Ways to display products in a supermarket that can contribute to a healthier society:

Ways to display products in a supermarket that can contribute to a

healthier society:

The influence of the Transparency of nutritional information and Assortment

structure on the Choice likelihood of healthier foods.

Gioia Langen

Ways to display products in a supermarket that can contribute to a healthier

society:

The influence of the Transparency of nutritional information and Assortment structure on

the Choice likelihood of healthier foods.

University of Groningen Faculty of economics and business

Marketing department Master Thesis

Author: Gioia Langen

Address: Celebesstraat 71A, Groningen Phone number: +316 43715024

Email: gioialangen@gmail.com Student number: S2909413

Supervisor: Prof. dr. ir. K. van Ittersum Second supervisor: J. J. Hidding (Msc.)

Abstract

The obesity rates in the world keep rising and people are getting more and more unhealthy. This research examines two interventions that can reduce the growth rate of obesity and the related global deaths, and potentially lead to a healthier society. The first intervention that is investigated in this paper is the provision of more transparent information about the nutritional content of products, due to a Nutri-score label. A Nutri-score label is a colour-coded front of package (FOP) label that indicates the healthiness of a product based on a score from a dark green A to a dark red E. It was expected that Transparency of nutritional information was positively related to the Choice likelihood of healthier foods, because more transparent nutritional information should make it easier for consumers to distinguish the healthy products from the unhealthy ones. The second intervention that has been researched in this paper is the structure of the assortment. It was expected that healthier products were chosen more often when placed above the unhealthy products, due to the primacy effect. Lastly, the moderating effect of Assortment structure on the relation between Transparency of nutritional information and the Choice likelihood of healthier foods was researched. It was expected that when the two interventions were combined (a structured assortment with visible Nutri-score labels), it becomes even easier for consumers to distinguish the healthy products from the unhealthy ones, which will enhance the effect of Transparency of nutritional information. A two-way ANOVA with interaction was performed to analyse these hypotheses. This analysis revealed that the direct effect of Transparency of nutritional information did significantly increase the choice likelihood of healthier foods. The direct and moderating effect of Assortment structure turned out insignificant. In additional ANOVA analyses, containing the variable General health interest, it turned out that Assortment structure did have a significant positive direct effect on the Choice likelihood of healthier foods for some of the participants. When the participants in this research were split up into two same sized groups based on their general health interest, Assortment structure turned out to have a significant effect on the Choice likelihood of healthier foods for people who have a relatively low general health interest.

Table of contents

1. Introduction 5

2. Theoretical framework 7

2.1 Transparency of nutritional information 8

2.2 Assortment structure 10

2.3 Moderating effect of Assortment structure 12

3. Methodology 13 3.1 Participants 14 3.2 Research design 14 3.3 Procedure 19 3.4 Analysis 20 4. Results 20 4.1 Data management 20

4.2 Two-way ANOVA with interaction across all eight product categories 22 4.3 Two-way ANOVA with interaction separately for the product categories 22

5. Discussion of the results 23

5.1 Direct effect Transparency of nutritional information 23

5.2 Direct effect Assortment structure 24

5.3 Moderating effect Assortment structure 25

6. Additional analyses General health interest 25

6.1 Two-way ANOVA with interaction 27

6.2 Two-way ANOVA with interaction, General health interest split 29 6.3 Discussion of the results of the additional analyses 30

7. Implications 31

8. Limitations and recommendations for future research 31

9. Conclusion 33

Reference List 34

Appendices 39

Appendix 1 39

1. Introduction

Today, one third of the world population is overweight or obese, twice as many as 40 years ago (von Lengerke & Krauth, 2011). The growing obesity rate is not without consequences, as according to the World Health Organization (2016) 2.8 million of the yearly global deaths are attributable to obesity. Furthermore, it also results in a lot of costs for society. Von Lengerke and Krauth (2011) found that in Europe, the costs regarding obesity account for 0.47-0.61% of the GDP, which again results in direct and indirect costs per individual ranging from €117 to €1873. One of the drivers of this rising number of people with obesity is the growth of unhealthy food choices (Vecchio & Cavallo, 2019).

Since the obesity rates and the corresponding costs for society keep rising, public health organisations, governments, consumers, academics and other interest groups are doing a lot of research to improve the health of individuals and try to come up with all kinds of interventions to promote healthier food choices (Vecchio & Cavallo, 2019). These interventions range from merely informing consumers about the healthiness of products to eliminating unhealthy choices entirely (Griffiths & West, 2015). Since 75% of our food comes from food stores, supermarkets are recognised as a key area for promoting healthy consumption and are therefore an interesting environment to conduct research in (Kalnikaitė et al., 2013; Milliron et al., 2012). For that reason, this paper will research two interventions in the context of a supermarket.

What is the effect of increasing the transparency of nutritional information via a FOP label on the choice likelihood of healthier foods?

Another intervention that is often used to improve the health of individuals is nudging. Nudging strategies are used to promote healthy eating behaviour and is, according to previous research, effective in doing so (Bucher et al., 2016). Nudging is defined as: changing the environment in such a way that consumers will be unconsciously steered into the direction of the healthier product, without changing the economic consequences or the number of available choices (Griffiths & West, 2015; Hertwig & Grüne-Yanoff, 2017). Nudging is a technique based on the theory that people are not fully capable of making informed choices (Griffiths & West, 2015). People often suffer from a gap between their intentions and actual behaviours. Most people have the intention to eat healthier, but they often fail in doing so (Burgoon et al., 2008; Griffiths & West, 2015). One of the reasons for this gap to occur in the context of a supermarket, is that people find it hard to distinguish the healthy products from the unhealthy ones due to the cluttered environment (Hagmann & Siegrist, 2020). One way of changing the environment in a supermarket that can serve as a nudge could be changing the assortment structure, so altering the way in which products are organized and presented to consumers. Previous research proved that visual attention, due to the positioning of the product, is correlated with product consideration (Chandon et al., 2016). The products that people encounter first, will receive more visual attention and therefore they are chosen more often than the products that people encounter later. Unfortunately, prior research is limited to the context of sales and neglects the importance of visual attention related to the choice for healthy products. Therefore, this paper will investigate if the same effect occurs for nudging people toward a healthier choice in a supermarket. Thereby, this paper aims for answering the following research question:

What effect does the assortment structure have on the choice likelihood of healthier foods?

the healthy products from the unhealthy ones. This is based on the representative heuristic which will be further explained in the theory section. The last research question of this paper is therefore:

Does the effect of the transparency of nutritional information via a FOP label on the choice likelihood of healthier foods depend on the assortment structure?

In this paper, the healthiness of a product will be operationalized based on the Nutri-score. The Nutri-score is a colour-coded FOP label that indicates the healthiness of a product based on a score from a dark green A to a dark red E (Julia & Hercberg, 2018). The Nutri-score is developed by the Programme National Nutrition Santé in France, as part of the France nutrition policy in 2014 (Julia et al., 2018; Julia & Hercberg, 2018). To calculate the score of a specific product the content of energy, fruit, vegetables and nuts, fibre, saturated fat, total sugar, sodium and protein is taken into consideration (Hagmann & Siegrist, 2020). The Nutri-score is reflecting the relative healthiness of the product choice compared to the other products in the category.

This research paper performs an experiment in the context of a supermarket. The original idea of this research was to make use of virtual reality technology, in order to create a realistic supermarket environment. Due to the Covid-19 pandemic, this method became impossible to execute, and for that reason, this research will proceed with an online questionnaire. This online questionnaire will simulate the effects of different factors that can influence the product choices that people make in an online supermarket. An online supermarket is an interesting area to conduct research in, since one fifth of the Dutch consumers is buying their groceries online at the moment and it is expected that this way of grocery shopping is only going to be more popular in the future (Smart Food Monitor, 2019).

The structure in this research paper is as follows. First, the theory section contains an overview of the theory of FOP labelling and assortment structure. Besides the expected direct effect of the two factors, the paper also examines the expected moderating effect of assortment structure. Thereafter, the methodology will be described, and the results will be presented. Followed by a discussion of the results, the practical implications, and the recommendations for future research. This paper ends with a conclusion, which will present a short summary of the research.

2. Theoretical framework

literature, the first two hypotheses will be articulated. The third part of the theory will contain the expected moderating effect of Assortment structure.

2.1 Transparency of nutritional information

Transparency of nutritional information, as well as its influence on the healthiness of product choices, has gained a lot of attention in academic articles in the past years. It is widely known that nutrition labels can foster the transparency of nutritional information. “Nutrition labels are designed to help consumers evaluate the healthiness of foods and to promote healthier food choices” (Hagmann & Siegrist, 2020, p. 1). Labelling a product in terms of healthiness can have a strong impact on the food choices consumers make (Irmak et al., 2011). Feunekes et al. (2008) state that labelling, regardless of its kind, will always contribute to a better understanding of the healthiness of products. As stated before, people suffer from a gap between their intention and behaviour in terms of healthiness (Burgoon et al., 2008). In the cluttered supermarket environment it is hard for people to distinguish the healthy products from the unhealthy ones (Hagmann & Siegrist, 2020). Providing people with a label that displays how healthy a product is, will improve the transparency and hence lead to a better understanding of the healthiness of the products. It will therefore diminish the intention-behaviour gap, which results in healthier food choices by consumers (Crosetto et al., 2017; Julia et al., 2015).

nutritional information provided and interpretive labels also provide an evaluation of the nutritional information. Reductive labels are able to offer nutritional information in a way that is less complex, more condensed and presented at a better accessible location compared to the NFP labels (Ikonen et al., 2019). However, several researchers concluded that this type of labelling is still time consuming and difficult to interpret for consumers, since the label does not evaluate the information in terms of healthiness (Hawley et al., 2013; Hersey et al., 2013; Talati et al., 2017). Depending on the degree of information aggregation, Ikonen et al (2019) divide the labelling category interpretive labels again into two separate categories: nutrient-specific and summary indicator. Whereby interpretive nutrient-specific labels contain an evaluation on the healthfulness of one or more individual nutrients of a product and interpretative summary indicator labels provide an evaluation of the overall healthiness of a product (Ikonen et al., 2019). The interpretive nutrient-specific labels are known to help consumers process and interpret information about the individual nutrients (Ikonen et al., 2019). This facilitates the consumers’ understanding about the individual ingredients a product contains, but still requires consumers to combine multiple sources of information to determine the product’s overall healthiness (Talati et al., 2017). Above that, consumers may overgeneralize the claims about individual nutrients to the overall healthfulness of the product. Andrews et al. (2000) studied the misleading effect of nutrition specific claims and confirmed overgeneralization of information. In their study, participants assumed that a low-salt product would also have more beneficial levels of other nutrients, while that was not the case. A summary indicator helps to overcome this issue, by summarizing the overall healthiness of a product into one indicator (Ikonen et al., 2019). This makes these labels easy to read and understand, and especially helpful for consumers who want to compare multiple alternative products based on healthiness at the point of purchase (Liem et al., 2012; Newman et al., 2018).

Hercberg, 2018). As shown in Figure 1, the label shows all five letters, but the one indicating the healthiness of the product is expanded. This score is calculated based on the product’s content of energy, fruit, vegetables and nuts, fibre, saturated fat, total sugar, sodium and protein (Hagmann & Siegrist, 2020).

Figure 1 - Nutri-score Based on the literature of FOP labelling the first hypotheses is defined as:

Hypothesis 1: The placement of the Nutri-score (vs no label) on products in a supermarket will positively affect the Choice likelihood of healthier foods.

2.2 Assortment structure

In the marketing and sales context, shelf management is frequently studied. Manufacturers want to allocate their brands to optimise product sales and retailers want to allocate shelf space to maximise category sales (Drèze et al., 1994). This paper will use this literature to come up with an assortment structure that is most efficient for the consumers in terms of healthiness, instead of the best pattern for creating sales for the manufacturer or the retailer.

option. Besides, they found that the volume of the consumption of healthier food increased when it was placed more to the left (Romero & Biswas, 2016).

Second, product assortments in supermarkets can be structured from the middle (i.e. horizontal centre) towards the outside of the shelf. Various studies show that people prefer a product that is placed towards the centre. Chandon et al. (2009) studied this type of structuring for pain relievers and soap bars in supermarkets. They found that the attention of consumers was more focussed towards the centre, which increased product notion by 22% and product choice by 17%. Moreover, in the research towards middle choice preference, Keller et al. (2015) studied consumer choice between three similar snacks in a row with different calorie contents. They found the same effect: the apple bar with the lowest calorie content was picked almost three times more often when it was placed in the middle (compared to the outside) (Keller et al., 2015). The eye-tracking study of Chandon et al. (2006) confirms this line of thinking, which shows that products near the centre of the shelf are significantly better noted by consumers than products located further away from the centre.

Third and final, the product assortment can be structured top to bottom. This is a way of structuring that has been researched a lot in the context of sales, but the literature is lacking about this in terms of nudging people towards a more healthier choice in a supermarket (Bucher et al., 2016; Hollands et al., 2013). Multiple studies found that products positioned higher on the supermarket shelf will be chosen more often compared to products lower on the shelf. For example, Drèze et al. (1994, p. 324) found that: “A couple of facings at eye level did more for a product than five facings on the bottom shelf”. Multiple eye-tracking studies confirm this effect. For example, Chandon et al. (2009) found that products located at the top of the shelf were noted 17% more often than products located at the bottom, and increased the chance of choosing the product by 20%. Attention is the main reason for this increase. People tend to look first at the shelves at eye level, therefore these shelves get more visual attention and subsequently these products are chosen more often (Chandon et al., 2009). The reason that attention can have such a big impact on purchase behaviour in a supermarket is due to the fact that there is a very low level of customer involvement for these products (Drèze et al., 1994). Choices in a supermarket are made very quickly, without a lot of research and often based on heuristics (Hoyer, 1984; Scheibehenne et al., 2007).

by Chandon et al. (2009), several other articles state that attention is positively related to choice (Russo et al., 1994; Tan & Ward, 2000).

If these theories are linked with the literature of nudging it is expected that placing healthier products on the spot where consumers tend to look first, will lead to an increase of consumers choosing these products. Visual attention for the healthier products is increased, which is expected to increase the preference for these products. In fact, the environment is changed in such a way that people are (unconsciously) steered towards healthier products. In an online supermarket, products are generally listed, and people scroll down their screen to see all the products in the category. Therefore, the first two structures mentioned are less executable in an online environment compared to the ‘top to bottom’ structure. For that reason, this research will examine the top to bottom assortment structure. When structuring the products from top to bottom in terms of healthiness, consumers will encounter the healthier products first, which is expected to result in more attention and an increased choice likelihood of healthier foods. This leads to the second hypothesis:

Hypothesis 2: Structuring supermarket assortments from top to bottom (vs randomly), in terms of healthiness, will increase the Choice likelihood of healthier foods.

2.3 Moderating effect of Assortment structure

As mentioned in section 2.1, Nutri-scores improve the ease of distinguishing healthy products from unhealthy ones. It serves as a way to simplify the information that needs to be processed to determine the healthiness of products. When combining this knowledge with the above-mentioned literature about assortment structure, it is expected that the assortment structure enhances the labelling effect as well. Therefore, a combination of the two interventions would further improve the ease of distinguishing healthy and unhealthy products. Hypothetically, by structuring the supermarket assortment from top to bottom, in terms of healthiness, and by including the related Nutri-scores, it is also visible to the consumer on what basis the structure is formed. This will result in the distribution of the products into categories. These categories will be based on the five different levels of the Nutri-score. The upper part of the assortment will only display A-products, the part below that will only display B-products, this will go on until the lowest vertical position in the assortment, where the E-products are displayed (if all categories are present).

between people’s environment and the mental representation of that environment will make the information easier to process (Romero & Biswas, 2016). It is thus expected that when products are displayed into categories, based on the Nutri-score, the information about the healthiness of products is easier to process. This will make it easier for consumers to differentiate the healthy products from the unhealthy ones. By organizing the assortment based on the Nutri-score, it is expected that the consumers will be better at interpreting the nutritional information and therefore able to make more healthier choices. This leads to the following final hypothesis:

Hypothesis 3: The effect of Transparency of nutritional information, due to Nutri-score FOP labelling, on the Choice likelihood of healthier foods will increase when supermarket assortments are structured from top to bottom (vs randomly), in terms of healthiness.

The conceptual model is shown in figure 2 below.

Figure 2 – Conceptual model

3. Methodology

This study investigates the relationship between three variables. The independent variables ‘Transparency of nutritional information’ and ‘Assortment structure’ are expected to positively influence the dependent variable ‘Choice likelihood of healthier foods’. Besides, the moderating effect of assortment structure will be researched.

Transparency of nutritional information:

Visibility of the Nutri-score (visible vs not visible) Choice likelihood of healthier foods

3.1 Participants

209 respondents participated in this study. The average age of the participants was 32 years old (M age = 31,78, SD age = 15,332). The ages of the participants ranged from 16 to 81 years old, with most participants aged between 20 and 25 years old (58,4%). Only 7,2% of the participants were younger than 21 years old, and the remaining 34,4% of the participants had an age that was evenly spread between 26 and 81 years old. Unfortunately, due to some technical issues, the gender of the participants is not recorded, therefore it is not possible to draw conclusions about the gender of the participants. Regarding level of education, most of the participants finished higher level of education. Namely, 78% of the participants has finished a university or university of applied science degree. Furthermore, all of the participants visit a physical supermarket for their groceries, either multiple times a week (56%), once every week (20,1%), every day (12,4%), or less than once a week (11,5%). Lastly, most of the participants never shop their groceries online (82,8%), from all participants only 9,6% is shopping their groceries online once a month or more often, the remaining 7,6% does shop some their groceries online but less than once a month.

3.2 Research design

Within this research, the collection of data has been done via an online questionnaire on the platform Qualtrics (Appendix 1). Participants were assigned to one of four conditions based on two factors: Nutri-score (visible vs not visible) and Assortment structure (random vs vertically structured based on healthiness). Based on this, four different conditions were formed. The different conditions and their characteristics are shown in Table 1 below.

Condition Nutri-score visible

(yes/no)

Assortment structure

(randomly/structured)

Nutri-score condition (n=53) Yes Randomly

Structure condition (n=52) No Structured

Nutri-score + structure condition (n=52) Yes Structured

Control condition (n=52) No Randomly

Table 1 – Condition descriptions

choices, each from a different product category. The sequence in which the participants got to see the product categories was randomized between the participants, to control for bias resulting from the influence of the sequence in which participants did see the product categories. Within each product category, the participants had to pick one product out of the 15 products. The products were displayed in five rows of three products. Lastly, efforts were made to control for the effects of pricing and brand preference. To reduce the effect of pricing, cheap, moderate, and expensive products were represented in every Nutri-score level, for every product category. To reduce the effect of brand preference, brands were only displayed one time per category (to the best of efforts). The following eight categories were used: snacks, drinks, breakfast cereals, desserts, products from the bakery, cookies, (rice)crackers, and crisps. These categories are chosen because they consist of products within all the five levels of the Nutri-score. Besides that, these products are all ready to eat and preparation is not needed. This means the nutritional value (thus the Nutri-score) of these products does not differ between the moment of buying and the moment of consuming.

The healthiness of the product choice is operationalized by using the Nutri-score level of the choice. That score is reflecting the relative healthiness of the product choice compared to the other products in the category. Product choices with a Nutri-score of A were labelled as 1, products with a Nutri-score of B were labelled as 2, up until products with a Nutri-score of E, that were labelled as 5. Thus, the lower the score, the healthier the product choice. The variable Transparency of nutritional information is manipulated by displaying the Nutri-score on each product vs. displaying the products without the Nutri-score. Figure 3 below shows an example of the visualization of the products with and without a Nutri-score.

Example product with a Nutri-score Example product without a Nutri-score

The manipulation of the assortment structure was created by the order in which the products were displayed to the participants (structured vs random). In the structured condition, people did see the products in a certain sequence, based on the healthiness of the products in terms of the Nutri-score. The first row contained only products with the Nutri-score A, up until the fifth row, where only products with an E label were displayed. Table 5 shows that type of product

displaying. For participants in the random structure condition, the products were displayed in a random order.

Product with Nutri-score A Product with Nutri-score A Product with Nutri-score A Product with Nutri-score B Product with Nutri-score B Product with Nutri-score B Product with Nutri-score C Product with Nutri-score C Product with Nutri-score C Product with Nutri-score D Product with Nutri-score D Product with Nutri-score D Product with Nutri-score E Product with Nutri-score E Product with Nutri-score E

Table 2 - Structured shelve based on healthiness

Assortment example cookies Nutri-score condition (order of displaying was randomized between participants)

Assortment example cookies structure condition

Assortment example cookies Nutri-score + structure condition

Assortment example cookies Control condition (order of displaying was randomized between participants)

Figure 4.2 – Visualization cookie assortment questionnaire

supermarket in The Netherlands. Besides that, also the prices of the products were identical to the prices at Albert Heijn. Furthermore, the participants were informed about their chance of winning the eight products of their choice, which contributes to a realistic shopping motivation among the participants. Lastly, the participants read an unrelated cover story about the purpose of the research. This will exclude that people behave differently because they have knowledge about the purpose of the experiment. The cover story explains that the purpose of the research is to collect information about consumer preferences in a supermarket.

3.3 Procedure

3.4 Analysis

At the beginning of this paper, three research questions have been formulated and based on previous findings in the literature several hypotheses were formed. The goal of this paper is to measure the effects of the two independent variables: Transparency of nutritional information and Assortment structure, on the dependent variable: Choice likelihood of the healthier foods and besides that, to find out if Assortment structure also enhances the effect of Transparency of nutritional information as a moderator.

The three relations between the variables were analysed using a two-way ANOVA with interaction. This test determines whether there are any statistical differences between the means of the different conditions in this research. This test is used because the two independent variables (Transparency of nutritional information and Assortment structure) have a nominal scale and the dependent variable (Choice likelihood of healthier foods) is measured on a ratio scale. Since this research examines the effects of two independent variables, whereby one of them is also expected to enhance the effect as a moderator, a two-way ANOVA with interaction was performed. The reason for the statistical fit of a two-way ANOVA with interaction to this study is that it comprises a first order single moderator with a dichotomous variable.

4. Results

This chapter will give an overview of the results of the analyses that were performed to test the hypotheses of this research paper.

4.1 Data management

incorrect patters found in the dataset. Two participants solely chose products with an A Nutri-score, but there is no reason to assume that this data would be incorrect.

A two-way ANOVA with interaction demands six requirements to be met. First, the dependent variable should be measured at a continuous scale. Since the dependent variable in this research: Choice likelihood of healthier foods, is operationalized on a scale from 1 to 5 (1=most healthy choice, 5=most unhealthy choice), this requirement is met. Second, the two independent variables consist of two or more categorical independent groups, since both independent variables are dichotomous (Transparency of nutritional information: Nutri-score or no Nutri-score & Assortment structure: structured assortment or random assortment), the second requirement is met as well. Moreover, since there is no relationship between the observations in each group or between the groups themselves, the third requirement that observations should be independent is met. To meet the last three requirements of a two-way ANOVA, three additional tests were performed. To ensure that the data contains no significant outliers, a boxplot analysis was performed. The Shapiro-Wilk test was performed to test whether the data is normally distributed, which is a requirement for a two-way ANOVA. The p values of the Shapiro-Wild test for the different conditions are the following: Nutri-score: p=0,450; No Nutri-score: p=0,698; Structure; p=0,242; Random: p=0,342. Therefore, this requirement is met, the data is normally distributed. The last additional test that has been performed is a Levene’s test for homogeneity of variances. This test confirmed homogeneity of variances since p was 0,138. Hereby, all the requirements to conduct a two-way ANOVA with interaction are met.

Details on the mean healthiness of the product choices for the different conditions are displayed in Table 3 below. The healthiness score is calculated based on the average healthiness across all product categories. Participants within the Nutri-score + structure condition chose the healthiest products (M = 2,40500; SD = 0,747257) compared to the participants in other conditions. As mentioned above, healthiness of the product choice is operationalized by using the Nutri-score label whereby the healthiest product choices are labelled as 1, up until the unhealthiest choices that are labelled as 5. Thus, the lower the score, the healthier the product choice. More detailed results of the analyses will be described in the following chapters.

Condition Control (n=53) Nutri-score (n=52) Structure (n=52) Nutri-score + Structure (n=52) Mean 2,882 2,491 2,716 2,405 St. Deviation 0,686 0,612 0,545 0,747 Minimum 1,630 1,250 1,380 1,000 Maximum 4,500 3,880 3,880 4,130

4.2 Two-way ANOVA with interaction across all eight product categories

To analyse the direct and moderating effects of the variables a two-way ANOVA with interaction was performed. The healthiness of the product choice was calculated by the average healthiness of the eight product choices of a participant. Across all eight product categories, it turned out that the variable Transparency of nutritional information had a significant effect on the Choice likelihood of healthier foods, F (1,205)= 15,158, p = 0,000. Therefore, the first hypothesis is accepted. Besides, the direct effect of the variable Assortment structure turned out to insignificantly affect the Choice likelihood of healthier foods, F (1,205)= 1,959, p = 0,163. Hence, the assortment structure does not positively influence the choice likelihood of healthier foods and hypothesis 2 is rejected. Lastly, hypothesis 3 is rejected as well, since the moderating effect of assortment structure on the relation between Transparency of nutritional information and the Choice likelihood of healthier foods was insignificant, the interaction effect was insignificant, F (1,205)= 0,199, p = 0,656. See Figure 5 for a visualization of the results.

Figure 5 – Estimated marginal means healthiness of the product choice

4.3 Two-way ANOVA with interaction separately for the product categories

To examine the direct and moderating effect of the variables to a further extend, the two-way ANOVA with interaction was performed for all the product categories separately. A significant

0,000 1,000 2,000 3,000 Random Structured Es ti m at ed m ar gin al m ea n s h ea lt h in es s p rodu ct choi ce ( 1 = m os t he alt hy , 5 = m os t un he alt hy ) Assortment structure

Choice likelihood of healthier foods

No Nutri-score Nutri-score

product categories: snacks; F(1,205)= 11,304, p = 0,001, drinks; F(1,205)= 8,575, p = 0,004, desserts; F(1,205)= 12,171, p = 0,001 and (rice)crackers; F(1,205)= 4,232, p = 0, 041. The Assortment structure turned out to only have a significant effect on the Choice likelihood of healthier foods in the product category bakery, F(1,205)= 5,774, p = 0,017. Lastly, the moderating effect was only marginal significant for the separate product category (rice)crackers, since the interaction effect was marginal significant for that product category, F(1,205)= 3,682, p = 0,056. Table 4 presents an overview of the p values for all product categories in this two-way ANOVA with interaction analysis.

Category / Variable Transparency of nutritional information

Assortment structure Interaction effect

Across all 8 categories 0,000** 0,163 0,656

Snacks 0,001** 0,689 0,834 Drinks 0,004** 0,119 0,664 Ontbijtgranen 0,261 0,836 0,171 Desserts 0,001** 0,556 0,817 Bakery 0,884 0,017** 0,200 Cookies 0,177 0,303 0,252 (Rice)crackers 0,041** 0,401 0,056* Crisps 0,293 0,489 0,526

* P value <0,10 = marginal significant ** P value <0,05 = significant

Table 4 – All P values two-way ANOVA with interaction analysis

5. Discussion of the results

This chapter provides a discussion on the results of the analyses described above. The main goal of this research paper was to examine three research questions: (1) What is the effect of increasing the transparency of nutritional information via a FOP label on the choice likelihood of healthier foods? (2) What effect does the assortment structure have on the choice likelihood of healthier foods? (3) Does the effect of the transparency of nutritional information via a FOP label on the choice likelihood of healthier foods depend on the assortment structure?

5.1 Direct effect Transparency of nutritional information

Therefore, hypothesis 1 is accepted. This is line with previous research on FOP labelling and more specific, the Nutri-score label. Research revealed that FOP labelling positively affects the healthiness of the product choice (Hagmann & Siegrist, 2020; Irmak et al., 2011; Hagmann & Siegrist, 2020; Feunekes et al., 2008). Furthermore, the Nutri-score label was stated to be the most effective label in doing so (Ducrot et al., 2015; Egnell et al., 2018; Hagmann & Siegrist, 2020). This research is in line with previous literature, it replicates the results of previous studies and therefore contributes to the literature of antecedents of healthy eating behaviours. Using a Nutri-score FOP label will increase the choice likelihood of healthier foods and can therefore be used as an intervention to increase the health of individuals.

5.2 Direct effect Assortment structure

work with others. The consumer characteristic General health interest is often suggested as one that could determine the effectiveness of health interventions (e.g. Mai & Hoffmann, 2012; Wansink, 2017). To investigate this, this paper will perform additional analyses in chapter 6.

5.3 Moderating effect Assortment structure

Lastly, this research did not find a significant moderating effect of Assortment structure on the relation between Transparency of nutritional information and the Choice likelihood of healthier foods. There is no statistical evidence to proof that the effect of the Transparency of nutritional information via a FOP label on the Choice likelihood of healthier foods depends on the Assortment structure. Therefore, the third hypothesis needs to be rejected. Only for the specific product category (rice)crackers, a marginal significant moderating effect was found. This could indicate that Assortment structure indeed serves as a moderator, but only for specific product categories. Since this research examined only eight product categories, it might be that the moderating effect occurs in product categories that were not part of this study. The overall findings of this research about the moderating effect of assortment structure are not in line with the literature about representative heuristics. It was expected that when products were displayed into categories, based on the Nutri-score label, the information about the healthiness of the products was easier to process, which in turn should have resulted in easier interpretation of nutritional information and therefore healthier product choices (Romero & Biswas, 2016). A possible explanation for this could be that 15 products are not an abundance of information for consumers. Since the effect is expected to occur on the basis that people split up their environment into categories to deal with an abundance of information (Rosch & Mervis, 1975; Chen et al., 2010), it might be that this effect will only occur when consumers are confronted with more products at the same time. This is the case in a real-life supermarket, where product shelves consist of more than 15 products. Future research could examine the moderating effect of assortment structure, whereby product shelves with more products should be presented to the consumers.

6. Additional analyses General health interest

General health interest is defined by Mai and Hoffmann (2012, p. 317) as: “The motivational

healthier foods. As mentioned in the theory section, it was expected that the Assortment structure could have such a big impact on customer behaviour in supermarkets, because consumers are little involved with this buying process. It is also known that consumers with a high General health interest will actively engage in healthy eating (Michaelidou & Hassan, 2008). It is therefore expected that this group of consumers is more involved with the process of buying groceries which could reduce the effect of Assortment structure. It is expected that the variable Assortment structure only affects the Choice likelihood of healthier foods for consumers who are less interested in their general health.

To examine the effect of the variable General health interest, the variable was added to the previous two-way ANOVA with interaction. The variable General health interest was computed out of four multi-item scaling questions in the questionnaire (Roininen et al., 1999). In the reliability analysis, the variable had a Cronbach’s Alpha of 0,770 and because the Alpha would only drop if one of these four questions would be left out, the variable General health interest was computed by the average of the four questions. Since the ANOVA requires the data to have categorical independent variables, the variable General health interest was converted into two same sized groups: low general health interest and high general health interest. This split was made based on the median (3,5) to create same sized groups. All participants with a General health interest below 3,5 were labelled as having a low general health interest (n=101). All participants with a general health interest of 3,5 and above were labelled as having a high general health interest (n=108).

6.1 Two-way ANOVA with interaction

Figure 6 – Visualization two-way interaction

Figure 7 – Visualization three-way interaction, General health interest low

2,000 2,200 2,400 2,600 2,800 3,000 3,200 3,400 Random Structured Es ti m at ed m ar gi n al m ea n s h ea lt h in es s p ro d u ct ch oic e (1 = m os t h ea lt h y, 5 = m os t u nh ea lt h y) Assortment structure

Choice likelihood of healthier foods

Low High General health interest 0,000 1,000 2,000 3,000 4,000 Random Structured Es ti m at ed m ar gi n al m ea n s he alt hi n es s pr oduc t ch oi ce ( 1= m os t h ea lt h y, 5= m os t u n h ea lt h y) Assortment structure

Choice likelihood of healthier foods at General health interest (median split): low

No Nutri-score Nutri-score

Figure 8 – Visualization three-way interaction, General health interest high

6.2 Two-way ANOVA with interaction, General health interest split

To further analyse the marginal significant moderating effect of General health interest, a two-way ANOVA with interaction was performed separately for participants with a low general health interest and for participants with a high general health interest. The mean healthiness of the product choices in different conditions for participants with high and low general health interest is visualized in Table 5 below.

The effectiveness of the variable Assortment structure did indeed depend on the General health interest of the participants. In line with expectations, the variable Assortment structure turned out to significantly influence the Choice likelihood of healthier foods for participants with a low general health interest, F (1,97)= 4,074, p = 0,046. For the participants with a high general health interest, this variable was insignificant, F (1,104)= 0,156, p = 0,693. For both analyses, Transparency of nutritional information was significant. For the low general health interest group, Transparency of nutritional information turned out to be significant, F (1,97)= 7,751 p = 0,006; and this variable was also significant for the high general health interest group, F (1,104)= 7,214 p = 0,008. Participants (with both low and high general health interest) who saw the products with a Nutri-score label made significant healthier food choices than participants who did not see the label. The interaction effect, to examine the moderating effect, was again insignificant for both

0,000 1,000 2,000 3,000 Random Structured Es ti m at ed m ar gi n al m ea n s he alt hi n es s pr oduc t ch oi ce ( 1= m os t h ea lt h y, 5= m os t u n h ea lt h y) Assortment structure

Choice likelihood of healthier foods at General health interest (median split): high

No Nutri-score Nutri-score

insignificantly enhance the effect of Transparency of nutritional information, F (1,97)= 0,005, p = 0,816. For the high general health interest group, the interaction effect was also insignificant, F (1,104)= 0,013, p = 0,909.

General health interest low

Condition Control (n=26) Nutri-score (n=22) Structure (n=23) Nutri-score + Structure (n=30) Mean 3,163 2,801 2,898 2,598 St. Deviation 0,653 0,514 0,357 0,731 Minimum 2,130 2,000 2,380 1,130 Maximum 4,500 3,880 3,380 4,130

General health interest high

Condition Control (n=29) Nutri-score (n=29) Structure (n=27) Nutri-score + Structure (n=23) Mean 2,516 2,207 2,578 2,180 St. Deviation 0,550 0,557 0,624 0,735 Minimum 1,630 1,250 1,380 1,000 Maximum 3,630 3,500 3,880 4,000

Table 5 - mean healthiness of the product choices in different conditions (for high and low general health interest)

6.3 Discussion of the results of the additional analyses

In line with expectations, General health interest positively influences the Choice likelihood of healthier foods. Michaelidou & Hassan (2008) found in their research that people who are more interested in their general health will be more motivated to consume healthy foods, this study replicates these findings.

customer involvement is not taken into consideration within this research, future research should determine if customer involvement is indeed the underlying variable that explains this effect.

All the other results from the additional analyses were in line with previous analyses. The direct effect of Transparency of nutritional information was still significant and the moderating effect of Assortment structure on the relation between the Transparency of nutritional information and the Choice likelihood of healthier foods was still insignificant.

7. Implications

The findings of this research are relevant for different parties, such as governments, insurance companies, supermarkets, and marketeers. As mentioned in the introduction, nutritional diseases such as obesity, are responsible for a lot of global deaths and result in a lot of societal costs. When interventions provoke people to make healthier product choices, this will result in a healthier society, which subsequently will reduce the societal costs as well. This research found statistical evidence for the fact that consumers make healthier food choices when confronted with products with a Nutri-score label, compared to consumers who are confronted with products without a Nutri-score label. Above that, we found evidence that a structured online supermarket assortment, from top to bottom in terms of healthiness, will also lead to healthier food choices among consumers who have a relatively low general health interest. Assortment structure did not lead to a significant healthier food choice among participants with a high general health interest, but there is also no statistical evidence that proofs that it did make the choices of these participants less healthy. Therefore, implementing both interventions would lead to a healthier society.

For governments and insurance companies, these findings can contribute to the research they are already performing in the area of health interventions. Since the Nutri-score label and a structured assortment will lead to healthier food choices it would make sense to investigate the possibilities to implement these interventions on a national scale. When these interventions are available in the marketplace, this will also have an impact on marketeers of food producing companies. Since consumers’ preference for healthy foods will increase when there is more transparent nutritional information available, marketeers should respond to this by increasing the availability of healthier foods in the marketplace.

8. Limitations and recommendations for future research

due to the Covid-19 pandemic this was no longer possible. Therefore, the purpose of this research was changed from measuring the effects of the two interventions for physical grocery stores to measuring the effects for online grocery stores. Since most of the consumers are still shopping their groceries in a physical supermarket, it is interesting to implement the interventions in that area as well. The effects are expected to occur in a physical supermarket, but further analysis is necessary to determine any conclusions about it. This could for example be done by using virtual reality technology to create a realistic supermarket environment or by conducting observational research in a real-life supermarket setting.

Second, the following factors might have influenced the representativeness of this research too: ages and education level. As mentioned, 78% of the participants had finished higher level of education (university / university of applied science) and 58,4% of the participants was aged between 20 and 25 years old. This young and highly educated sample is probably the effect of the snowball effect that was used to spread the questionnaire. Unfortunately, this is not a correct representation of the Dutch society and this might have an effect on the representativeness of the results. Especially, since educational level is a predictor of income, and income could have an effect on the healthiness of the product choices that people make (Pesta et al., 2012). There is evidence that proves that people with a higher social economic status follow healthier diets compared to people with less income (Power, 2005). This might have therefore positively influenced the Choice likelihood of healthier foods for these participants. Future research should take this into consideration when conducting research in this area.

Third, as mentioned in the methodology, the influence of pricing and brand preference was reduced in the questionnaire to the best efforts. Yet, despite the fact that this was possible for almost every product category, some categories did contain a specific brand more often than once or could not display a cheap, moderate, and expensive product for each Nutri-score level within the assortment. This might have weakened the effects of Transparency of nutritional information and Assortment structure on the choice likelihood of healthier foods because participants’ preference for brands or prices might have influenced their product choice. Future research could investigate this, to reinforce the results of this research.

The fourth limitation lies with the representativeness of the assortments. Within this research, every assortment only contained of 15 products. A real supermarket assortment contains often more products than 15. Moreover, this research investigated eight product categories. Future research could replicate this study with more products in each category and other categories could be investigated as well.

they chose as part of the research. This contributed to a realistic shopping motivation among the participants, but there might be also a chance that some respondents chose products they would like to win, but normally not buy (e.g. more expensive products). Since the respondents did not have to check out the products as they would normally do in a supermarket, it might be that they chose different products. To overcome this, future research could perform this experiment in a real-life shopping setting, for example by conducting observational research.

9. Conclusion

Reference List

Andrews, J. C., Burton, S., & Netemeyer, R. G. (2000). Are some comparative nutrition claims misleading? The role of nutrition knowledge, ad claim type and disclosure conditions. Journal of Advertising, 29(3), 29-42.

Anesbury, Z., Nenycz‐Thiel, M., Dawes, J., & Kennedy, R. (2016). How do shoppers behave online? An observational study of online grocery shopping. Journal of Consumer

Behaviour, 15(3), 261-270.

Bucher, T., Collins, C., Rollo, M. E., McCaffrey, T. A., De Vlieger, N., Van der Bend, D., Truby, H., & Perez-Cueto, F. J. A. (2016). Nudging consumers towards healthier choices: a systematic review of positional influences on food choice. The British Journal of Nutrition, 115(12), 2252–2263.

Burgoon, J. K., Pete Blair, J., & Strom, R. E. (2008). Cognitive Biases and Nonverbal Cue Availability in Detecting Deception. In Human Communication Research 34(4), 572–599. Byrd‐Bredbenner, C., Wong, A., & Cottee, P. (2000). Consumer understanding of US and EU

nutrition labels. British Food Journal, 102(8), 615–629.

Chae, B., & Hoegg, J. (2013). The Future Looks “Right”: Effects of the Horizontal Location of Advertising Images on Product Attitude. Journal of Consumer Research, 40(2), 223–238. Chandon, P., & Wansink, B. (2012). Does food marketing need to make us fat? A review and

solutions. Nutrition Reviews, 70(10), 571–593.

Chandon, P., Wesley Hutchinson, J., Bradlow, E. T., & Young, S. H. (2009). Does In-Store Marketing Work? Effects of the Number and Position of Shelf Facings on Brand Attention and Evaluation at the Point of Purchase. Journal of Marketing, 73(6), 1–17.

Chandon, P., Wesley Hutchinson, J., Bradlow, E., & Young, S. H. (2016). Measuring the Value of Point-of-Purchase Marketing with Commercial Eye-Tracking Data. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.1032162.

Chen, Y., Iyer, G., & Pazgal, A. (2010). Limited Memory, Categorization, and Competition. Marketing Science, 29(4), 650–670.

Cohen Kadosh, R., Brodsky, W., Levin, M., & Henik, A. (2008). Mental representation: what can pitch tell us about the distance effect? Cortex; a Journal Devoted to the Study of the Nervous System and Behavior, 44(4), 470–477.

Cowburn, G., & Stockley, L. (2005). Consumer understanding and use of nutrition labelling: a systematic review. Public Health Nutrition, 8(1), 21–28.

Drèze, X., Hoch, S. J., & Purk, M. E. (1994). Shelf management and space elasticity. Journal of Retailing, 70(4), 301–326.

Ducrot, P., Méjean, C., Julia, C., Kesse-Guyot, E., Touvier, M., Fezeu, L., Hercberg, S., & Péneau, S. (2015). Effectiveness of Front-Of-Pack Nutrition Labels in French Adults: Results from the NutriNet-Santé Cohort Study. PloS One, 10(10).

Egnell, M., Talati, Z., Hercberg, S., Pettigrew, S., & Julia, C. (2018). Objective Understanding of Front-of-Package Nutrition Labels: An International Comparative Experimental Study across 12 Countries, Nutrients. 10(10).

Feunekes, G. I. J., Gortemaker, I. A., Willems, A. A., Lion, R., & van den Kommer, M. (2008). Front-of-pack nutrition labelling: testing effectiveness of different nutrition labelling formats front-of-pack in four European countries. Appetite, 50(1), 57–70.

Griffiths, P. E., & West, C. (2015). A balanced intervention ladder: promoting autonomy through public health action. Public Health, 129(8), 1092–1098.

Hagmann, D., & Siegrist, M. (2020). Nutri-Score, multiple traffic light and incomplete nutrition labelling on food packages: Effects on consumers’ accuracy in identifying healthier snack options. Food Quality and Preference, 83, 103894.

Hawley, K. L., Roberto, C. A., Bragg, M. A., Liu, P. J., Schwartz, M. B., & Brownell, K. D. (2013). The science on front-of-package food labels. Public Health Nutrition, 16(3), 430–439. Hayes, A. F. (2017). Introduction to Mediation, Moderation, and Conditional Process Analysis,

Second Edition: A Regression-Based Approach. Guilford Publications.

Hersey, J. C., Wohlgenant, K. C., Arsenault, J. E., Kosa, K. M., & Muth, M. K. (2013). Effects of front-of-package and shelf nutrition labeling systems on consumers. Nutrition Reviews, 71(1), 1–14.

Hertwig, R., & Grüne-Yanoff, T. (2017). Nudging and Boosting: Steering or Empowering Good Decisions. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 12(6), 973–986.

Hollands, G. J., Shemilt, I., Marteau, T. M., Jebb, S. A., Kelly, M. P., Nakamura, R., Suhrcke, M., & Ogilvie, D. (2013). Altering micro-environments to change population health behaviour: towards an evidence base for choice architecture interventions. BMC Public Health, 13, 1218. Hoyer, W. D. (1984). An Examination of Consumer Decision Making for a Common Repeat

Purchase Product. In Journal of Consumer Research, 11(3), 822.

Ikonen, I., Sotgiu, F., Aydinli, A., & Verlegh, P. W. (2019). Consumer effects of front-of-package nutrition labeling: an interdisciplinary meta-analysis, Journal of the Academy of Marketing Science, 1-24.

Jayanti, R. K., & Burns, A. C. (1998). The antecedents of preventive health care behavior: An empirical study. Journal of the academy of marketing science, 26(1), 6-15.

Julia, C., Etilé, F., & Hercberg, S. (2018). Front-of-pack Nutri-Score labelling in France: an evidence-based policy. The Lancet Public Health, 3(4), 164.

Julia, C., & Hercberg, S. (2018). Big Food’s Opposition to the French Nutri-Score Front-of-Pack Labeling Warrants a Global Reaction. American Journal of Public Health, 108(3), 318–320. Julia, C., Kesse-Guyot, E., Ducrot, P., Péneau, S., Touvier, M., Méjean, C., & Hercberg, S. (2015). Performance of a five category front-of-pack labelling system – the 5-colour nutrition label – to differentiate nutritional quality of breakfast cereals in France. BMC Public Health, 15(1). Kalnikaitė, V., Bird, J., & Rogers, Y. (2013). Decision-making in the aisles: informing,

overwhelming or nudging supermarket shoppers? Personal and Ubiquitous Computing, 17(6), 1247–1259.

Keller, C., Markert, F., & Bucher, T. (2015). Nudging product choices: The effect of position change on snack bar choice. Food Quality and Preference, 41, 41–43).

Liem, D. G., Miremadi, F., Zandstra, E. H., & Keast, R. S. J. (2012). Health labelling can influence taste perception and use of table salt for reduced-sodium products. Public Health Nutrition, 15(12), 2340–2347.

Iacobucci, D., Posavac, S. S., Kardes, F. R., Schneider, M. J., & Popovich, D. L. (2015). Toward a more nuanced understanding of the statistical properties of a median split. Journal of Consumer Psychology, 25(4), 652-665.

Iacobucci, D., Posavac, S. S., Kardes, F. R., Schneider, M. J., & Popovich, D. L. (2015). The median split: Robust, refined, and revived. Journal of Consumer Psychology, 25(4), 690-704.

Mai, R., & Hoffmann, S. (2012). Taste lovers versus nutrition fact seekers: how health consciousness and self‐efficacy determine the way consumers choose food products. Journal of Consumer Behaviour, 11(4), 316-328.

McClelland, G. H., Lynch Jr, J. G., Irwin, J. R., Spiller, S. A., & Fitzsimons, G. J. (2015). Median splits, Type II errors, and false–positive consumer psychology: Don't fight the power. Journal of Consumer Psychology, 25(4), 679-689.

Michaelidou, N., & Hassan, L. M. (2008). The role of health consciousness, food safety concern and ethical identity on attitudes and intentions towards organic food. International journal of consumer studies, 32(2), 163-170.

Milliron, B. J., Woolf, K., & Appelhans, B. M. (2012). A point-of-purchase intervention featuring in-person supermarket education affects healthful food purchases. Journal of Nutrition Education and Behavior, 44(3), 225–232.

Newman, C. L., Burton, S., Craig Andrews, J., Netemeyer, R. G., & Kees, J. (2018). Marketers’ use of alternative front-of-package nutrition symbols: An examination of effects on product evaluations. Journal of the Academy of Marketing Science, 46(3), 453–476.

Park, C. W., Iyer, E. S., & Smith, D. C. (1989). The effects of situational factors on in-store grocery shopping behavior: The role of store environment and time available for shopping. Journal of consumer research, 15(4), 422-433.

Power, E. M. (2005). Determinants of healthy eating among low-income Canadians. Canadian journal of public health/revue canadienne de sante'e publique, 37-42.

Pesta, B. J., Bertsch, S., McDaniel, M. A., Mahoney, C. B., & Poznanski, P. J. (2012). Differential epidemiology: IQ, neuroticism, and chronic disease by the 50 US states. Intelligence, 40(2), 107-114.

Roininen, K., Lähteenmäki, L., & Tuorila, H. (1999). Quantification of consumer attitudes to health and hedonic characteristics of foods. Appetite, 33(1), 71–88.

Romero, M., & Biswas, D. (2016). Healthy-Left, Unhealthy-Right: Can Displaying Healthy Items to the Left (versus Right) of Unhealthy Items Nudge Healthier Choices? Journal of Consumer Research, 43(1), 103–112.

Rosch, E., & Mervis, C. B. (1975). Family resemblances: Studies in the internal structure of categories. Cognitive Psychology, 7(4), 573–605.

Rucker, D. D., McShane, B. B., & Preacher, K. J. (2015). A researcher's guide to regression, discretization, and median splits of continuous variables. Journal of Consumer Psychology, 25(4), 666-678.

Russo, J. E., Edward Russo, J., & Leclerc, F. (1994). An Eye-Fixation Analysis of Choice Processes for Consumer Nondurables. Journal of Consumer Research, 21(2), 274. https://doi.org/10.1086/209397.

Scheibehenne, B., Miesler, L., & Todd, P. M. (2007). Fast and frugal food choices: uncovering individual decision heuristics. Appetite, 49(3), 578–589.

Talati, Z., Norman, R., Pettigrew, S., Neal, B., Kelly, B., Dixon, H., Ball, K., Miller, C., & Shilton, T. (2017). The impact of interpretive and reductive front-of-pack labels on food choice and willingness to pay. The International Journal of Behavioral Nutrition and Physical Activity, 14(1), 171.

Tan, L., & Ward, G. (2000). A recency-based account of the primacy effect in free recall. Journal of Experimental Psychology. Learning, Memory, and Cognition, 26(6), 1589–1625.

Temple, N. J. (2020). Front-of-package food labels: A narrative review. Appetite, 144, 104485. Turilli, M., & Floridi, L. (2009). The ethics of information transparency. Ethics and Information

Technology, 11(2), 105-112.

von Lengerke, T., & Krauth, C. (2011). Economic costs of adult obesity: a review of recent European studies with a focus on subgroup-specific costs. Maturitas, 69(3), 220–229. Wansink, B. (2017). Healthy profits: An interdisciplinary retail framework that increases the sales

of healthy foods. Journal of Retailing, 93(1), 65-78.

WHO. (2016). Obesity and overweight, Fact sheet N° 311. Retrieved from https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight

Appendices

Appendix 1

Gioia Langen Thesis

Start van blok: Welkom

Welkom bij deze vragenlijst,

U helpt mij enorm door deze vragenlijst in te vullen!

Voor mijn scriptie doe ik onderzoek naar verschillende supermarktvormen en de voorkeur van mensen tijdens het boodschappen doen. Het invullen van de vragenlijst duurt hooguit 5 minuten.

De vragenlijst zal anoniem zijn en de data die wordt verkregen door middel van deze vragenlijst zal uitsluitend worden gebruikt voor mijn onderzoek.

Alvast heel erg bedankt voor het invullen van de vragenlijst!

Gioia Langen

In een supermarkt moeten klanten constant keuzes maken tussen verschillende producten. De voorkeur van een consument kan afhangen van verschillende product factoren, denk hierbij bijvoorbeeld aan: smaak, prijs, verpakking, gezondheid, diervriendelijkheid en duurzaamheid.

Er bestaan verschillende labels om bijvoorbeeld de duurzaamheid, diervriendelijkheid en gezondheid van een product te duiden.

De Albert Heijn gebruikt bijvoorbeeld onderstaande logo om aan te duiden dat producten duurzaam zijn:

Ook gebruiken supermarkten ranking labels om verschillende factoren aan te duiden. Het beter leven label is hier een voorbeeld van, waarbij het aantal sterren aangeeft hoe diervriendelijk een product is.

Nutri-score is ook zo'n label, dit ranking label geeft aan in hoeverre een product een gezonde keuze is. Waarbij een A staat voor de meest gezonde keuze en E staat voor de meest

ongezonde keuze.

Stelt u zich bij de volgende vragen voor dat u boodschappen aan het doen bent.

U krijgt telkens een supermarkt schap te zien, waarbij u 1 product moet kiezen. Als u het product heeft gekozen dat u zou willen kopen, klik dan op de pijltjes onderaan de pagina (>>) om naar de volgende categorie te gaan. In totaal moet u 8 product keuzes maken uit

verschillende product categorieën.

Let op: u maakt kans om deze producten aan het eind van de enquête ook echt thuisgestuurd te krijgen.

Pagina-einde

Einde blok: Welkom

Start van blok: Nutri-score conditie / Structure conditie / Nutri-score + structure condition / Control condition

Kies 1 product

Welke van de onderstaande tussendoortjes wilt u kopen?

Pagina-einde

Kies 1 product

Welke van de onderstaande dranken wilt u kopen?

Pagina-einde

Kies 1 product

Kies 1 product

Welke van de onderstaande desserts wilt u kopen?

Pagina-einde

Kies 1 product

Welke van de onderstaande producten van de bakkerij wilt u kopen?

Pagina-einde

Kies 1 product

Welke van de onderstaande koekjes wilt u kopen?

Pagina-einde

Kies 1 product

Welke van de onderstaande crackers / rijstwafels wilt u kopen?

Kies 1 product

Welke van de onderstaande chips wilt u kopen?

Einde blok: Nutri-score conditie

Start van blok: Koert and David nutri-score condition / Structure conditie / Nutri-score + structure conditie / Control conditie

Bedankt voor uw 8 keuzes!

In het volgende blok krijgt u weer 8 schappen te zien van verschillende product categorieën, geef bij de vraag onderaan het schap aan hoe gezond u denkt dat de producten in deze categorie zijn.

Alle keuzes - Verborgen en zichtbaar overbrengen van "Kies 1 product Welke van de onderstaande tussendoortjes wilt u kopen?"

Onderstaande ziet u een supermarkt schap met tussendoortjes.

Wat is volgens u de gemiddelde Nutri-score van deze product categorie? Waarbij A de meeste gezonde keuze is en E de meest ongezonde keuze

o

o

o

o

o

Alle keuzes - Verborgen en zichtbaar overbrengen van "Kies 1 productWelke van de onderstaande dranken wilt u kopen? "

Onderstaande ziet u een supermarkt schap met drinken.

Wat is volgens u de gemiddelde Nutri-score van deze product categorie? Waarbij A de meeste gezonde keuze is en E de meest ongezonde keuze

Alle keuzes - Verborgen en zichtbaar overbrengen van "Kies 1 productWelke van de onderstaande ontbijtgranen wilt u kopen? "

Onderstaande ziet u een supermarkt schap met ontbijtgranen.

Wat is volgens u de gemiddelde Nutri-score van deze product categorie? Waarbij A de meeste gezonde keuze is en E de meest ongezonde keuze

o

o

o

o

o

Alle keuzes - Verborgen en zichtbaar overbrengen van "Kies 1 productWelke van de onderstaande desserts wilt u kopen? "

Onderstaande ziet u een supermarkt schap met desserts.

Wat is volgens u de gemiddelde Nutri-score van deze product categorie? Waarbij A de meeste gezonde keuze is en E de meest ongezonde keuze