to explain human behaviour, and the main claim is that lifestyle is how consumers mentally link products to the attainment of live values (Grunert & Brunsø, 1993). This will be discussed more in-depth in chapter 2.

Vrije Universiteit Amsterdam

10

2006

1.2.2

Cultural Background & Food

Another special challenge in discerning target groups is the diversity of cultural background of consumers in the Netherlands. Patterns of global migration resulted that large numbers of people have moved to western nations, influencing the dietary habits and lifestyle of many migrants (Shatenstein & Ghadirian, 1997). Eating habits contribute to define a society, a family, and a personality. Cultural tastes, personal preferences, and financial status all influence persons’ lifestyle and eating habits (Gochman, 1988). Because culture is one of the many factors that can influence the development of lifestyle (Gochman, 1988), and numerous studies link lifestyles to cultural differences (Douglas & Urban, 1977, Joy & Laroche, 1990) it is interesting to know whether belonging to a particular food related lifestyle is as well determined by a person’s cultural background.

Furthermore there is evidence showing that cultural identity and food habits are strongly linked and highlight the strength of traditional food habits in different cultural environment (Carlson et al., 1984). Food is a cultural symbol (Edles, 2004); it is one of the cultural traits that people learn first from childhood, and one that consumers change with the greatest reluctance at older age (Gabaccia, 1998). Cultural differences lead to differences in the habitual consumption of certain foods and in traditions of preparation, and in certain cases can lead to restrictions such as exclusion of meat and milk from the diet (Lau et al., 1984). Awareness and trial of ethnic cuisine has increased in the past decades as a consequence of the growing international trade, globalisation, migration and tourism. Because cultural differences leads to different food habits in general it is therefore interesting to investigate as well if culture influences the consumption of functional food.

1.2.3

Aims of the Study

As overall food habits are proven to be influenced by lifestyle and culture, this study aims to explore in an aggravated research design the exact nature of the relationship between lifestyle, cultural background and the consumption of functional food. In sum, the aim of this study is threefold: (1) to assess factors that belong to a particular lifestyle and how lifestyle correlates with the consumption of functional food;

Vrije Universiteit Amsterdam

11

2006

(2) to explore whether cultural background, in particular Turkish background segments citizens into a particular lifestyle (3) to explore the influence of cultural background, specifically Turkish background on the consumption of functional food. The results of this research can be useful for both marketing efforts as well as for government policies aimed at promoting the consumption of functional food.

1.2.4

Motivation for Studying Citizens with a Turkish Background

In the past decades many people from different countries such as Indonesia, Suriname, Hungary, Italy, and more recently Turkey, came to the Netherlands. Nowadays the largest group of immigrants consists of people with a Turkish background. According to a study of the CBS (Central Bureau of Statistics) there will be 425.000 Turks living in the Netherlands in 2015. With dietary habits and lifestyle undergoing transition as a result of migration, cultural background has become more relevant to for instance overweight and obesity in the Netherlands. For example, the risk of overweight among both children and adult women is higher for Turkish than for Dutch citizens (Brussaard, 2001). Because of the existing differences in eating habits including both preparation and kind of nutrition intake, it is interesting to investigate to what extent cultural background and lifestyle influences one’s consumption of functional foods. Diverse minority groups have been under-represented in research related to eating habits. In order for health promotion interventions to be based on an understanding of cultural values, this study set out to examine the cultural and lifestyle influences on the consumption of functional foods. Given the different cultural background and eating habits, it is plausible to assume a deviation in the consumption of functional food of Turkish people, compared with people with a Dutch background. Efforts directed toward the promotion of functional food, both by government as well as by industry, will benefit from specific knowledge concerning this target group. Finally: the fact that the financial resources of these citizens show an increase underlines to the relevance of this target group for the food industry.

Vrije Universiteit Amsterdam

12

2006

1.3

Problem Statement

Based on marketing, psychology and nutrition literature, concepts will be explored to explain different factors, which affect cross-cultural food-related consumer behaviour and their consumption of functional food in The Netherlands.

What is the relationship between cultural background, lifestyle, and functional food consumption in the Netherlands? 1.3.1

Sub-questions

In order to be able to answer the problem statement the following sub-questions are formulated: 1. What is the relationship between respondents representative of a certain lifestyle segment and their consumption of functional food? 2. What is the relationship between respondents with a Dutch or Turkish background and the segmentation into different lifestyle segments? 3. What is the relationship between respondents with a Turkish background and their consumption of functional food?

1.3.2

Academic Relevance

It is generally agreed that food related behaviour is complex and determined by the interplay of many factors, including physiological factors; socio-demographic characteristics such as education, income, ethnicity and availability of food; behavioural and lifestyle factors; and knowledge and attitudes related to diet and health (Slack, 1996). Despite increasing attempts to include people of culturally and linguistically diverse backgrounds in research, these groups have been consistently under-represented in research more generally (Small et al. 1999) and also understudied in relation to eating habits and physical activity more specifically (Treloar, 1999). More specific research is needed in order to specify the exact nature of these relationships. However modest, this study will add to a better understanding of lifestyle and functional food consumption in relation to cultural background. This study uses the food related lifestyle instrument developed by Grunert and Brunsø in 1993, which – to my knowledge - has not been used prior in the Netherlands. Vrije Universiteit Amsterdam

13

2006

1.3.3

Managerial Relevance

Research has been conducted in the Netherlands among citizens with a Dutch and Turkish background.

As mentioned before, this particular sub-culture has been chosen because of the

increasing number of citizens with a Turkish background in The Netherlands. As a result of increasing interest in improving or maintaining health in a proactive and convenient approach (Jong et al. 2003), consumers have become more concerned about nutrition, health, and quality of the food they eat (Gil et al., 2000). Consequently the field of diet and health is growing fast and the food industry is focused on developing products with positive nutritional advantages. For that reason it is interesting to investigate whether this increasing group of consumers behave according this latest health trend.

The results of this research support the communication of functional benefits that are credible and convincing, easy to understand, and representative of science. The need for health promotion interventions to be based on an understanding of the cultural values concerning food choice, eating customs, physical activity and body shape has recently been highlighted (Haikerwal et al. 2002).

Because both groups (citizens with a Dutch and Turkish background) have access to the same information data, e.g. magazines, government, news and sport facilities it is therefore an honest comparison to make between those two groups. The results of this study will show whether food industry companies can best target specific ethnic groups or specific lifestyle groups.

The combination of demographics (necessary for targeting) with psychographics (necessary for understanding) symbolizes an important step forward and, I hope, puts aside the argument over whether demographics or lifestyles are more important. Finally, any marketer involved in brand development and management will welcome the chance to get deeper into their consumers’ psyche without losing site of practical marketing concerns such as how to reach the segments once they have been identified. Just like the promotion of brown bread since the 1960’s, which elicited great changes

Vrije Universiteit Amsterdam

14

2006

in consumption of this product it must be possible to reach and change consumers minds about the consumption of functional food.

1.3.4

Limitations of this Study

The results need to be interpreted critically considering the limitations encountered in this study. The research is of a quantitative, explorative nature. The first limitation of this study is that the quantitative data is collected in Amsterdam among students with a Turkish, and Dutch background. Thus the results do not apply for all Turks, and Dutch people in the Netherlands. The second limitation of the study can be ascribed to data obtained from self-administered questionnaires filled in by a limited amount of citizens who are not representative of the Dutch population.

1.3.5

Overview of this Study

In this chapter the aim of the study as well as its theoretical and practical relevance are presented. To summarize, the aim is to enhance the insight in the relationship between lifestyle, cultural background and functional food consumption. In chapter two the relevant literature and research results will be presented and discussed. The choice for research instruments will be substantiated. Based on the selected literature hypotheses have been formulated, as building blocks for the research design presented in chapter three, in order to finally be able to answer the sub-questions. The results of the empirical part of the study will be presented in chapter four. The last chapter will place the results in a general perspective of the problem statement: what is the relationship between lifestyle, cultural background and functional food consumption.

Vrije Universiteit Amsterdam

15

2006

2

Discussion of Relevant Literature

The survey of theories in this chapter spring from the problem area discussed in previous chapter. What will be described in this chapter will be the foundation for continued work. In paragraph 2.1 lifestyle segmentation instruments will be evaluated and the food-related lifestyle concept will be presented as well as aspects which potentially influence functional food consumption behaviour. By doing this the first sub-question is discussed and the first step is taken to answer the problem statement. In paragraph 2.2 culture will be discussed in relationship with food consumption. The first part of paragraph 2.2 addresses sub-question 2; differences in food consumption habits of citizens with a Turkish background in the Netherlands are given and placed in relation to food-related lifestyle segments. The second part of paragraph 2.2 deals with sub-question3; here are the aspects placed in relation to the consumption of functional food. Dealing with the three sub-questions is a first step necessary to eventually be able to answer the problem statement. Finally a conceptual framework will be presented representing the variables included in the research project.

2.1

Lifestyle Segmentation

“Lifestyle” has been defined simply as “how one lives” (Kucukemiroglu).

In marketing however, lifestyle is described as the behaviour of individuals, a small group of interacting people, and large groups of people (e.g. market segments) acting as potential consumers. Lifestyle relates to the economic level at which people live, how they spend their money, and how they allocate their time (Anderson & Golden, 1984). Lifestyle segmentation research measures people’s activities in terms of: •

How they spend their money;

•

What interests they have and what importance they place on their immediate surroundings;

•

Their views of themselves and the world around them; and

Vrije Universiteit Amsterdam

16

2006

•

Some basic demographic characteristics

The concept of lifestyle has been very popular in academic as well as in practical marketing research in the latest decades. It is an attempt to collect purchase related, but not product specific data about consumers. For this reason it should be basically appropriate to identify long term trends in consumer markets that fulfil the criteria stated above: that they can be correlated to the attractiveness of a company’s present and future offerings, in this case functional food products.

2.1.1

Lifestyle Approaches in Marketing Research

In surveying consumers, lifestyle instruments have been used to detect major trends in order to analyse differences and similarities across markets in search of global or at least cross-European segments. The lifestyle concept was introduced to consumer research in 1963 by Lazer (Lazer, 1964). The concept became popular quickly and was regularly characterised as an example for psychographic segmentation variables (Wells, 1975), which meant to say that the variable is subjective (in contrast to objective segmentation criteria like demographics or usage rate), and that it was not product specific (unlike product attitudes or preferences), but rather a general characteristic of the consumer.

The most widely used approaches to lifestyle measurements were developed in the seventies: The AA and AIO approaches.

Hustad and Pessemier (1972, 1974) developed AA which stands for activities and attitudes. Activities can be defined as visible actions, while attitudes can be defined as learned, enduring beliefs to react positively or negatively to an object or group of objects.

The AIO approach is the most well known one. Wells and Tigert (1971) proposed it originally in cooperation with the Leo Burnett advertising agency. AIO stands for activities, interests, and opinions, which are complemented by demographics, considered as the most important element of lifestyles.

Vrije Universiteit Amsterdam

17

2006

As well AA as AIO followed a pattern which is still typical of lifestyle research today, including the international instruments like CCA, RISC, or VALS. They are based on a large set of AIO items that are reduced analytically to a few, usually only two, dimensions. The resulting space, sometimes called an attitude map or a value map, is then used to classify consumers on the remaining dimensions, which leads to lifestyle segments.

2.1.2

Criticism on Lifestyle Research

The lifestyle research instruments mentioned in the previous sub-paragraph have been criticised by academic marketing scholars on several grounds. Besides that there is no agreement on what lifestyle actually means (Anderson & Golden, 1984), there are two points of criticism which led to the development of the food-related lifestyle approach: 1) The instrument is not guided by theory - According to Grunert and Bronsø (1996) lifestyle types are based on dimensions derived by exploratory data analysis techniques like factor analysis or correlation analysis. These techniques are related to sets of items, the creation of which is not theoretically guided either, but is very much based on common sense reasoning. Grunert and Brunsø (1996) noted that the lifestyle concept actually has roots in the social sciences, which can give inspiration for theoretical foundation. 2) The cross cultural validity of the international lifestyle instruments remains to be demonstrated – The cross-European studies like RISC and CCA offer data which try to identify similar lifestyle segments across borders, as well as Douglas & Urban (1977), Linton & Broadbent (1975) have tried to identify cultural differences in lifestyle. Gathering data from different cultures with the goal to obtain comparative results requires a measurement instrument is cross-cultural valid, i.e. that translation and measurement equivalence are guaranteed or at least tested. According to Grunert and Brunsø (1996) there are reasons to believe that such equivalence cannot be taken for granted in many studies, and this would apply all the more in such a culture-dependent area as food intake.

Vrije Universiteit Amsterdam

18

2006

Because of the points of criticism on existing lifestyle research instruments, which are evaluated above, Grunert and Brunsø (1996) came up with a new definition for lifestyle and developed an outline of a model for lifestyle segments in the context of food and drink. It is for these same reasons that the food-related lifestyle approach is used in this study.

2.1.3

Food Related Lifestyle Aspects

While the concept of lifestyle research was initially introduced to consumer research by Lazer (1963), the food-related lifestyle research has been proposed by the Centre for research on customer relations in the food sector (MAPP Institute in Denmark) in the early 1990s. The food related lifestyle instrument (FRL) initiated by Brunsø and Grunert is a 69-item questionnaire, and attempts to explain behaviour toward food purchase through examining the food related lifestyle of an individual by looking at the importance of five interrelated aspects: (1) ways of shopping, (2) quality aspects for evaluating food products, (3) meal preparation methods, (4) consumption situations and (5) purchase motivations. More conventional questionnaire methods have been developed for the investigation of eating disorders, including the restraint scale (Polivy, Herman & Warsh, 1978), the Three Factor Eating Questionnaire (Stunkard & Messick, 1985) and the Dutch Eating Behaviour Questionnaire (Van Strein et al., 1986). These measures are more concerned with cognitive constraints on food intake and the situations surrounding excessive consumption than with the aspects influencing food choice covered by the FRL instrument, which is evaluated on in the next part. (1) Ways of shopping seeks to determine how people shop for food products, by studying whether their decision-making is characterized by impulse buying, or by extensive deliberation. Do they read labels and other product information, or do they rely on the advice of experts, like friends or sales personnel? In which shops—one-stop shopping versus specialty food shops? It also tries to find out if the consumer is an organised shopper, i.e. if he/she makes shopping lists. (2) Quality aspects refer not to concrete attributes of products, but relate to product's benefits such as health, taste, organic, quality and freshness. It also included novelty, which tries to establish whether the consumer enjoys trying new foods.

Vrije Universiteit Amsterdam

19

2006

(3) Meal preparation methods relate to the time taken to prepare a meal; perhaps the consumer is more convenience oriented and does not have any interest in cooking. Is preparation characterized by efficiency, or by indulgence? To which extent is it planned or spontaneous? Meal preparation methods also determine if cooking is a social event involving all the family or if cooking is solely the woman's responsibility. (4) Purchasing motives tries to discover what is expected from cooking and eating. Certain consumers do not like changing and feel secure in eating foods which are familiar to them. Other consumers derive self-fulfilment from cooking and being praised for their exploits. Purchasing motives also tries to discover whether consumers believe that sharing a meal aids them in developing social relationships. How important are social aspects, hedonism, tradition and security? (5) Consumption situations refer to whether the consumers have set meals or if they eat a number of small snacks during the day instead. This element refers to consumers' social eating habits, such as if sharing a meal is an important part of their social lives. How are meals spread over the day? How important is eating-out?

By looking at the importance of those five aspects the FRL model attempts to explain behaviour toward food purchase behaviour. The FRL initial analysis with regard to cross-cultural validity as well as its intra-cultural validity has proved stable over time, and the instrument has been successfully applied to various European and non-European food cultures (Askegaard and Brunsø, 1999; Bredahl and Grunert, 1997; Brunsø et al, 1996;; Brunsø and Grunert, 1995; Scholderer et al., 2002).

Since the food-related lifestyle determinants have been identified as important when considering food purchasing motives in general it is very likely that they will also relate to purchase behaviour of functional food. Therefore, food-related lifestyle issues as well as specific functional food related lifestyle questions should be incorporated in order to explain the consumption of functional food.

Vrije Universiteit Amsterdam

20

2006

2.1.4

Food Related Lifestyle Segments

Research has shown that lifestyle can be divided into several differentiated groups with help of wellknown lifestyle instruments like RISC, CCA and VAlS. Nevertheless, the food-related lifestyle instrument will be used here instead, because Van Raaij and Verhallen (1994) justified the use of domain specific segmentation techniques. They state that, staying as close as possible to a behavioural domain ensures that strategic marketing decisions are based on specific market knowledge. In this subparagraph the characteristics of food related lifestyle segments will be evaluated and related with the typification of functional food consumers revealed in previous research, for each segment a hypothesis will be formulated.

The FRL instrument will be used as a tool to measure attitudes to food, i.e. how people link food to the attainment of life values, and to compare these between (non)functional food consumers. Therefore, the original 69-item questionnaire (seven-point scales, from ‘totally disagree’ to totally agree’), covering five interrelated aspects will be used, to segment citizens to different lifestyle dimensions. Specific questions concerning the consumption of functional food will be added in order to eventually be able to answer the problem statement.

Previous research accomplished with the food related lifestyle instrument in European countries, have resulted in the identification of five basic cross-cultural food consumer segments. These segments count the uninvolved food consumer, the careless food consumer, the conservative food consumer, the rational food consumer and the adventurous food consumer. These segments have also been found to be stable over time, and may thus be used for detecting trends over time (Reid, et al., 2005). These food related lifestyle segments will be described beneath. Characteristics of the different segments are compared with features of functional food consumers which resulted in five hypotheses, this is the first step in answering sub-question one. The information about the food-related lifestyle segments comes from the study by Grunert & Brunsø (1996). The hypothesis will be formulated in ‘more’ or ‘less’ inclination towards functional food consumption, which refers to the mean in the developed scale.

Vrije Universiteit Amsterdam

21

2006

Segment 1 – The careless food consumer – This type of food consumers have below average scores on all dimensions regarding shopping, and score lowest of all on using shopping lists, ie, they shop spontaneously. They are considerably less interested than other consumers in all quality aspects. They are especially uninterested in health and ecological products. Their interest in cooking is at the level of the population average. Meal preparation should preferably be fast and easy for these consumers, and is again characterised by a higher degree of spontaneity. The scores on all purchase motives are low, and they are especially low on creating security by traditional eating patterns. No purchasing motive is particularly important for this group; they score lowest of all on security by traditional eating patterns. This segment is thus generally uninterested in shopping for and preparing of meals. The only aspect which raises some interest for them is new products, if they do not require effort in the kitchen. As price is of lesser importance than in any other segment, and given the fact that functional food products have a premium price, one could say that members of this segment are more inclined to consume functional food. Second argument that this segment is more inclined to consume functional food, is the fact that this group is only influenced by the product characteristic ‘novelty’ and functional food products are defined as innovative products claiming beneficial psychological effects beyond those ordinarily associated with typical nutrients (Jong et al., 2003). Thirdly, members of this segment do not put much effort in preparing their food, this timesaving aspect can be an argument for consuming functional food.

Based on the three arguments above, the following hypothesis is

formulated in regard to this segmentation and functional food consumption:

H1: Members of the careless food consumer segment are more inclined to consume functional food.

Segment 2 – The uninvolved food consumer – These consumers are the ones least interested in specialty shops, and are relatively uninterested in product information. They are quite uninterested in most aspects of food with the possible exception of price and the social aspect. They are really uninvolved in food and meals, and this pervades all aspects of their food-related lifestyle. Their foodrelated lifestyle is characterised by the fact that they do not use food to achieve basic values; they

Vrije Universiteit Amsterdam

22

2006

achieve them through other channels. Consumers belonging to this segment are far less interested in products characteristics as healthiness and ecology/naturalness. They are least interested in price and because functional food is generally more expensive and like Hilliam (1996) posited that consumers of functional food in Europe reflect a higher willingness or ability to pay a price premium it can be expected for those reasons that uninvolved food consumers are positively related to the consumption of functional food. Because consumers belonging to this food-related lifestyle segment are the least interested in price, and novelty is the only aspect which they are interested in, therefore the following hypotheses is formulated:

H2: Members of the uninvolved food consumer segment are more inclined to consume functional food.

Segment 3 – The rational food consumer – Consumers which belong to this segment are very interested in food, and plan their shopping. They are also very interested in product information, and use specialty shops more than the others. They are very price conscious, and shopping lists are more often used in this segment than in the others. Healthiness and ecology/naturalness in particular are highly valued in this segment. Rational food consumers believe in one’s own impact on personal health (Hilliam, 1996), believe in food-disease prevention concept (Wrick, 1996), believe in the disease-preventative nature of natural food (Childs & Poryzees, 1997), and opinions of the relationship between food and health. The novelty aspect is not as important for the rational food consumer as for the careless and adventurous food consumers. However, rational food consumers do like innovation and challenging cooking, e.g., experimenting with new recipes. Most of these consumers still consider the entire cooking process to be the women’s job. Food and food products are an important part of these consumers’ lives, and are essential for achieving such basic values as self-fulfilment, recognition, social togetherness and security. This group of consumers is interested in food and cooking, because it is a part of tradition and can therefore help create a feeling of security in their lives. This leads to an interesting-critical shopping behaviour, which is characterised by a strong interest in product information, price, and major quality aspects such as healthiness and

Vrije Universiteit Amsterdam

23

2006

ecology/naturalness. Based on the fact that they have a particular interest in ecological natural products and novelty is not that important to them, the following hypothesis is formulated.

H3: Members of the rational food consumer segment are less inclined to consume functional food.

Besides formulating a hypothesis, the title of this segment (rational food consumer) is disputable after identification. Given the characteristics of this segment, e.g., strong belief in the preventative nature of natural food and the perception of the cooking process to be the women’s job, it is maybe better to change the name to ‘the Traditional food consumer’ segment.

Segment 4 – the conservative food consumer – Consumers in this segment are rather involved in shopping. Among all segments, they are the ones most positive with regard to advertising and the most price-conscious. They study product information, and they use shopping lists. They also use specialty shops more than the average consumer. These consumers are not at all keen on new products, and they are especially interested in taste, moreover food must be healthy and fresh. Food is an important part of the conservative food consumers’ lives. Food and drink products create stability and security in their lives. This is reflected in the careful planning of both shopping and cooking, and an aversion to anything new. Functional food is generally more expensive and like Hilliam (1996) stated that consumers of functional food in Europe reflect a higher willingness or ability to pay a price premium, and because functional food products are innovative products claiming beneficial psychological effects beyond those ordinarily associated with typical nutrients (Jong et al., 2003) the following hypothesis is formulated.

H4: Members of the conservative food consumer segment are less inclined to consume functional food.

Vrije Universiteit Amsterdam

24

2006

Segment 5 – the adventurous food consumer – These consumers are the ones most interested in product information and this is the segment using speciality shops the most. They use shopping lists more often and enjoy shopping, but their attitude towards advertising is about average. They do to some extent check prices when shopping. These consumers are very interested in quality aspects. Among all segments they are the ones most interested in health, new products, fresh, natural and ecological products, whereas taste has a lower score. Unanimously positive correlations between those beliefs and acceptance or purchasing interest for functional food have been reported. Growing health awareness, and anxiety about physical appearance drive demand for functional food (Reuters, 2000). The price-quality relation also has an above average importance. They like cooking more than all other segments and reject the idea that cooking should be quick and easy. They score highest in experimenting with new recipes. They also score highest in cooking as a family affair, and correspondingly lowest in cooking as a woman’s task. They plan their cooking activities more than the population average. They also have the highest average in dining out and eating with friends, and less tendency to snack. These consumers score high on self-fulfilment and recognition by cooking, and also the social motive is of importance. They are not interested in security by traditional ways of cooking. These consumers are quite involved with food. They are interested in quality and new products, and like speciality shops where they can get advice. Given the indications that functional food products are likely to be used by individuals who do not fit in the initial target group: individuals who already have a healthy lifestyle are more likely to buy functional food (Jong et al., 2003), this probably applies for the adventurous food consumer segment. Cooking is characterised by a conscious attempt to experiment and find new ways. Given the characteristics of this consumer segment, i.e. interested in product information and new products, which correlates with characteristics of functional food consumers (Boque & Ryan, 2000), the following hypothesis is formulated.

H5: Members of the adventurous food consumer segment are more inclined to consume functional food.

Vrije Universiteit Amsterdam

25

2006

In this paragraph a general introduction is given about the basics of lifestyle measurement instruments. After evaluating the existing criticism on well-known lifestyle instruments like RISC, CCA and VAlS, the food related lifestyle approach is introduced. Because the latter is of specific interest for realizing this particular study, there is elaborated extensively on this research method. Justification for choosing this research instrument is given at the start of this paragraph. The five different segments covered by the 69-item questionnaire are explored, and the characteristics of the different consumer segments are linked with the characteristics of functional food consumers identified in previous research, and based on that hypotheses are formulated. By doing this the first sub-question is partially answered. Investigating respondents on their food-related lifestyle characteristics and functional food consumption behaviour will give the opportunity to reject or confirm these 5 hypotheses, and is the second step in answering the first sub-question.

2.2

Culture & Food Related Lifestyle

In this paragraph the second sub-question will be discussed. By evaluating and summarizing previous research about cultural influences on food habits a first step is taken to answer the second sub-question of this study. Particular studies exploring food habits of Turkish immigrants in the Netherlands give insight in the relationship between cultural influence and food consumption behaviour.

Culture is a mayor determinant of what we eat and food habits tend to be long lasting and resistant to change. Culturally determined values, attitudes, and beliefs form the framework within food habits develop. Cultural patterns affect food choices, food preparation and food consumption (Harper, 1983). The type of food eaten and the nutritional status are strongly influenced by the food habits and eating patterns. Influence of food habits is through family tradition, which is affected by country of origin, by ethnicity and religious or traditional beliefs. Cultural factors, which distinguish food selection from food avoidance, are not rational but stem from beliefs passed on from one generation to another. Cultural issues include religious beliefs involving food, food preference and food restriction. Food habits persistence can clearly be seen in immigrants trying to adapt to the way of life of their adopted nation. Food habits are transmitted tot the second and third generations of immigrants in a

Vrije Universiteit Amsterdam

26

2006

new country, and gradually modified and weakened. Most studies following dietary habits of those living in other countries have shown that some changes occurred by substitution, addition or modification (Pan et al., 1999).

Based on a review of studies regarding the phenomenon of immigrant cuisines, Mennell et al. (1992) noticed that immigrants and ethnic minorities tried to maintain their own cooking and eating habits as long as possible. One of the factors which relate to health is food, eating habits are primarily determined by cultural and social-economic background. Migration in general leads to a change in the cultural and social environment and from that health of migrants will be influenced (Meulmeester, 1988). Researches from inland and abroad revealed that migrants often adopt unfavourable eating habits, which can result in diseases as a result (Offerman et al., 1978). Previous research pointed out that adjustments in eating habits of migrants mainly rest in the use of ordinary Dutch products prepared in a traditional manner (Meulmeester, 1988). Research carried out among Turkish families showed that they fall short in vitamin-D intake, and tooth decay is higher (Buys et al., 1979).

Beardsworth and Kiel (1997) pointed out that particular foods and food combinations; in certain cultures can be associated with festivity and celebration, with religious observance and ritual, and with the rites of passage which mark essential status transitions in the life cycle. Food items may also develop associations with health, moral righteousness and spiritual purity. The reverse can also be found in specific cultural and historical contexts, when particular food items may carry negative meanings associated with the dangers of diseases, immorality or ritual pollution (Beardsworth and Kiel, 1997). These differentiations create a kind of sensitivity to what might be considered “good” or “bad” by persons from varying cultures. In general these attitudes regarding specific food and drink products do not change among migrants (Knegt et al., 1982).

2.2.1

Turkish Food Consumers

This part of the paper investigates the consumption of Turkish citizens in the Netherlands with specific attention to their distinctive functional food preferences i.e. consumption, coupled with a tradition of

Vrije Universiteit Amsterdam

27

2006

food symbols which link Turkish subcultures. To identify citizens with a Turkish background, different definitions are being used and a correct description of the various population groups is subject of discussion among researchers concerned with migrants in the Netherlands (Bruynzeels, et al.,1999). First generation migrants are persons born outside the Netherlands with at least one parent also born outside the Netherlands. Second generation migrants are , according to the narrow definition, persons born in the Netherlands with both parents born outside the Netherlands. According to the broad definition, second generation migrants are persons born in the Netherlands with at least one parent born outside the Netherlands (Brussard, et al., 2001). To carry out this research Turkish migrants from the first and second generation were investigated. The latter definition for defining the second generation will be used in this research for Turkish citizens in the Netherlands.

Culture is a major determinant in human food choice. Traditions, beliefs and values are amongst the main factors of food consumption within a cultural group, affecting nutrition intake, preference, and mode of food preparation and serving (Rozin, 1990). Markets are becoming more and more international, and the drive of industry to initiate and expand food exports to consumers with different food preference patterns has created a increasing interest in cross-cultural research on food perception and preference, with an aim of enhancing understanding of the specific individual cultures (Million et al. 2002).

Previous cross-cultural studies focussing on the perception of tastes have found no differences between cultures. Druz & Baldwin recorded no difference in the four basic taste thresholds between Nigerian, Korean and American subjects; American and Japanese thresholds for sweetness and saltiness did not significantly differ in their study. Perception of small differences in taste intensity (Laing, 1993) and intensity ratings for various basic tastes levels in foods or drinks showed no differences in discrimination either between Caucasian Australians and Japanese (Prescott & Bell, 1993), or between Caucasian Australians and Malaysians (Holt et al., 2000). Despite the lack of research studying cultural differences in perception of sensory properties other than taste (odour, flavour, texture), taste-experiencing abilities appear generally similar between cultures.

Vrije Universiteit Amsterdam

28

2006

Perceptions and attitudes, which are strongly founded in cultural values, are hard to change by information means such as health claims (Frewer et al., 2004), i.e. the effectiveness of health claims depends among other things on the strength of the association between consumers’ values and their attitudes to functional food. The harmony and mastery values (Schwartz, 1994) are examples of cultural dimensions that may influence consumers’ attitudes to functional food. Mastery means active and self-assertive manipulation of the social and natural environment, where harmony gives emphasis to co-existence with nature and rejects the manipulation of natural sources. Even though other consumer values may also be associated with functional enrichment, research has shown that ‘mastery’ and ‘harmony’ influence consumers’ attitudes to novel foods as well as the dominance of the health aspect in their food related life-style (Brunsø et al., 2004).

Thus besides taste, other factors determine cultural differences in food and drink consumption patterns. In many Turkish families for instance the domestic tasks are still traditionally divided nowadays, so the women are assigned with these tasks primarily (Brussaard, 2001). At the same time preparing/cooking food is still more a social matter in Turkish families. Therefore with regard to food it can be presumed that citizens with a Turkish background lead a more conservative food related lifestyle.

The study accomplished by Meulmeester in 1988 concerning the food consumption among Dutch, Turkish and Moroccan children in the Netherlands, revealed that 20% of Turkish children skip breakfast in comparison with 10% of Dutch children. For lunch one third of the Turkish children have a warm meal in comparison with hardly any of the Dutch children. Most Turkish children have bread specially added at diner. At breakfast, lunch, diner, and in between meals most of the Turkish children drink tea or water in comparison with milk and soft drinks consumed mostly by Dutch children. As a result Turkish children and presumably as well Turkish adults’ calcium intake is lower than that of Dutch citizens and 20% lower than recommended. An attractive alternative for most Turkish consumers instead of milk is yoghurt, because in regards to taste it is acceptable for them (Meulmeester, 1988). As snack Turkish children primarily consume fresh fruit and nuts or chips while

Vrije Universiteit Amsterdam

29

2006

Dutch children mostly eat biscuits or something sweet. The study of Meulmeester showed as well that Turkish children in general consume more fresh fruit and vegetables than Dutch children. As regards to the consumption of fat/lipids, Turkish children consume more ‘good’ fats than Dutch children, but nevertheless the cholesterol level of Turkish boys is somewhat higher because they consume more eggs and liver. As well Dutch as Turkish children do not consume enough fibres. The diet of citizens with a Turkish background shows both positive and negative differences with the general Dutch diet. Surprisingly the risk of overweight is among both children and adult women with a Turkish background higher than for Dutch citizens (Brussaard, 2001). Because of the existing differences in eating habits concerning preparation as well as kind of nutrition intake two hypotheses as well as one explorative research question is formulated.

Taking the aspects, given in the previous part, in mind the subsequent hypothesis is formulated:

H6: Citizens with a Turkish background are more likely to be classified as being part of the rational/traditional food consumer segment than citizens with a Dutch background.

This hypothesis is based on the two characteristics of this rational/traditional consumer segment; (1) consumers in this segment consider the cooking process to be a women’s job – which is the case in most Turkish families and (2) food and food products are an important part of these consumers’ lives and essential for achieving basic values, such as social togetherness – which is true for most Turkish families as well.

The second hypothesis concerning segmenting citizens with a Turkish background is formulated in the following manner.

H7: Citizens with a Turkish background are less likely to be classified as being part of the careless food consumer segment than citizens with a Dutch background.

Vrije Universiteit Amsterdam

30

2006

This seventh hypothesis is formulated as characteristics belonging to this particular segment do not correspond with the characteristics of Turkish food consumers, for example members of this segment score especially low on creating security by traditional eating patterns and do not put any effort in the kitchen. These features are exactly opposite to the characteristics of a Turkish food consumer and therefore it is hypothesized that the only a small amount of Turkish people belong to this segment.

2.2.2

Turkish & Functional Food

A hypothesis concerning the consumption of functional food cannot yet be formulated in any particular direction without speculation. Although it has been proved in previous research (Brussaard, 2001; Buys, 1979; Meulmeester, 1988) that citizens with a Turkish background have a healthier diet and that they consume more fruit and vegetables than citizens with a Dutch background it can not be hypothesised whether their need for products with added vitamins is likely to be lower because their diet already contains these vitamins. Cornelisse (2003) revealed in her study that Turkish respondents show the lowest prevalence for intake of extra vitamins/mineral, despite the fact that immigrants feel less healthy than the Dutch and report more chronic health problems. On the other hand citizens with a Turkish background do not consume enough milk to reach their recommended daily calcium intake and for that reason people could argue that they have an interest in consuming products with added calcium, for instance in yoghurt products. Even though there is information about the differences of nutrition intake between citizens with a Turkish and Dutch background there is no information whatsoever about the consumption of functional food, and therefore the following exploratory question evolved:

•

What is the difference in the consumption of functional food between citizens with a Dutch and a Turkish background?

By evaluating and combining previous research literature about influences on food habits in general and particularly in respect to a Turkish background, the second and third sub-question is discussed. It is expected that cultural background influence the segmentation of citizens into a particular food

Vrije Universiteit Amsterdam

31

2006

related lifestyle. However the exact nature of this relationship is not yet clarified. Therefore an exploratory research question is formulated. To investigate whether differences in the consumption of functional food exist among citizens with a different cultural background specific questions are included in the questionnaire.

Vrije Universiteit Amsterdam

32

2006

2.3

Conceptual Framework

To visualize and integrate all theories and facts, discussed in the previous sections of this thesis, the following framework is developed.

Food Related Lifestyle H6, 7

Cultural

H1,2,3,4,5 ERQ

Consumption of

Background

Functional Foods

H = Hypothesis ERQ = Explorative Research Question

The framework shows which relationships are expected to exist between food related lifestyle (extensively discussed in paragraph 2.1), consumption of functional food (hypotheses 1-5) and cultural background (hypotheses 6 & 7). Citizens with either a Dutch or Turkish background will be first segmented, with help of the FRL instrument, into one of the five food related lifestyle segments, and are hypothesized to be divided unequally over the 5 segments, which is formulated in hypothesis six and seven. An exploratory research question is formulated for exploring the relation between cultural background and the consumption of functional food, because there is not enough scientific/academic evidence to formulate a hypothesis.

Summarizing the study is to investigate whether there exists a significant relationship between persons belonging to a particular food-related lifestyle segment and/ or persons belonging to a particular cultural background (Turkish and Dutch) and the consumption of functional food. With the conceptual framework this chapter is finished, in the next chapter will be evaluated on the methods and means of

Vrije Universiteit Amsterdam

33

2006

data collection to accomplish this study. This chapter was concluded with a conceptual model illustrating the expected relationship between persons belonging to a particular food-related lifestyle segment and/ or persons belonging to a particular cultural background (Turkish and Dutch) and the consumption of functional food. In the next chapter methods and means of data collection will be discussed.

Vrije Universiteit Amsterdam

34

2006

3

Method and Means of Data Collection

The purpose of this chapter is to explain and justify the methods and processes which were used to carry out this research. This chapter contains a description of the procedures how the hypotheses were tested. The implementation will be discussed in chapter 4. It contains a description of the choices made concerning the research methods, and the arguments for these. It also contains a presentation of the discussion concerning which existing questionnaire elements will be used and why certain questions are added.

3.1

Sample

To give a reliable answer on the research question it is of importance to collect sufficient information, within a large enough sample. A minimum of 50 citizens with a Turkish background and 50 citizens with a Dutch background took part in this study to test the conceptual model. The study has been conducted among students with a Turkish and Dutch background in the Netherlands. The argumentation for choosing these two particular groups is given in chapter 1.

An analysis of the consumption of functional foods is carried out in the period between May 5 and 20, 2006, on a sample of citizens with a Dutch and Turkish background. The sample size has taken the objectives of the research and the resources available into consideration. Students of the VU have been asked to fill-in the questionnaire via email. Based on the demographical questions (i.e., place of birth of the respondent and of their parents), respondents were grouped in respondents having a Turkish or Dutch background. A respondent born in Turkey or born in the Netherlands with at least one parent born in Turkey is placed in the group ‘having a Turkish background’. Results could only be based on at least 50 questionnaires filled in by respondents with a Dutch background and 50 respondents with a Turkish background. After emailing students form the Vrije Universiteit, there were initially not enough respondents with a Turkish background. The next step in reaching students with a Turkish background was sending an e-mail at a Turkish student association with the request to forward the

Vrije Universiteit Amsterdam

35

2006

questionnaire to their members. As a result the aim of surveying 50 respondents with a Turkish background was reached.

3.2

Questionnaire

To evaluate the construct ‘food-related lifestyle’ the 69 questions of the cross-cultural valid foodrelated lifestyle instrument (Grunert & Brunsø, 1993), were used. The instrument of Brunsø & Grunert (1993) has been applied to studies in Denmark, France, Great Britain and Germany. The instrument covers the main aspects of buying, preparing and consuming food. All questions are rated on a seven-point Likert scale. The questionnaire consists of 23 dimensions with three items each (23 x 3 = 69). All 23 dimensions are tested on scale validity and scale reliability before the 69 items are converted into sum scores for the 23 dimensions. The sum scores will then be used as input to a clustering procedure based on K-Means cluster method. The selection of the cluster solution will be based on criteria derived from previous research as well as interpretability. The interpretation of the clusters rests mainly on how the pattern of cluster means corresponds to that found by Grunert & Brunsø (1993). They revealed the following segments: 1. The Careless food consumer 2. The Uninvolved food consumer 3. The Rational food consumer 4. The Conservative food consumer 5. The Adventurous food consumer

The questionnaire is extended with questions concerning the consumption of functional food, in order to be able to investigate the relationship between segments, cultural background and functional food consumption. Respondents were asked to define their functional food consumption by indicating how many of the 15 selected functional food products they consume. In addition to these, several questions were related to socio-demographics and to the economic status of the respondent’s household. The questionnaire is shown in appendix 1, it is translated into Dutch before it was used.

Vrije Universiteit Amsterdam

36

2006

3.3

Data Analysis

After coding the questionnaire, data was analysed using Statistical Package for Social Sciences (SPSS). Descriptive analysis, for example mean and standard deviation, will be carried out to determine the respondents’ segment membership and consumption of functional food.

3.3.1

Testing Hypothesis 1 & 2 & 3 & 4 & 5

After segmenting respondents in one of the five food related lifestyle segments, the study further investigated how the members of a specific food-related lifestyle segment relate to the consumption of functional food. Analysis of variance will be used to check whether significant differences exist between the five different segments and the amount of functional food products consumed. The first five hypotheses were tested after segmenting respondents into one of the five segments, and measuring if the mean scores of functional food consumption differed significantly between segments. The next section describes how the first five hypothesis will be tested. First of all an appropriate test must be selected to examine the validity of the hypotheses. This depends on the scale of the variables and whether the sample has a normal distribution or not. The Kolmogorov-Smirnov test will be used to test for normality. The method used for the analysis of variance depends on whether the variables are normally distributed or not. In the case of normal distribution the mean scores will be analysed with Anova, when the variables are not normally distributed the mean scores will be analysed with the Kruskal-Wallis test. When the significance level of the test outcome is above .05, it can be said that the consumption of functional food does not significantly differ across segments.

When the outcome of the ANOVA or Kruskal-Wallis test is below .05, or one segment scores obviously higher or lower on functional food consumption than the others (despite of a test outcome above .05), the Post Hoc Test can clarify the difference in functional food consumption for each segment and another segment separately. With the Post Hoc Test the mean scores of functional food consumption for each segment is compared with the mean scores of the other segments. Again when

Vrije Universiteit Amsterdam

37

2006

the results of the Post Hoc Test are above .05, it can be concluded that there is no significant difference between membership of a particular segment and the consumption of functional food.

So the hypotheses one till five will be tested with ANOVA or the Kruskal-Wallis test and for a further explanation of the difference between segments the Post Hoc Test will be carried out. Hypothesis one till five will be rejected when the outcomes of these tests are above .05., because this means there is no significant difference between the consumption of functional food and membership of a particular food-related lifestyle segment.

When the outcome of the ANOVA and Kruskal-Wallis test is below .05 the Post Hoc Test can specify the exact segment which differs significantly from the others. For example, when the ANOVA or Kruskal-Wallis test results are above .05, hypothesis 1 will be rejected. H1: Members of the careless food consumer segment are more inclined to consume functional food. But when the outcome of the Post Hoc Test points out that the careless food consumer consumes significant more functional food than for instance the conservative food consumer p< .05, the hypothesis will be accepted. The four other hypotheses will be dealt with in the same manner.

Food Related Lifestyle H 1,2,3,4,5

Consumption of Functional Foods Figure 3.1: Hypotheses related to FRL and the consumption of functional food.

Vrije Universiteit Amsterdam

38

2006

By investigating the first five hypotheses it will become clear whether membership of a particular food related lifestyle segment shows a relationship with the consumption of functional food.

So far, the relationship between memberships of a particular food related lifestyle segment and the consumption of functional food. The next step in answering the problem statement is investigating the relationship between cultural background and membership in a particular food related lifestyle segment. How this will be done is described in the next sub-paragraph.

3.3.2

Testing Hypotheses 6 & 7

It is of specific interest how citizens with a Dutch or Turkish background are divided among the different segments.

Food Related Lifestyle H 6, 7 Cultural Background

Figure 3.2: Hypotheses related to FRL and cultural background

Measuring the difference in cultural background of the members of a particular food-related lifestyle segment, is relevant for hypothesis six and seven. Hypothesis 6 and 7 will be answered using the Ttest in case of normal distribution or otherwise with the Binomial test.

Hypothesis 6 concerns:

H6: Citizens with a Turkish background are more likely to be classified as being part of the rational/traditional food consumer segment than citizens with a Dutch background.

Vrije Universiteit Amsterdam

39

2006

and hypothesis 7 concerns:

H7: Citizens with a Turkish background are less likely to be classified as being part of the careless food consumer segment than citizens with a Dutch background.

The number of members of a particular segment with a Turkish background can simply be revealed by splitting the data file according to background in SPSS and than count the segment members. Whether there is a significant difference between the numbers of members with a Dutch or Turkish background within a particular segment will then be investigated with help of a statistical test. The Chi-Square test, will make clear whether the Turkish and Dutch respondents are equally divided among the five segments.

When the members of the five different segments are normally distributed, the T-test will be used to test hypothesis six and seven, if not the Binomial test will be carried out. The hypotheses will be rejected when the outcomes of the tests are above .05, which means that there is no reason to assume cultural background influences membership into a particular segment.

By rejecting or accepting hypotheses six and seven it will come clear whether cultural background is a predictor of membership of a particular five food-related lifestyle segment. By doing this the relationship between cultural background and food-related lifestyle segments can be determined, which is the second step of the research design.

3.3.3

Investigating the Exploratory Research Question

Thirdly, the exploratory research question will be addressed, which aims to explain the potential relationship between the construct ‘cultural background’ and the construct ‘consumption of functional food’.

Vrije Universiteit Amsterdam

40

2006

ERQ Cultural

Consumption of

Background

Functional Food

Figure 3.3: Exploratory research question related to cultural background and the consumption of functional food

The study will test whether there is a significant difference between the mean scores of functional food consumption between respondents with a Turkish and Dutch background. In case of normal distribution this will be done with the T-test and otherwise with the Mann-Whitney test. When the outcome of the test is below .05, it can be concluded that the consumption of functional food differs among respondents with a different cultural background.

In addition, the potential relationship between cultural background and the consumption of functional food will be investigated using a regression analysis. The purpose of this analysis is to assess the relative importance of cultural background as a predictor of the consumption of functional food. Regression analysis is a powerful and flexible procedure for analysing relationships between a metric and one or more independent variables (Malhotra, 1999). Only when the coefficients of the regression model have a significance level below .05 regression is supposed to be linear.

3.3.4

Partial Correlations

The last step is to investigate whether the relationships influence each other, therefore tests for revealing partial correlations will be carried out. The relationship between (the construct) lifestyle and (the construct) functional food consumption may be moderated by cultural background. Next, a test will be conducted to examine whether membership of a particular segment is a moderator for the relationship between cultural background and the consumption of functional food.

Through evaluating the outcomes of the research it will be possible to determine the existence of significant relationships between the three constructs (within the limits of this research design). This

Vrije Universiteit Amsterdam

41

2006

study provides an insight into the lifestyles of food consumers in the Netherlands and tries to identify the drivers of the demand for functional food. The outline of the questionnaire used to conduct this research is presented next. In chapter 4 the results of the analyses will be presented and in chapter 5 these results will be discussed. 3.3.5

Outline of Questionnaire

Constructs

Food-Related Lifestyle

Functional Food Consumption

Data

69 Statements included in the five domains: 1. Ways of Shopping 2. Quality Aspects 3. Cooking Methods 4. Consumption Situations 5. Purchasing Motives

Question 70-75 Frequency of purchase of functional food products

Background

Question 76-80 Having a Turkish or native Dutch background.

SocioDemographic Data

Question 81-86 Age, gender, income, , education status, household size, occupation status

Cultural

Analyses

K-Means Cluster Analysis

H 1,2,3,4,5 Kruskal-Wallis or Anova Post Hoc Test H6&7 Chi-Square Binomial Test or T-Test

ERQ Mann-Whitney Or T-Test Partial correlations

Figure 3.4: Outline of the questionnaire

Vrije Universiteit Amsterdam

42

2006

4

Analysis of the Collected Data

In this chapter the analysis of empirical results will be presented. First of all some preanalyses are given to describe the sample distribution. Next the demographics are given, which give an idea about the respondents used in this study. The analysis of the different foodrelated lifestyle segments is the second step in this chapter. The analysis of the consumption of functional food is the third step taken. Besides this there will be analysed how the different segments correlate to the consumption of functional food. At last there will be looked after the partial correlations of demographics..

4.1

Pre-Analysis

Normal Distribution To begin with, several variables were tested whether they have a normal distribution. This is done with the Kolmogorov-Smirnov test. When the significance level is below .05 it can be said that the variable is not normally distributed. It is important to know whether a variable is normally distributed or not because this determines if a parametric or non-parametric test should be used to test the hypotheses. In this study the tested variables do not have a normal distribution, therefore the hypotheses are confirmed or rejected based on non-parametric tests. In appendix 2 the results of the KolmogorovSmirnov test are shown. The skewness and kurtosis are given as well to give more information about the shape of the distribution. Neither values are extreme which means that the distribution of the variables is close to normal distribution.

Outliers The second pre-analysis is testing whether there are any extreme values among the respondents with regard to age, education, gender, segments or amount of functional food consumed. Bar charts and frequency tables show that the data does not consist any outliers. This means that the data does not consist any extreme values which might have influenced the mean scores of any variable in a particular direction. In appendix 3 these figures are shown. Vrije Universiteit Amsterdam

43

2006

4.1.1

Socio-Demographics

In the table below the socio-demographic characteristics of the respondents are given. It is shown that there are slightly more women than men in the sample and that most respondents are higher educated. The first three income levels comprise almost 80 percent of the respondents and 40 percent of them live in a household which consists of more than 4 persons.

Gender Background Education Man 43 Turkish 50 MAVO Woman 57 Dutch 50 HAVO/VWO MBO HBO/WO Total

100 Total

100 Total

1 1 1 93 100

Income >1000 1000- 2000 2000-3000 3000-4000