To Market Food Trends for the Greater Good:

(1)

To Market Food Trends for the Greater Good:

Using Food Trends to Nudge Consumers towards the Right Choice

Bas Poorta

S2356384

1

st

_{Supervisor: Prof. dr. L. M. Sloot}

2

nd

_{Supervisor: MSc J. A. Koch}

Master Marketing Management 2017/2018

ABSTRACT: In the past decades, the interest for healthy and sustainable food consumption

has grown like never before. Using four food trends, an experimental research is conducted.

The trends are experience, convenience, healthiness, and sustainability. The existence of

these trends raises the question: How can experience and convenience be used to increase

healthy and sustainable sales? The findings indicate higher purchase intentions and the

willingness to pay a price premium for healthy and sustainable food. These effects appear to

be enhanced by consumer need for convenience, which is strongest for utilitarian shoppers.

(2)

2

Index

1. Introduction ... 3 1.1.Background ... 3 1.2 Problem statement ... 4 1.3 Relevance of study ... 4 1.4 Thesis structure ... 5 2. Literature review ... 6 2.1 Food trends ... 6

2.2 Decision making process ... 7

2.3 Purchase intentions and the willingness to pay a price premium ... 8

2.4 Product attributes ... 9

2.5 Consumer needs ... 11

2.6 Shopping motivations ... 13

2.7 Conceptual model ... 15

3. Research methods ... 16

3.1 Study 1 – Qualitative research – expert interview ... 16

3.1.1 Method ... 16

3.1.2 Findings and Conclusions ... 16

3.2 Study 2 – Quantitative research ... 17

3.2.1 Research context ... 17

3.2.3 Questionnaire design ... 17

3.2.3. Data collection process ... 20

3.2.4 Descriptive statistics ... 20

3.2.4 Analysis ... 22

4. Results ... 23

4.1 Factor analysis and grouping of variables ... 23

4.2 Manipulation check ... 25

4.3 Testing of hypotheses ... 25

5. Conclusion ... 36

6. Discussion and Limitations ... 37

6.1 Discussion ... 37

6.2 Limitations ... 39

7. Managerial Implications and Contributions to Literature ... 40

7.1 Managerial implications ... 40

7.2 Contributions to literature ... 40

References ... 41

(3)

3

1. Introduction

1.1.Background

A recent development in food marketing is companies trying to restore their image. The food industry is being compared to the tobacco industry, using similar marketing tactics (Chandon & Wansink, 2012). In addition, the corporate strategies that target children are controversial given the link between food marketing and childhood obesity (Kraak & Story, 2015). Last year, Coca Cola even was sued for false and misleading marketing regarding sugar (Foxnews, 2017).

“Whereas many consumers would like a healthy and sustainable diet and lifestyle, their good intentions are thwarted by non-conscious buying patterns and factors such as familiarity, taste expediency, price and how options are presented” (Lassen et al, 2016, p. 124). Viviani (2013) states that food is now fashionable, a social phenomenon that interests consumers and represents their lifestyles, and that it is important to understand that food lifestyle today is the result of individual choices dictated by tastes and trends. In other words, food is fashion, and trendy. “At its most basic, a trend can be defined as the direction in which something (and that something can be anything) tends to move and which has a consequential impact on the culture, society or business sector through which it moves” (Raymond, 2010, p. 14).

(4)

4

1.2 Problem statement

The concern consumers have with health and nutrition issues never have been stronger due to information streaming from manufacturers, retailers, government agencies, and health professionals (Trivedi, Sridhar & Kumer, 2016). Healthiness is a known and discovered trend, and the health concern seems to be stronger than ever. Nevertheless, obesity is a very serious issue: “the past two decades have witnessed a rapid increase in obesity among U.S. consumers. According to the Center for Disease Control, 34% of U.S. adults are obese, up from 23% in 1988” (Thomas, Desai & Seenivasan, 2011, p. 126). The same applies to sustainable products. Most consumers are aware of the fact that buying healthy and sustainable food is good for themselves and the world, but they still do not always buy such products. This may have several reasons. Van Herpen, Nierop, and Sloot (2012) state that the relatively high price of fair trade and organic foods can limit their demand, but on the other hand may increase quality perceptions. Consumer preferences differ per person, especially in such a versatile market as food. According to Chandon and Wansink (2012) many consumers want tasty, inexpensive, varied, convenient, and healthy foods on the short term, in that order of importance, and that by stimulating those interest marketers have contributed to the global obesity problem. The aim of this study is to discover how trends can be combined to increase sales of healthy and sustainable food products, using the EFMI food trend model as guideline. Hereby the four food trend drivers will be measured against the willingness to pay a price premium and purchase intentions. Health and sustainability are inherent to a product, where convenience and experience cope with the process of acquiring the product. The research question addressed in this study is: How can experience and

convenience be used to increase healthy and sustainable sales? Hereby convenience and experience

are used as moderating variables, where the other trends are part of manipulated products.

1.3 Relevance of study

(5)

5

communicating, delivering, and exchanging offerings that have value for customers, clients, partners, and society at large”. Thiele and Wiess (2003) investigated consumer demand for variety in food consumption and they state that knowledge of consumer preferences may serve as a criterion for market segmentation and assist firms in adapting marketing strategies more effectively to consumer needs. By analysing customer preferences, this thesis may provide insights on how companies, consumers, and the society at large all may benefit from the right products. Chandon and Wansink (2012) state that future research should ideally combine the best aspects of studies from consumer research, nutrition, and health. As such, they would provide the necessary link between specific marketing actions, individual short-term food choices, and long-term population weight gain. In this study the long-term food trends, and short-term food choices are investigated for health economics.

1.4 Thesis structure

(6)

6

2. Literature review

2.1 Food trends

The Erasmus Food Management Instituut, (from now on referenced to as EFMI) is a Dutch business school, which functions as an academical knowledge institute for the food sector (EFMI, 2018). In 2006, the EFMI developed the EFMI food trend model. This model describes the four biggest food trends in the Netherlands, which are health, sustainability, convenience and experience (EFMI, 2006). These trends were classified by means of continues written data (EFMI shopper monitor) amongst 1736 consumers in the period between July 2004 to June 2005, statistical sources and other relevant publications (EFMI, 2006). These four trends constitute the basic concepts of this thesis.

As mentioned earlier, a trend is the direction in which something tends to move and which has a consequential impact on the culture, society or business sector through which it moves (Raymond, 2010). The existence of these food trends is due to several reasons. Because of the changing lifestyle of western consumers, the demand for convenient food such as pre-cut fruit, fruit salads and ready-to-eat meals is increasing. Furthermore, consumer concerns regarding animal welfare, environmental issues and social aspects, such as salary and working conditions, bring about demand for products related to fair trade, organic production. (Trienekens et al., 2003).

Nijssen and Douglas (2008, p. 84) state that: “As advances in communications technology shrink the impact of geographic distance, consumers are likely to become more aware of and familiar with products and services in other parts of the world, as well as global social and ethical issues”. Throughout their paper, they prove that this is the case for food products as well. In their research among Dutch consumers their “findings provide important insights for international managers who market products that are ecologically friendly or have an environmental/ethical positioning or food products positioned as traditional products from another country or culture” Nijssen and Douglas (2008, p. 98). An example is “for instance, over the past 40 years, foods like yogurt and granola have gone from being foreign oddities to favourite staples. Knowing what created these new norms could help engineer sustainable healthy food trends of the future” (Wansink, 2017, p. 67).

(7)

7

2.2 Decision making process

Consumer decision making is definitely not a new field of research. Olshavsky and Granbois (1989) state that scientists who have performed research on consumer decision making may use different terminology, but all agree on the following: 1) there is a choice between two or more alternatives, 2) evaluative criteria facilitate the forecasting of each alternative's consequences for the consumers goals or objectives, 3) the chosen alternative is determined by a decision rule or evaluative procedure, and 4) information sought from external sources and/or retrieved from memory is processed in the application of the decision rule or evaluation procedure. Basically, this means that there are factors that distinguish the available choices from each other (product attributes), factors that influence the decision making context (contextual attributes), and consumer needs. In other words, the decision making process consists of consumer needs, product attributes, and contextual attributes. More recently, Puccinelli et al. (2009) propose a model on the consumer decision-making process, which is similar to other models in the very core. Their model consist of the steps 1) goals, schema, and information processing, 2) memory, 3) involvement, 4) attitudes, 5) affect, 6) atmospherics, and 7) attributions and choices. Again, it can be argued that this model consist of the same fundamental basic factors that influence the consumer decision-making process: product attributes, contextual attributes, and consumer needs. Applying these factors on the different food trends provides the following table:

Product Attributes Contextual attributes Consumer needs

Health 0% fat Diets Healthy lifestyle

Sustainability Eco-friendly Recyclable package Taking care of the planet

Experience Taste Ambience Enjoyment of food

Convenience Quick cook Delivery time Time reduction

Table 2.1 – food trend examples

(8)

8

2.3 Purchase intentions and the willingness to pay a price premium

If all products were priced the same, but some are superior to others, just the superior products would be sold. Unfortunately, it is more expensive to eat in a healthy and sustainable way (Barosh, Friel, Engelhardt & Chan, 2014). In this study, the willingness to pay a price premium relative to inferior products will be investigated. According to Homburg, Koschate, and Hoyer (2005, p. 86) companies may potentially charge a premium price for products and services that have a higher level of customer satisfaction. In their article, they conceptualize the willingness to pay as “a measure of the value that a person assigns to a consumption or usage experience in monetary unit”. In other words, the more satisficing a product or service is, the more money a consumer wants to pay for it. It can be argued that products, which are more healthy or sustainable than a comparable, conventional product, are more satisficing and thus consumers want to pay more. Harris’ (1997) findings underline this effect; parents were willing to pay premium prices for organic baby foods compared to conventional baby foods, because they perceived it as safer and healthier, which satisfied them.

Where willingness to pay a price premium may be important to distinguish the effects of different food trends, purchase intentions is the second dependent variable. This second variable is added for accuracy, as accurately measuring consumer willingness to pay is critical (Miller et al., 2011). Moreover, it would not make much sense to know the willingness to pay a price premium if there are no purchase intentions at all. Knowing the purchase intentions per trend makes it possible to accurately target consumers on the trend they are most attracted to.

(9)

9

2.4 Product attributes

The World Health Organisation defines health since 1948 as "a state of complete physical, mental, and social well-being and not merely the absence of disease or infirmity". Healthy food contributes to that. According to the still growing body of research, there is an increasing interest in healthy food. According to Wansink (2017), phenomena of recent years are the emergence of vigilant consumers (consumers that are highly informed, conscious of calories, and are influenced by nutrition information), and grocers who are motivated to sell healthy, profitable foods. According to EFMI (2006), the healthy trend is characterized by a healthy lifestyle, weight, functionality, and safety of foods. In agreement with QM, this study conceptualizes the healthy food trend as products that are fit to consume on a daily basis. The Google Trend analysis shows that ‘healthy food’ is a popular search term, and even gained interest over the period from 01-01-2007 to 13-03-2018. As Google trend needs a certain amount of searches to recognize a concept, this implies there is still a substantial amount of interest for the healthy food concept among consumers.

Figure 2.1 – healthy food (Google Trends, 2018)1

Healthy food also comes, at least in the minds of consumers, with downsides. One of those downsides may be that consumers think “they need to eat unhealthy to eat tasty or that eating healthy means eating food that is not tasty” (Raghunathan, Naylor & Hoyer, 2006, p. 177). Another perceived downside is that the desire to eat healthy may compete with the desire to fulfil one’s appetite (Finkelstein & Fishbach, 2010). Despite the perceived disadvantages of healthy food, it is expected that consumers are willing to pay more for healthy products in an unbiased situation, because they now it is better to do so. The discrepancy between choice and knowledge is called the attitude-behaviour gap (De Pelsmacker Driesen & Rayp, 2005). This theory explains that one’s attitudes and behaviours bias information processing, and lead to distorted attitudes (Samson & Voyer, 2012). The expectation is that unbiased consumers are, compared to conventional products, more satisfied by healthy products and as satisfied consumers are willing to pay more the first hypotheses are:

Hypothesis 1a1: If a healthy component is added to a food product, the WTPP increases.

Hypothesis 1a2: If a healthy component is added to a food product, the PI increase.

(10)

10

The same accounts for sustainable products. Although consumer interest in sustainable products may be growing, sustainable food markets remain niche markets, attracting only very specific consumers (Vermeir & Verbeke, 2006). According to EFMI, the sustainability trend is about taking the well-being of humans, animals, nature, and the environment into account. In the literature, sustainability is an overarching concept. Sustainable products have attributes and consequences that contribute to economic, social, and environmental goals (Vermeir & Verbeke, 2006). In accordance with QM, the trend for sustainable foods will be conceptualized as the tendency for food that is environmentally friendly and does no harm to humans or animals. The Google trend analysis shows a slight increase in consumer interest over the period from 01-01-2007 to 13-03-2018.

Figure 2.2 – sustainable food (Google Trends, 2018)

The interest seems to be there, so consumers are expected to know that sustainable products are better for themselves, society and the planet. A reason that consumers do not buy sustainable products may be the existence of “a consumers’ lay theories about the relationship between a firm’s intentions and its allocation of resources. Specifically, when a company intends to make a product better for the environment, consumers assume that in order to make a product more environmentally friendly, the company diverted resources away from product quality” (Newman, Gorlin & Dhar, 2014, p. 833). In other words, products that are superior on the sustainability aspect may lead consumers to think that the product is inferior on other aspects. This theory comes close to the unhealthy is tasty

intuition of Raghunathan, Naylor, and Hoyer (2006). The same attitude-behaviour gap appears to be

present here. Thus, as with healthy products, it is expected that unbiased consumers are more satisfied by sustainable products, and hence are willing to pay more for such products. Which leads to the following hypotheses:

Hypothesis 1b1: If a sustainable component is added to a food product, the WTPP increases.

(11)

11

2.5 Consumer needs

Where the first two hypotheses focussed on product attributes and unbiased consumers, the following paragraph focusses on consumer needs, with respect to the remaining trends experience and convenience. As mentioned earlier, the conceptualization of experience and convenience will focus on how consumer needs influence the decision making process, hereby attending to the attitude-behaviour gap, nudging consumers towards the right choice.

Retailers, restaurants, and food service companies can influence the ambient characteristics of the point of purchase and of the point of consumption. Elder and Krishna (2009) for instance, show that advertisements for food may even influence the perceived taste of consumers. Also, some atmospheric dimensions, such as temperature, have a direct physiological effect (Chandon & Wanskink, 2012). According to EFMI the experience trend is characterized by the palatability of food, enjoyment of shopping, trying new things, and authenticity (EFMI, 2006). For purpose of this research, consumer need for experience will be investigated, at which the experience food trend will be defined as the tendency for the enjoyment of food. Looking at the Google trend graph, there is clear evidence of an upward trend towards experience food in the period from 01-01-2007 to 13-03-2018.

Figure 2.3 – experience food (Google Trends, 2018)

Dhar, Huber, and Khan (2007) show that the nature of the shopping experience may affect the goals that a consumer pursues. This experience may lead people to be distracted from the choice they know is right. The unhealthy is tasty bias may lead consumers to be less willing to buy healthy products if there is no experience component. Another research by Mai and Hoffmann (2015) shows that taste perceptions have more influence on food decisions than healthiness expectations. Altogether, this means that if a healthy product is enjoyed less, a consumer will buy something else when the need for this enjoyment is there. He or she might then value this enjoyment over the positive product attributes. According to literature, the need for experience attends the attitude-behaviour gap in such a way that it is more important than healthiness, and this expectation leads to the following hypotheses:

Hypothesis 2a1: Need for experience will have a negative influence on the WTPP for products with

(12)

12

Hypothesis 2a2: Need for experience will have a negative influence on the PI for products with healthy

elements.

With regard to food products with sustainable elements, the relationship may be the other way around. In her research Sims (2009) shows that the experience of local food products – which had all kinds of environmental and socially sustainable benefits – was the reason of consumer interest. Thus, the sustainability of the products was the experience in this case. As said, consumers have knowledge about sustainability, but as experience is more intense than education it is memorized better (Hoch, 2002). This might result in repeated consumer interest, when the sustainable aspect is enjoyed.Hence, this leads to the following hypotheses:

Hypothesis 2b1: Need for experience will have a positive influence on the WTPP for products with

sustainable elements.

Hypothesis 2b2: Need for experience will have a positive influence on the PI for products with

sustainable elements.

The focus on improving the convenience of food preparation and consumption is one of the strongest trends in food marketing (Chansdon & Wansink, 2012). According to EFMI the convenience trend is defined in terms of the convenience of shopping and cooking, by means of easy packaging and convenience stores. For purpose of this research, the convenient food trend will be defined as the tendency for food products that are easy to acquire and prepare. Consumer need for convenience will be used as independent variable to measure the effects of the convenience trend, hereby attending the need for easiness to prepare and acquire food products. As for all trends, the slope of the line of the Google trend analysis shows an increase in consumer interest in the period from 01-01-2007 to 13-03-2018.

(13)

13

Walker, Keane and Burke (2010) state that the choice for convenience often comes at the sacrifice of costs and quality of the products as convenience stores offer less healthy, more expensive food. Ease of preparation is a strong driver of intake, and while it is known that food products that are easy to prepare are less healthy, the demand for such products is increasing (Chandon & Wansink, 2012). Fast food may very well be the ultimate convenient food, and a study by Lassen et al. (2016) shows that the majority of fast food consumers would like healthier and more sustainable menu options, but that price was more important. In other words, the willingness to pay a premium is not big enough to make the best decision in terms of healthiness and sustainability. However, as the demand seems to be there, the efforts could also be combined. So, what if a healthy product is easy to acquire and quick to prepare? Making healthy products convenient may nudge consumers to buy them. Thus, if a healthy product meets the consumer need for convenience, the following hypotheses are constructed:

Hypothesis 3a1: Need for convenience will have a positive influence on the WTPP for products with

healthy product elements.

Hypothesis 3a2: Need for convenience will have a positive influence on the PI for products with

healthy product elements.

The same accounts for sustainable food. Hjelmar (2011) showed that making organic products more convenient increased the sales volumes. One can hereby think of products as pre-cut, biological poultry. This leads to the following hypotheses:

Hypothesis 3b1: Need for convenience will have a positive influence on the WTPP for products with

sustainable product elements.

Hypothesis 3b2: Need for convenience will have a positive influence on the PI for products with

sustainable product elements.

2.6 Shopping motivations

(14)

14

consumers who can be described as hedonic shoppers, or shoppers that are interested in the emotive aspects of shopping, have a need for experience as well.Thus, the following hypothesis will be tested:

Hypothesis 4a: Hedonic shopping motivations have a positive influence on need for experience.

The same accounts for utilitarian shopping motivations and convenience. The preparation time and the effort of acquiring facets of the convenience trend, and the task-related and rational motivational shopping aspects are expected to be related too. Therefore, the last hypothesis that will be tested is:

Hypothesis 4b: Utilitarian shopping motivations have a positive influence on need for convenience.

This chapter will end with an overview of all hypotheses and the conceptual model. H1a1 If a healthy component is added to a food product, the WTPP increases. H1a2 If a healthy component is added to a food product, the PI increase.

H1b1 If a sustainable component is added to a food product, the WTPP increases. H1b2 If a sustainable component is added to a food product, the PI increase.

H2a1 Need for experience will have a negative influence on the WTPP for products with healthy elements.

H2a2 Need for experience will have a negative influence on the PI for products with healthy elements.

H2b1 Need for experience will have a positive influence on the WTPP for products with sustainable elements.

H2b2 Need for experience will have a positive influence on the PI for products with sustainable elements.

H3a1 Need for convenience will have a positive influence on the WTPP for products with healthy product elements.

H3a2 Need for convenience will have a positive influence on the PI for products with healthy product elements.

H3b1 Need for convenience will have a positive influence on the WTPP for products with sustainable product elements.

H3b2 Need for convenience will have a positive influence on the PI for products with sustainable product elements.

H4a Hedonic shopping motivations have a positive influence on need for experience. H4b Utilitarian shopping motivations have a positive influence on need for convenience.

(15)

15

2.7 Conceptual model

Figure 2.5 – Conceptual model

(16)

16

3. Research methods

In this section, the methodology of this research is described. Preceding the main, quantitative analysis a small qualitative study in the form of an expert interview is conducted in section 3.1. The conclusions from this interview will be used in the main analysis, which is described in section 3.2 and further. 3.1 Study 1 – Qualitative research – expert interview

In order to review the insights from literature, an interview with an expert on the field of research was arranged. Simon Bunt was interviewed. Mister Bunt is currently working as business developer at Questionmark (QM) Intelligence. Before that, he used to work at a big corporate supermarket chain, leading projects on sustainability and organizing health interventions. As QM intelligence is a database that collects information on food, he is a true expert on the field of food trends. His insights can be used to compare literature to practice. QM intelligence tries to make information on food products more transparent and to make this possible, they collect data on sustainability, nutritional value, and production impact on hundreds of thousands of products.

3.1.1 Method

Qualitative data is particularly interesting as it offers real and full insights to phenomena (Sinkovics, Pens, & Ghauri, 2005). To get good information from the respondent open questions were asked, as open questions are potentially rich in content (Bachmann, Elfrink & Vazzana, 1999). The interview took place at the office of QM intelligence in Amsterdam on the 26th_{of March, 2018.}

3.1.2 Findings and Conclusions

(17)

17

3.2 Study 2 – Quantitative research

The expert interview confirmed that the classified trends, product attributes, and decision attributes are important in the Dutch food market and that there is need for a research like this. Thus, the second study is a quantitative study, which will be set out in this section.

3.2.1 Research context

This study is a master thesis for the master marketing. For purpose of this research, a 2x2 between subjects experimental design was chosen. According to Malhotra (2010) a between subjects design is a design wherein each respondent is exposed to only one treatment condition. The rest of the survey questions were the same for every respondent. This way, the differences between the groups can be investigated. The different treatment conditions were health (yes/no) x sustainability (yes/no) creating four different conditions. Respondents were randomly assigned to one of the conditions, using the questionnaire program ‘Qualtrix’.

3.2.3 Questionnaire design

(18)

18

emissions of the animals (QM, 2018). Because QM is relatively new in business the human rights scores are not known for every product, therefore the manipulated products are locally produced, as human rights violation in production processes is almost non-existent in the Netherlands (Maplecroft, 2016). Based on that reasoning, some yogurts get a so-called biological label, which is strictly given to products that have a high ranking on animal welfare and environmental impact. In short, this means that the biological yogurt products are eco-friendly products, considering animal welfare. This resulted in the following four products:

Figure 3.1 – Manipulated products

The questionnaire started with some context of the research, stating that this research is a master thesis on eating habits and buying behaviour. In addition, a situation is created where respondents were shopping for groceries in their supermarket. The trends themselves were not mentioned so that the respondents would not be affected by that knowledge.

(19)

19

reasonable price for the products, resulting in the following question: ‘I would pay the following amount for this product’. Respondents could check the following boxes: <€0,89; €0,99; €1,09; €1,19; €1,29; €1,39; €1,49; €1,59; €1,69; 1,79; €1,89. This way the means can easily be compared, and the average premium can be calculated. Hereafter some control questions about shopping at Albert Heijn, attitude towards low-fat yogurt, and yoghurt buying behaviour were asked in order to ensure the validity of the results.

The consumer need for convenience was measured on a 7-point Likert scale, where respondents had to indicate whether they agreed or disagreed with certain opinions. Scales of Maher, Marks and Grimm (1997), and Chandon and Wansink (2002) were used, resulting in the following statements for convenience: ‘When it comes to shopping for food, convenience is the most important thing to me’, ‘I always buy my groceries in the nearest supermarket’, ‘I like products that are ready to consume without any further preparation the most’, and ‘I prefer products that are very easy and convenient to consume’.

The same accounted for measuring experience: a 7-point Likert scale, where respondents had to indicate whether they agreed or disagreed with certain opinions. The scales were derived from the unhealthy is tasty intuition measuring the enjoyment of food (Raghunathan, Naylor & Hoyer, 2006). The statements are: ‘Food should particularly be tasty’, ‘Food always must be delicious’, ‘Food must be appealing’, and ‘Enjoying food is the most important for me’.

The next part of the survey was about whether a consumer is a hedonic or a utilitarian shopper, and is measured by agreeing or disagreeing with several statements. Batra and Ahtola (1990) suggest that in order to have the highest validity and reliability, hedonic motivations should be measured in terms of, amongst others, pleasantness, fun, inspiring and enjoyable. Utilitarian motivations should be measured in terms of useful, necessary, practical, and valuable. For each of the eight terms respondents had to indicate on a 7-point Likert scale to which degree the terms were applicable to their person.

(20)

20

3.2.3. Data collection process

As Dutch food trends were investigated, respondents had to be consumers on the Dutch grocery market. To make the survey as convenient as possible and to enlarge the possible target audience, all questions were in Dutch. Before the survey was distributed, a pre-test was done. Pre-testing is the testing of the questionnaire on a small sample of respondents to identify and eliminate potential problems (Malhotra, 2010). Some mistakes were taken out of the survey afterwards. As said, the survey was made in Qualtrix. The program provides the user with a link to the survey, which randomly assigns respondents to one of the conditions, equally distributing them. This link was distributed on social media (Facebook and LinkedIn), by email, and by Whatsapp. Also, various people shared the link with colleagues and business contacts. In the end, 251 respondents filled in the survey.

3.2.4 Descriptive statistics

The following tables show information about 184 of the 251 respondents that took the survey. Out of the 251 respondents, 223 completed the survey. Seven of those respondents had conflicting diets or allergies and were disregarded. The rest of the discarded responses failed the manipulation check, which will be elaborated on in section 4.2. Hereafter, some responses had too many missing values leaving the sample with 184 genuine responses. There were some missing values and outliers replaced by the mean.

Gender Women: 84 Men: 100

Age Range: 18-80 Mean: 38 SD: 16

Pre-tax income (monthly) Range: < €1000 - > €4000 Mean: €2343 SD: €1365

Level of education (N) VMBO: 2

HAVO: 4 VWO: 3 MBO: 30 HBO: 61 WO: 83 Other: 1

Table 3.1 – Descriptive statistics

(21)

21

relatively well educated. As said, Qualtrix randomly distributed each respondent to one of the four package types. An overview of the descriptive statistics of the respondents per package type is provided next.

Base condition (N=49)

Pre-tax income (monthly) Range: < €1000 - > €4000 Mean: €2275; SD: €1398

Table 3.2 – Descriptive statistics base condition

Healthy condition (N=43)

Table 3.3 – Descriptive statistics healthy condition

Sustainable condition (N=44)

Table 3.4 – Descriptive statistics sustainable condition

Healthy and Sustainable condition (N=48)

Table 3.5 – Descriptive statistics healthy and sustainable condition

(22)

22

3.2.4 Analysis

In order to test the hypotheses a variety of statistical analysis in SPSS will be carried out on the data provided by the survey. First of all the data will be prepared, following steps provided by Malhotra (2010). First of all the questions need to be coded to provide a better overview of the questions in SPSS. Secondly, the data must be cleaned. In order to do so the data was checked on outliers, extreme values, and missing values in order to provide consistency. As said, those responses were discarded or replaced by the mean. Hereafter some variables needed scale transformation, and this was done by rescaling them in SPSS. When the data was prepared, a data analysis strategy was selected.

First of all, multiple questions that aimed to measure the same variables should be grouped together. This will be done using factor and reliability analysis. Hereafter, the means of the WTPP and PI are of interest. This will be measured by carrying out an analysis of variance (ANOVA). This test shows the mean values of different treatment groups, which in this study are the different package types. The ANOVA also shows which variables cause significant change in the means of the different treatment groups. Hypotheses 1a1, 1a2, 1b1 and 1b2 will be tested via this analysis. Hereafter, the different

package types will be regressed against the dependent variables to quantify the individual relationships per package type with the dependent variables. Some dummy variables have to be created in order to properly run those tests.

To check for the effect of the moderating variables, an analysis of covariance will be carried out. A covariate is a factor that causes difference, but cannot be controlled for. Meaning that its effect must be measured. When the effect of the covariate is known, one knows what causes the differences in the means of the different treatment groups. Liu (2010) states that in comparison to analysis of variance without covariates, analysis of covariance (ANCOVA) yields a smaller within-treatment error variance, which helps decrease the standard error of the adjusted mean difference. When the standard error, is decreased the precision of the measurement of the mean is increased. This way hypotheses 2a1, 2a2, 2b1, 2b2, 3a1, 3a2, 3b1 and 3b2 will be tested in order to measure the effect of consumer need

for experience and convenience.

(23)

23

4. Results

This chapter is about the results of the survey, along with the acceptance or rejection of the hypotheses.

4.1 Factor analysis and grouping of variables

Factor analysis is the class of procedures primarily used for data reduction and summarization (Malhotra, 2010). First of all the data has to be checked on the appropriateness of factor analysis by the Kaiser-Meyer-Olkin measure of sampling accuracy and Bartlett’s test of sphericity. If the KMO has a value between 0,5 and 1,0 and Bartlett’s is significant, the factor analysis is appropriate. Hereafter, the amount of variables to be grouped is determined by looking at the eigenvalues of the variables (>1) and the percentage of variance explained by the variables (>5% each; 60% in total). When the right number of variables is determined the Cronbach’s Alpha of the combined variable must be above 0,6 to ensure internal consistency of the new variable. An overview of the grouped variables is provided in table 4.1. The questions measuring the need for convenience had a KMO of 0,64 and a significant score on Bartlett’s test, so factor analysis was appropriate. Extracting all four questions provided an eigenvalue of 1,949, all questions explained >5%, and the total variance explained was 48%. Extracting three questions enlarged the total variance explained, but lowered all other tests. Thus, all four questions were grouped together. This provided a Cronbach’s Alpha of 0,65 being a reliable variable.

The questions measuring the need for experience had a KMO of 0,74 and a significant score on Bartlett’s test, so factor analysis was appropriate again. Extracting all four questions provided an eigenvalue of 2,232, all questions explained >5%, and the total variance explained was 56%. Again, extracting three questions enlarged the total variance explained, but lowered the other tests. Thus, all four questions were grouped together. This provided a Cronbach’s Alpha of 0,73 which is a reliable variable. The questions measuring hedonism had a KMO of 0,77 and a significant score on Bartlett’s test, so again, factor analysis was appropriate. Extracting all four questions provided an eigenvalue of 2,785, all questions explained >5%, and the total variance explained was 70%. Thus, once more, all four questions were grouped together. This provided a Cronbach’s Alpha of 0,85 being a very reliable variable.

(24)

24

enlarged the total variance explained, but lowered all other tests. Thus, all four questions were grouped together. This provided a Cronbach’s Alpha of 0,51 being a not a very reliable variable. Where a minimum value of 0,6 is the most common rule, Peterson (1994) also takes lower values into account (Nunally: 0,5-0,6; Davis: 0,5) in his meta-analysis of the Cronbach’s Alpha. Unfortunately, the ‘Cronbach’s Alpha if item deleted’ even lowered the total Cronbach’s Alpha in all cases, meaning the unreliable variable was the best option for this data. As this is not the first study to work with a variable with a low Cronbach’s Alpha, and the variable is only of interest for one hypothesis, it will be maintained. This ensures some limitation with regard to hypothesis 4b and will be elaborated on in the discussion session. As mentioned before, table 4.1 contains all variables.

Variable Questions α

Healthiness Manipulated n/a2

Sustainability Manipulated n/a

PI Which product do you prefer, at first sight? n/a

WTPP I would pay the following amount for this product: n/a

Convenience

When it comes to shopping, convenience is the most important to me.

0,65 I always buy my groceries in the nearest supermarket.

I like products that are ready to consume without any further preparation the most.

I prefer products that are very easy and convenient to consume.

Experience

Food should particularly be tasty.

0,73 Food always must be delicious.

Food must be appealing.

Enjoying food is the most important for me.

Hedonism Pleasantness 0,85 Fun Inspiring Enjoyable Utilitarianism Useful 0,513 Necessary Practical Valuable

Table 4.1 – Overview of variables

(25)

25

4.2 Manipulation check

Two types of manipulation checks were performed. One to check whether the respondents had the right product in mind throughout the survey, and one to check if the manipulation of the product attributes was perceived the way it should have been perceived. To check whether the respondents had the right product in mind, some questions about the package types were asked. The two most important questions were about the added sugar and sustainability of the product, asking if the package that was shown had those elements. When respondents gave the wrong answer on both of the questions the manipulation was considered as failed. In this case this respondent was disregarded in order to get genuine results about the products. Twenty-four responses were disregarded, mainly because respondents thought the base condition had sustainable, as well as healthy elements. When one, for instance, perceived the base condition as healthy and sustainable, the WTPP and PI are not reliable in comparison to the real healthy or sustainable products.

The perceived healthiness and sustainability was controlled for as well, bearing in mind that it is “rarely safe to assume that the operations used manipulate psychological and sociological variables will represent the precise concepts the researcher has in mind” (Perdue and Summers, 1986, p. 317). Respondents were asked how healthy and sustainable they perceived the shown product. On average, the base condition had lower mean values on healthiness and sustainability than the other manipulated products. These results indicate that the manipulated products were indeed perceived healthier and more sustainable than the base condition, meaning that the product manipulation succeeded. With regard to the perceived healthiness, the results were significant with F(4,183) = 2,233 and P = 0,09. The results measuring the sustainability were insignificant F(4,183) = 1,277 and P = 0,28.

4.3 Testing of hypotheses

(26)

26

*Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

Table 4.2 – ANOVA results

The results show that there is a significant relation between package type and both WTPP and PI. Meaning that if the package type is changed, a significant change in the mean value per package type occurs with F(3,183) = 4,849 and P = 0,003 for the WTPP, and F(3,183) = 2,415 and P = 0,068 for the PI. The mean values per package type are displayed below. As all control variables are highly insignificant, they will be left out for further analyses.

WTPP PI

Package Type Mean SD Mean SD

Base condition (N=49) 5,55 2,39 0,33 0,75

Healthy (N=44) 6,23 2,22 0,37 0,79

Sustainable (N=43) 7,09 2,62 0,45 0,85

Healthy and Sustainable (N=48) 7,17 2,36 0,75 0,98

Table 4.3 – Overview of means and standard deviations

A graphical overview of the means is provided in figure 4.1. The results show that the average WTPP and PI for the base condition is lowest in all conditions. With regard to both the WTPP and the PI, the means move in the same directions. The package with both sustainable and healthy elements has the highest mean, followed by the sustainable package and the healthy package. At first sight, respondents appear to have a higher preference for sustainable than for healthy packages.

WTPP PI

Dependent Variables F-value P-value F-value P-value

Package type 4,849 0,003*** 2,415 0,068*

Control variables F-value P-value F-value P-value

Yoghurt buying behaviour 0,125 0,945 0,125 0,945

Grocery buying behaviour 0,078 0,972 0,078 0,972

Attitude towards low-fat yoghurt 0,203 0,894 0,203 0,894

Gender 0,600 0,616 0,600 0,616

Level of education 0,412 0,744 0,412 0,744

(27)

27

Figure 4.1 – Overview of means

As said, regression analyses are performed next, in order to check the individual influence the different packages have on both dependent variables. The results of the regression analyses with WTPP as dependent variable are displayed in the following table:

Dependent variable = WTPP B B (SE) β T-value P-value Adj. R2

Constant _6,844 _0,208 _32,924 _0,000*** _0,048

Base condition (B=1, 0) _-1,293 _0,403 _-0,232 _-3,211 _0,002***

B B (SE) β T-value P-value Adj. R2

Constant _5,551 _0,331 _16,794 _0,000*** _0,011

Healthy (H=1, B=0) _0,682 _0,483 _0,147 _1,410 _0,162

Constant _5,551 _0,357 _15,529 _0,000*** _0,078

Sustainable (S=1, B=0) _1,540 _0,425 _0,297 _2,963 _0,004***

Constant _5,551 _0,339 _16,377 _0,000*** _0,096

Healthy and sustainable

(HS=1, B=0) 1,616 0,482 0,325 3,353 0,001*** *Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

Table 4.4 – Regression analysis

0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 1,00 1 2 3 4 5 6 7 8 9

Base condition Healthy Sustainable Healthy and Sustainable

Overview of means

WTPP PI

(28)

28

The regressions show the relationship between the different packages types and the willingness to pay a price premium. The B-value shows the intercept of the regression line, β shows the direction of the relationship. Between the base condition and WTPP a negative significant relationship exists (T = -3,211 and P = 0,002). When adding healthy or sustainable elements to a product, this relationship becomes positive with T = 1,410 and P = 0,162 for healthy products, T = 2,963 and P = 0,004 for sustainable products, and T = 3,353 and P = 0,001 for the package with both elements. The intercept of these relations is the same, and the effects of the package types become increasingly larger when adding healthy, sustainable, or both elements. The regression analysis of the healthy condition is not significant, and compared to the other packages it seems that sustainability is increases the WTPP the most. Considering these results, the following tables show regression analyses that measure the differences between the healthy condition and the condition with both elements (table 4.5), as well as the sustainable condition and the condition with both elements (table 4.6).

Constant _6,233 _0,350 _17,816 _0,000*** _0,030

(HS=1, H=0) 0,934 0,482 0,201 1,939 0,056* *Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

The relationship is positive and significant, with T = 1,939 and P = 0,056. Meaning that adding a sustainable element to a healthy product significantly increases the WTPP.

Constant 7,091 0,375 18,919 0,000 -0,011

(HS=1, S=0) 0,076 0,519 0,015 0,146 0,146 *Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

The relationship is positive, but insignificant with T = 0,146 and P = 0,146. Meaning that adding a healthy element to a sustainable product does not significantly increase the WTPP.

As hypothesis 1a1 expected the WTPP to increase when a healthy component is added, this hypothesis

is rejected. The WTPP seems to increase, but this relation is insignificant. With regard to hypothesis 1b1, a significant result is found, hence this hypothesis is accepted. Adding a sustainable element does

(29)

29

The same regression analyses were run again, this time with purchase intentions as dependent variable. Table 4.7 reports the results of the tests.

Dependent variable = PI B B (SE) β T-value P-value Adj. R2

Constant 0,533 0,073 7,266 0,000*** 0,006

Base condition (B=1, 0) -0,207 0,142 -0,107 -1,454 0,148

Constant _0,327 _0,109 _2,984 _0,004*** _-0,010

Healthy (H=1, B=0) _0,046 _0,160 _0,030 _0,285 _0,777

Constant _0,327 _0,114 _2,871 _0,005*** _-0,004

Sustainable (S=1, B=0) _0,128 _0,165 _0,081 _0,774 _0,441

Constant _0,327 _0,124 _2,630 _0,010** _0,057

(HS=1, B=1) 0,423 0,177 0,239 2,399 0,018** *Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

(30)

30

Constant 0,372 0,136 2,731 0,008*** 0,033

(HS=1, H=0) 0,378 0,188 0,209 2,014 0,047** *Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

Again, the regression shows that there is a positive effect on the PI if a sustainable element is added to a healthy product. This effect is significant with T = 2,014 and P = 0,047.

Constant _0,455 _0,138 _3,283 _0,001 _0,015

(HS=1, S=0) 0,295 0,192 0,160 1,541 0,127 *Significant at P < 0,10; **Significant at P < 0,05; ***Significant at P < 0,01

Once more, the relation is similar to the aforementioned results. Adding a healthy element to a sustainable product increases the PI, but with T = 1,541 and P = 0,127 this result is insignificant.

As the relations between PI and both the healthy condition and the sustainable condition are positive but insignificant, hypothesis 1a2 and hypothesis 1b2 are rejected. Adding just a healthy or a sustainable

element is not enough to result in significantly higher purchase intentions, a combination of both elements appears to be needed.

In order to measure the influence of the consumer need for convenience and experience between the treatment groups, two between-subject analyses of covariance are conducted. Hypotheses 2a1, 2a2,

2b1, 2b2, 3a1, 3a2, 3b1 and 3b2 are tested this way. Before an analysis of covariance is conducted, the

assumption of the homogeneity of slopes has to be tested in order to make sure that the covariates have the same influence on all variables throughout the sample. In other words, the effects of convenience and experience should be independent of the different package types. As all of the regressions were insignificant, the ancova could be continued.

Hypotheses 2a1, 2a2, 2b1, and 2b2, test the effect of the covariate experience on the package type

(31)

31

difference between the different package types, controlling for the need for experience. The results of the ancova for experience are presented in table 4.10.

WTPP PI

Independent Variables F-value P-value F-value P-value

Package type 4,762 0,003*** 1,045 0,133

Covariate F-value P-value F-value P-value

Experience 0,025 0,875 1,045 0,413

Table 4.10 – ANCOVA experience

With regard to the main effects, different package types still have a significant influence on the WTPP as F(4,183) = 4,762 and P = 0,875 while controlling for the effect of covariate experience. The covariate itself creates no significant results as F(4,183) = 0,025 and P = 0,875, meaning that experience does not explain a significant amount of the variance between the package types. The same accounts for the purchase intentions, as experience generates an F-value of 1,045 and a P-value of 0,413. The main effect here is insignificant as well as F(4,183) = 1,045 and P = 0,13.

Table 4.11 and figure 4.2, illustrate the adjusted mean values for the different package types. The mean values move in the same directions, relative to each other, as the unadjusted mean values do. The base condition generates the lowest mean, the package with both elements creates the highest mean, and the sustainable package and the healthy package are in between. All mean values went up relative to the unadjusted means, except the mean WTPP of the sustainable condition. This is contrary to the expectations as both the PI and the WTPP were supposed to increase for the sustainable condition, and decrease for the healthy condition. Also, all standard deviations went up, decreasing measurement precision.

WTPP PI

Base condition (N=49) 5,35 3,27 0,38 1,02

Healthy (N=44) 6,29 3,13 0,47 1,03

Sustainable (N=43) 7,02 3,07 0,52 1,03

(32)

32

Figure 4.2 – Overview of adjusted means (experience)

Taking into account that the covariate experience was insignificant for both cases, that the measurement precision decreased, and the changes in the mean values, all tested hypothesis are rejected. For clarity, hypotheses 2a1,2a2, 2b1, and 2b2, all propose an effect of consumer need for

experience. According to the ancova this influence is insignificant. Hypotheses 2a1 and 2b1 proposed a

negative effect of consumer need for experience on WTPP and PI. The adjusted means of those variables even went up. In addition, the adjusted mean value of the WTPP in the sustainable condition went down, contrary to hypothesis 2a2. The only adjusted mean value that moved in the proposed

direction was the PI for the sustainable condition. However, as said, this result was insignificant.

Again, an ancova was run, this time checking for a statistically significant difference between the different package types controlling for the consumer need for convenience. Hypotheses 3a1, 3a2, 3b1,

and 3b2 are tested this way. The results of this ancova are reported in table 4.12.

WTPP PI

Independent Variables F-value P-value F-value P-value

Package type 5,330 0,002*** 3,028 0,031**

Covariate F-value P-value F-value P-value

Convenience 2,485 0,117 8,121 0,005***

Table 4.12 – ANCOVA convenience

0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 1,00 1 2 3 4 5 6 7 8 9

Overview of adjusted means

WTPP PI

44 (N) 43 (N) 48 (N)

(33)

33

This time, both main effects are significant with F(4,183) = 5,330 and P = 0,002 for the WTPP and F(4,183) = 3,028 and P = 0,031 with regard to the PI. There is a significant effect of the covariate convenience on the purchase intentions with F(4,183) = 8,121 and P = 0,005. This means that consumer need for convenience has a significant influence on the variance in the means for the different package types for PI. The adjusted means for PI are reported in table 4.13 and figure 4.3, along with the adjusted mean values of the WTPP. The effect of the covariate convenience on the WTPP is insignificant as F(4,183) is 2,485 and P = 0,117.

WTPP PI

Base condition (N=49) 5,51 2,398 0,30 0,832

Healthy (N=44) 6,20 2,396 0,35 0,831

Sustainable (N=43) 7,14 2,399 0,48 0,832

Table 4.13 – Overview of means (convenience)

Once more, the adjusted mean values are related to each other the same way as the unadjusted means. The base condition still is the least wanted package, and the healthy and sustainable package has the highest WTPP and PI. The effect of convenience lowers the mean value of the base condition with regard to both dependent variables. The WTPP increases for both the healthy condition and the sustainable condition, but as the effect of the covariate is insignificant, hypotheses 3a1 and 3b1 are

rejected. Adding a healthy element lowers the adjusted mean value of the PI relative to the unadjusted mean. However, the difference between the adjusted mean values of the base condition and the healthy condition increases. As the effect of the covariate is significant and the influence is as proposed, hypothesis 3a2 is accepted. Hypothesis 3b2, that proposes a positive influence of experience

(34)

34

Figure 4.3 – Overview of adjusted means (convenience)

In order to test hypotheses 4a and 4b, two correlation tests were carried out. These tests show how, and to what extend the variables are related to each other. Note that utilitarian shoppers are hypothesized to have a need for convenience, and hedonic shoppers should have need for experience. The results of the tests are displayed in table 4.14.

Correlation Coefficient P-value Correlation between need for experience and

hedonic shopping motivations 0,211 0,002**

Correlation Coefficient P-value Correlation between need for convenience and

utilitarian shopping motivations 0,097 0,095*

Table 4.14 – Correlation analysis

For both tests, a significant positive result was found. Hedonic shopping motivation has a positive significant correlation with consumer need for experience with r = 0,211 and P = 0,002. Therefore, Hypothesis 4a is accepted. The same accounts for utilitarian shopping motivations and convenience. The effect is somewhat weaker, but still significant and positive with r = 0,097 and P = 0,095, meaning that hypothesis 4b is accepted as well.

0,00 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 1,00 1 2 3 4 5 6 7 8 9

Base Healthy Sustainable Healthy and Susainable

Overview of adjusted means

WTPP PI

44 (N) 43 (N) 48 (N)

(35)

35

This section ends with table 4.15, providing an overview of all accepted and rejected hypotheses. Section 5, conclusion, is next, followed by section 6, discussion.

H1a1 If a healthy component is added to a food product, the WTPP increases. Rejected H1a2 If a healthy component is added to a food product, the PI increase. Accepted H1b1 If a sustainable component is added to a food product, the WTPP

increases.

Rejected

H1b2 If a sustainable component is added to a food product, the PI increase. Rejected H2a1 Need for experience will have a negative influence on the WTPP for

products with healthy elements.

Rejected

H2a2 Need for experience will have a negative influence on the PI for products with healthy elements.

Rejected

H2b1 Need for experience will have a positive influence on the WTPP for products with sustainable elements.

Rejected

H2b2 Need for experience will have a positive influence on the PI for products with sustainable elements.

Rejected

H3a1 Need for convenience will have a positive influence on the WTPP for products with healthy product elements.

Rejected

H3a2 Need for convenience will have a positive influence on the PI for products with healthy product elements.

Accepted

H3b1 Need for convenience will have a positive influence on the WTPP for products with sustainable product elements.

Rejected

H3b2 Need for convenience will have a positive influence on the PI for products with sustainable product elements.

Accepted

H4a Hedonic shopping motivations have a positive influence on need for experience.

Accepted

H4b Utilitarian shopping motivations have a positive influence on need for convenience.

Accepted

(36)

36

5. Conclusion

The aim of this study was to combine the four EFMI food trends, using consumer need for experience and convenience to increase the sales of healthy and sustainable food products. A quantitative study in the form of a controlled experiment was conducted in order to find a solution for the problem.

The results show that respondents are willing to pay a price premium, and have higher purchase intentions towards healthy and sustainable products. Unfortunately, some results were insignificant. Hence, hypothesis 1a1, 1b2 and 1b1 were rejected, while hypotheses 1a2 was accepted. Adding both

elements at the same time appeared to be the most effective. Consumer need for experience was proposed to positively influence the purchase intentions and willingness to pay a price premium for sustainable, and to decrease those values for healthy products. However, no statistical evidence was found for these relationships. Therefore, hypotheses 2a1, 2a2, 2b1, and 2b2 were rejected. Statistical

evidence was found with regard to the influence of convenience. The willingness to pay a price premium for sustainable products, and the purchase intentions for both healthy and sustainable products were positively influenced. The influence of convenience on the willingness to pay a price premium for healthy products was positive as well, but insignificant, so hypothesis 3a1 had to be

rejected. Hypotheses 3a2, 3b1, and 3b2 were all accepted.

Lastly, shopper types were included in the study to check if the need for different trends among shoppers varied. Significant positive correlations were found between utilitarian shoppers and consumer need for convenience. Likewise, positive significant relationships were found between hedonic shoppers and consumer need for experience. Hypothesis 4a and 4b were both accepted because of these results.

(37)

37

6. Discussion and Limitations

6.1 Discussion

People wrestle with decision difficulty on an almost daily basis, wanting one thing while knowing they should opt for another thing (Ehrich & Irwin, 2005). The aim of this study was to find how consumers can be nudged in making the decision that is the best for themselves, being healthy and sustainable products. The 184 respondents representing consumers were aged 38 and earned €2343 on average, at the time of the questionnaire. A side note is the relatively low income and the amount of well educated respondents. This result probably occurred due to many students filling in the survey. As the standard deviation of these findings is high and the variables had no significant influence throughout the sample, the external validity and generalizability of this study is quite good.

Five out of the fourteen hypotheses were accepted, leaving the other nine to be rejected. Five of the rejected hypotheses were insignificant, but indicated the right influence. This is probably due to the relatively small sample. The other four rejected hypothesis all dealt with the non-existent covariant consumer need for experience. Altogether, most of the expectations seemed to be confirmed, making this an interesting study.

The results indicate that the consumer need for convenience is a better tool to increase the sales levels of healthy and sustainable food products than the consumer need for experience. These results are partially in accordance with the expectations underlying the hypotheses. The finding that increasing convenience makes products more wanted is supported by marketing literature (Chandon & Wansink, 2002).The current study shows that convenience could function as an incentive for consumers to buy the product they know is the best choice for themselves. Translating the results into prices that consumers are willing to pay indicates: €1,34 for the base condition, €1,41 for the healthy package, €1,50 for the sustainable package, and €1,51 for the package with both healthy and sustainable elements. Considering the latter package, consumers appear to be willing to pay a 13% premium for products that are better for themselves and the environment. Utilitarian consumers appeared to have a bigger need for convenience than hedonic shoppers did. This makes sense as Batra and Ahtola’s (1990) characteristics of utilitarianism are quite similar to the benefits of convenience. However, as the Cronbach’s Alpha of this variable was relatively low, this relationship is somewhat untrustworthy. The low Cronbach’s Alpha might also be the reason that the correlation was only just significant with P = 0,095. Hedonism, on the other hand had a really high Cronbach’s Alpha, and a positive relation with need for experience.