HEALTHINESS OF SHOPPING BASKET
The effect of the introduction of a healthy food product and its
timing on the healthiness of a grocery shopping basket
Author: Esther Vos, s2031809 Supervisor: Prof. Dr. K. van Ittersum Co-Reader: Mr. Y. Joye
Completion Date: 19-06-2015 Abstract
Worldwide obesity rates are soaring and will continue to rise if precaution measurements are not taken. Supermarkets and manufacturers are facing increasing public pressure to address the issue. This study was one of the first that examined the effect of the introduction of a healthy food product during people’s grocery shopping trip on the healthiness of a shopping basket. This was tested via an online survey that simulated an online shop. The effects were compared against the introduction of a relatively unhealthy and neutral product introduction. Results indicated that a the healthiness of the shopping basket slightly improved after the introduction of a healthy food product, but the results were insignificant. Nevertheless, this study brought some relevant insight into the relationship between health consciousness and the healthiness of a shopping basket as this is found to be positively related. More research is needed to find out how supermarkets can make consumers more health conscious.
Title: Healthiness of shopping basket
Subtitle: The effect of the introduction of a healthy food product and its timing on the healthiness of a grocery shopping basket
Author: Esther Vos
Department: Marketing
Qualification: Master Thesis Marketing Management Completion Date: 19-06-2015 Address: Zeeburgerpad 127 1019 DX Amsterdam Phone number: 06 511 99 628 E-mail: e.vos.4@student.rug.nl Student number: s2031809
Acknowledgements
I would like to express the deepest appreciation to Prof. Dr. Van Ittersum, Professor of Marketing and Consumer Well-Being at the University of Groningen, my supervisor, who has advised and guided me from the very beginning of this research. I want to thank him for his supervision, contribution, continuous support and helpful suggestions during all stages of this study.
I would like to express my gratitude to Prof. Veenstra, Researcher and Professor at the NHL University of Applied Science, who has helped me with the SPSS analyses and gave me constructive feedback on my results.
Executive Summary
Obesity is a worldwide problem that affects the lives of many adults and children. In recent years, the prevalence of dietary-related diseases such as diabetes, cancer and cardiovascular diseases has increased. Part of the obesity epidemic can be explained by lack of exercise and poor consumer choices, but the food industry is often blamed as well. Given the fact that most of all food purchases are done within supermarkets, manufacturers of food products and supermarkets are facing increasing public pressure to address the issue of obesity. Many studies have investigated how in-store food marketing can influence food-purchase behaviors, but only few studies have examined the effects of a supermarket’s healthy food product assortment on consumer’s purchase intentions. Nonetheless, supermarkets might be in the position to steer consumers toward making healthier food choices if they increase their healthy food product assortment. Therefore, this study examined the effect of the introduction and its timing of a healthy food product during people’s grocery shopping trip on the healthiness of a shopping basket. This was tested via an online survey that simulated an online shop. The effects were compared against the introduction of a relatively unhealthy and neutral food product introduction as well as a control condition. Results from this study revealed that the introduction of a healthy food product resulted in a slightly healthier shopping basket than the introduction of a relatively unhealthy and neutral food product introduction. In addition, early introduction of the relatively healthy food product led to a less healthy shopping basket. However, the results were insignificant. This suggests that it is not easy for supermarkets to improve the healthiness of consumer’s shopping basket simply by extending their healthy food product assortment. Nevertheless, this study brought some relevant insight into the relationship between health consciousness and the healthiness of a shopping basket as this was found to be positively related. Supermarkets and (inter) national institutions would be wise to launch campaigns focusing on making consumers more health conscious if they are to promote healthier living to fight against the rise in obesity. Future research is needed to find out how supermarkets and (inter) national institutions can make consumers more health conscious, eventually encouraging them to make more healthy food product choices.
Table of content
ACKNOWLEDGEMENTS 3
EXECUTIVE SUMMARY 4
TABLE OF CONTENT 5
1. INTRODUCTION 6
2. LITERATURE REVIEW 9
2.1 BACKGROUND 9
2.2 HEALTHY FOOD 10
2.3 SUPERMARKET’S HEALTHY FOOD ASSORTMENT 11
2.4 CONSUMER’S LIMITED COGNITIVE ABILITY TO MAKE THE RIGHT DECISIONS 12
2.5 CONSUMER’S HEALTH CONSCIOUSNESS 14
2.7 CONCEPTUAL MODEL 15
3. METHODOLOGY 16
3.1 DATA COLLECTION METHOD & STUDY DESIGN 16
3.2 PRE-‐TEST 16
3.3 PROCEDURE 16
3.4 STIMULI 18
3.5 MEASUREMENT OF VARIABLES 19
3.5.1 HEALTHINESS OF SHOPPING BASKET 19
3.5.2 HEALTHINESS OF FOOD PRODUCT INTRODUCTION 20
3.5.3 TIMING 20
3.5.4 HEALTH CONSCIOUSNESS 20
3.6 DATA ANALYSIS 20
4. ANALYSIS & RESULTS 22
4.1 SAMPLE 22 4.2 RELIABILITY 22 4.3 HYPOTHESES TESTING 22 4.5 HYPOTHESES OVERVIEW 28 5. DISCUSSION 29 5.1 GENERAL DISCUSSION 29 5.2 LIMITATIONS 30 5.3 MANAGERIAL IMPLICATIONS 31 5.5 FUTURE RESEARCH 32 5.6 CONCLUSION 33 APPENDICES 37
I. QUESTIONNAIRE – COOKIES CATEGORY 37
II. TOTAL QUESTIONNAIRE – CONTROL CONDITION 39
1. Introduction
what manufacturers will offer and as such, manufactures of food products should produce and offer more relatively healthy foods. Nowadays, manufacturers are often blamed for adding too much sugar or salt in foods and use misleading labels letting consumers believe the product is a healthy choice (Basu, 2014; Palou et al., 2009).
All in all, preventing the rise in overweight and obesity is a complex task as it depends on consumers’ limited cognitive ability to make deliberate (healthy) food choices and various factors outside of people’s control: the type products offered by manufactures and supermarkets. Yet, despite an extensive body of research on the relationship between grocery shopping behavior, the role of supermarkets and obesity, much remains to be learned in the fight against the rise of the obesity epidemic. Until now, little research is available on how the introduction of healthy foods during consumer’s shopping trip might affect consumer’s subsequent purchase intentions and the overall healthiness of their shopping basket. Since supermarkets receive more attention as possible solution to prevent obesity and overweight, research is needed to shed light on it and to prove whether an increased access to healthy food products actually leads consumers to buy more of these. The purpose of this study is therefore to examine the effects of the introduction of a healthy, neutral and unhealthy product introduction on the healthiness of shopping basket. More specifically, the main research question guiding this study is:
What is the effect of the introduction of a healthy food product during consumer’s shopping trip on the healthiness of a shopping basket?
Moreover, this study is also interested in whether the timing of the introduction and health consciousness will have an effect on the healthiness of a shopping basket. As such, a sub research question has been developed:
Do timing and/ or health consciousness have a moderating role on the relationship between the introduction of a food and the healthiness of the shopping basket?
2. Literature review
In this section, hypotheses will be developed through a review of relevant behavioral literature that provides insights into the dynamics of consumer’s shopping behavior when they are exposed to a healthy food product assortment during their grocery shopping trip. This study is interested to see how this might affect the overall healthiness of consumer’s shopping basket. First, the background of this study will be discussed, followed by a definition of healthy food and by a theoretical discussion.
2.1 Background
People are getting fatter and fatter (Hausman, 2010). The health consequences of overweight and obese are serious and can be deadly. If we do not take prevention measurements, the obesity epidemic will continue to rise. According to Anker et al., (2011), part of the worldwide obesity problem is due to people’s poor decision making. However, some also blame the food industry (Payne et al., 2014).
2.2 Healthy food
For the purpose of this study, it is important to first define what is meant with relatively healthy food and how the healthiness of consumer’s shopping basket will be measured. However, researchers among the world have not yet agreed on a uniform definition for healthy food. This study will therefore make use of the Traffic Light Labeling system as developed by the United Kingdom Foods Standards Agency to determine what type of products can be categorized as relatively healthy. The Traffic Light system uses green, amber and red colors to inform individuals about the levels of fat, saturated fat, sugar, and salt in foods (Roberto et al., 2012). Thereby indicating that red means high in content signaling that people should not consume too much of it, amber indicates that the food is neither high nor low in content and means that is an ok option whereas green refers to low in content (Food Standards Agency, 2015). The selection of the four nutritions is based on their impact on public health (Hieke and Wilczynski, 2011). The European Regulation for Nutrition and Health Claims has decided the cut-off points for the change in traffic light colors. Figure 2.1 below depicts the cut-off points.
Figure 2.1: Traffic Light Label cut-off points
This is in line with the green color of the Traffic Light Labeling system. Food products that are neither unhealthy nor healthy will be labeled as ‘neutral’ in this study and contain between 5 and 15 grams of sugar, 3 and 20 grams of fat, between 1.5 and 5 grams of saturated fat and between 0.3 and 1.5 grams of salt per 100 grams. Unhealthy foods are those that contain over 15 grams of sugar and more than 20 grams of fat, 5 grams of saturated fat and more than 1.5 grams of salt per 100 grams. More details on the categorization of the food products and the measurement of the healthiness of consumer’s shopping basket will be provided in the methodology of this study.
2.3 Supermarket’s healthy food assortment
among consumers. As the aim of this research is to examine whether the introduction of healthy food products during people’s shopping trip affect their subsequent food product choices and whether it actually leads to an increase in the healthiness of a shopping basket, the following hypothesis has been formulated:
H1. The introduction of a healthy food product during consumers’ shopping trip will have a more positive effect on the healthiness of a shopping basket than the control condition and than the introduction of a neutral- and unhealthy food product.
2.4 Consumer’s limited cognitive ability to make the right decisions
Another theory that might explain consumer’s unconscious decision making during their shopping trip is the theory of the ‘shopping momentum effect’. The shopping momentum effect occurs when the purchase of driver item, that is the initial, unrelated item, increases the likelihood of purchasing a target, which is the subsequent item (Dhar, Huber and Khan, 2007). In a purchase context, ‘momentum can viewed as an outcome of a switch from deliberative to implementation orientation evoked by the initial purchase’. Similar to this, Kahn and Dahr (2006) found ‘licensing’ effects in consumer choice. This means that people give themselves a license to purchase a ‘virtue’ (e.g. vegetables, organic food) product, after they have purchased a ‘vice’ (e.g. ice cream, beer) product, since the latter positively influences people’s self-concept. The reverse also holds; buying vice categories reduces the consumer’s self-concept and additional purchases of vice categories increase the negative self-attribution. When looking at the healthiness of a shopping basket, it is assumed that at any moment during the shopping trip, the extent of the licensing effect is modeled by the current balance of vice/virtue products (Hui, Bradlow and Fader, 2009). Thus, within the model of this research, it is expected that when a basket contains more virtue categories than vice ones, the licensing effect should occur, increasing the likelihood that consumers will buy from vice categories afterwards and vice versa. However, this might also imply that due to licensing effects, consumers might end up with a ‘balanced’ shopping basket. When looking at the effect of the timing of the introduction of healthy food products on the healthiness of people’s shopping basket, it is expected that early or late introduction do not show different results because of licensing effects. Even though consumers may buy more healthy food products from virtue categories after the introduction (that is at the first or second half of the shopping trip), they will replace the initial healthiness of their basket by buying from vice category products afterwards. Since this research is interested in whether the ‘timing’ of the introduction of healthy food products makes a difference in the healthiness of a shopping basket, the following hypothesis has been formulated:
2.5 Consumer’s health consciousness
Besides the external role of supermarket’s product assortment on consumers buying intentions and food intake, attention needs to be paid to the possibly moderating effect of consumer’s health consciousness when looking at the relationship between the introduction of the healthy food products and the healthiness of a shopping basket. Becker et al., (1977) defined health consciousness as ‘the readiness to undertake health actions.’ Consumers that are health consciousness are aware and concerned about their state of well being (Michaelidou and Hassan, 2008). These people are also motivated to improve and/or maintain their health next to preventing illness by means of engaging in healthy behaviors, such as healthy eating and being aware of their own health (Newsom et al., 2005; Michaelidou and Hassan, 2008); Jayanti & Burns, 1998). Moreover, these people tend to be ‘conscious about, and involved with nutrition and physical fitness’ (Michaelidou and Hassan, 2008). According to Duta-Bergman (2004), health consciousness can be defined as ‘the extent to which health concerns are integrated into a person’s daily activities’. This study therefore assumes the following hypothesis:
2.7 Conceptual model
Belown an overview of the conceptual model for this research.
Figure 2.1: Conceptual model
3. Methodology
Chapter 3 describes the research methodology of the research at hand. It will also provide the data collection process, the detailed description of the questionnaire and the measurement of the variables. The analyses and results of the study will be presented in the next chapter. 3.1 Data collection method & study design
This research wants to find out if the introduction of a healthy food product affects the overall healthiness of a shopping basket. In order to find these effects, an online questionnaire is conducted which will be discussed in more detail later this section. Online questionnaires have a lot of advantages (e.g. structured, data quality, speed) that makes the data collection an easier and more accurate method (Malhorta and Birks, 2007). The study design was a 4 (type of introduced food; control, neutral, healthy, unhealthy) by 2 (timing of introduction; early introduction - at the first half of the shopping trip, late introduction - at the second half of the shopping trip) between subjects design. Participants were people responsible for most of the household food purchases.
3.2 Pre-‐test
A pre-test was run among 5 people to test whether everything was clear to the respondents and to correct for mistakes so that a most reliable and valid final questionnaire could be established. After the pre-test and after a critical look of both professor Mr. Van Ittersum and Mrs. McWhorter (panels project manager Qualtrics), the study was distributed among American citizens via Qualtrics.
3.3 Procedure
an online shop simulation (Van Ittersum et al., 2014). Each of the sixteen product categories carries three options of one product: a relatively unhealthy and a relatively healthy version of the product and a neutral option. The neutral food item was added as a control variant and to enrich the choice options in order ‘to avoid all-white and black scenarios’ (Van Ittersum, 2015). The differentiation for the three food items is based on the classification of the Traffic Light Labeling as developed by the Food Standards Agency (2015), which has been described in the literature review of this paper already. However, it must be noted that some categories, such as ice cream and chips, are in nature less healthy than other categories such as Yoghurt. These categories are not considered to be healthy according to the Traffic Light Labeling and it was therefore impossible to offer a healthy, neutral and unhealthy option according to the Traffic Light Labeling standards. To differentiate between the products within these categories, relatively healthier options were included. Table 3.1 depicts an overview of the sixteen product categories:
Product type:
Product category: Unhealthy Neutral Healthy
Breads Great Value White Bread Great Value Brown Bread Great Value Whole Wheat Bread Milk Great Value Fat 2% Milk Great Value Low Fat 1% Milk Great Value 0% Fat Milk Yoghurt Mountain High Original Plain Style yoghurt Mountain High Plain Lowfat Yoghurt Mountain High Plain Lowfat Yoghurt Cookies Chips Ahoy! Chunky Chocolate Chip Cookie Chips Ahoy! Original Chocolate Chip Cookie Chips Ahoy! Oatmeal Chocolate Chip Cookie Cold Cereal Great Value Sugar Frosted Flakes Great Value Corn Flakes Great Value Bran Flakes Spreadable Butter Land O Lakes Butter with Canola oil Land O Lakes Butter with Olive Oil Land O Lakes Butter Light Peanut Butter Skippy Creamy Peanut Butter Skippy Natural Creamy Peanut Butter Skippy Reduced Fat Creamy Peanut Butter Chips Pringles Original Potato Crisps Pringles Original Reduced
Fat Potato Crisps Pringles Original Fat Free Potato Crisps
Cola Pepsi Cola Pepsi Next Pepsi Diet
Ice cream Ben & Jerry’s Chocolate Ice Cream Ben & Jerry's Greek Frozen Yoghurt Ben & Jerry's Frozen Yoghurt Half Baked Macarroni & Cheese Hospitality Mac & Cheese Dinner Kraft Premium Three Cheese Macaroni & Cheese Dinner Annie’s Homegrown Organic Classic Mild
Cheddar Macaroni & Cheese Spaghetti Great Value Spaghetti Great Value Whole Wheat Spaghetti Barilla Whole Grain Spaghetti Pasta Sauce Newman's Own Pasta Sauce Bertolli Organic Traditional
Sauce
Gina Rispoli All Natural Sauce
Beef Patties Ground Beef Patties 73% Lean Ground Beef Patties 85% Lean Ground Beef Patties 93% Lean Salad Dressing Great Value Creamy Ranch Salad Dressing Great Value Classic Ranch
Light Dressing
Great Value Fat Free Creamy Ranch Salad Dressing
To minimize bias on brand and taste selection, products within one category were all from the same brand and almost all had the same flavor (e.g. Skippy Creamy Peanut butter, Skippy Natural Creamy Peanut Butter and Skippy Reduced Fat Creamy Peanut Butter 40 oz). Four categories carried different brands within their category, as there was no relatively unhealthy or healthy option available. The product details that respondents were provided in the questionnaire included: the name of the product (e.g. Skippy Creamy Peanut Butter, 40 oz) a 200x200px photo, price in dollars and nutritional value per 100g - calories in kcal, sugar, fat, saturated fat, and salt. In this way, it was ensured that respondents had enough information to make well-considered decisions. The photos, prices and nutrition information for the products used in this study are obtained from the online shop of the US retailer Walmart (Walmart.com).
3.4 Stimuli
Control Unhealthy food introduction Neutral food introduction Healthy food introduction Early introduction: cookies category fixed at number 3 C1.1 C2.1 (early introduction of unhealthy cookie) C3.1 (early introduction of neutral cookie) C4.1 (early introduction of healthy cookie) Late introduction: cookies category fixed at number 11 C1.2 C2.2 (late introduction of unhealthy cookie) C3.2 (late introduction of neutral cookie) C4.2 (late introduction of healthy cookie) Table 3.2: Conditions
3.5 Measurement of variables
3.5.1 Healthiness of shopping basket
As this study is predominantly interested in the healthiness of consumer’s shopping basket, its measurement needs to be clarified. The healthiness of consumer’s shopping basket is the sum of the scores on all the sixteen food product categories. This measurement is based on a study by Waterlander et al. (2012) who looked at the effects of different discount levels on healthy products in their study. They counted the number of relatively healthy and relatively unhealthy food products in consumer’s shopping basket. In order to do so, all items for the categories of this study were recoded: ‘-1’ for relatively unhealthy food products, ‘0’ for neutral food products, and ‘+1’ for relatively healthy items. The three new cookies that were introduced to the respondents were coded in a similar way. As a result, the healthiness of a consumer’s shopping basket is a continuous variable and its values vary from -16 (unhealthy shopping basket) to +16 (healthy shopping basket).
been made. Moreover, this measurement may result in negative values when a respondent has bought fewer relatively healthy options than relatively unhealthy.
3.5.2 Healthiness of food product introduction
To test the effect of the healthy food product introduction, a control condition was created. Furthermore, a neutral cookie and a relatively unhealthy cookie were added to the survey alongside the relatively healthy cookie introduction. As a result, a dummy variable was created where ‘0’ stands for the control condition, ‘1’ for the unhealthy food product introduction, ‘2’ for the neutral-, and ‘3’ for the healthy food product introduction.
3.5.3 Timing
In order to measure the effect of timing on the healthiness of a shopping basket, a dummy variable will be created with two levels where ‘0’ stands for early introduction of the cookies category and ‘1’ for late introduction.
3.5.4 Health consciousness
Respondents’ ‘Health Consciousness’ is measured based on six items that were derived from a study by Dutta-Bergman (2009). The items are as follows: ‘Living life in the best possible health is important to me’, ‘Eating right, exercising, and taking preventive measures will keep me healthy for life’, ‘My health depends on how well I take care of myself’, ‘I actively try to prevent disease and illness’, ‘I do everything I can to stay healthy’, and ‘Calorie information is important to me’. Respondents were asked to rate these items based on a likert scale from 1 (strongly disagree) to 7 (strongly agree). Therefore, the scores on health consciousness per respondent have values between 6 (not health consciousness at all) and 42 (very health consciousness). Afterwards, the median of all scores will be computed. Based on this score, a dummy variable will be created with two levels where ‘0’ indicates that people are not health conscious and ‘1’ indicates that people are health conscious. This dummy variable is created, as a three-way ANOVA analysis requires that the independent variables should be categorical (statistics.leard.com, 2015).
3.6 Data Analysis
4. Analysis & Results
This chapter analyses the results obtained from the survey discussed in Chapter 3. It also provides a clear outline on which analyses have been applied to the data from the survey. First, the sample of this survey will be presented. Afterwards, the results are going to be discussed in a statistically and visually way. The last part is the verifying of the hypotheses through a three-way ANOVA analysis.
4.1 Sample
In total, 304 Americans participated (38 respondents per condition) in the online questionnaire. However, 22 respondents did not fill in questions regarding their health consciousness or concerning their demographics. Since these questions are important for this study, they have been reported as missing values and deleted. After cleaning the data, a sample of 282 respondents remained of which 195 participants are female (70%) and 87 are male (30%). The average age of the respondents was 43 years old, with the youngest respondent being 18 years old and the oldest 76 years old.
4.2 Reliability
After selecting the items for people’s health consciousness, the reliability of the scales for health consciousness was measured. It is necessary to check reliability of each scale in order to see if there’s internal consistency among the scales. Internal consistency refers to the degree in which the items that make up the scale are all measuring the same underlying construct. The most commonly used indicator of the internal consistency is the Cronbach alpha coefficient. In an ideal and reliable way, the Cronbach alpha coefficient should be above 0.7. The Cronbach alpha coefficient concerning Health Consciousness is 0.856, making it highly reliable. After looking at the Cronbach alpha if item deleted within the variables, there are no items that have a higher coefficient than the current alpha’s, therefore no items need to be deleted. See appendix III for an overview of the results of this test.
4.3 Hypotheses testing
certain conditions. First, the dependent variable of your study should be measured at the continuous level. ‘Healthiness of a shopping basket’ meets this criteria as its values vary between -16 and 16, making it an interval variable. Furthermore, the three independent variables of this study all consist of two or more categorical, independent groups. The first variable of this study, ‘healthiness of the introduction’ has four categorical independent groups and the moderators of this study ‘timing’ and ‘health consciousness’ have two independent groups as well. Moreover, this study makes use of a between-participant subject design as there are different participants in each condition with no participant being in more than one of the conditions. The final condition that needs to be met when analyzing your data with the three-way ANOVA is that there should not be significant outliers. Outliers can have a negative on the three-way ANOVA, reducing the accuracy of the results. After checking the data for outliers in SPSS, there appeared to be no outliers in the scores of the healthiness of respondent’s shopping basket and in the scores of health consciousness when applying the outlier-labeling rule as identified by Hoaglin et al. (1986; 1987). Next, data needs to be checked for normal distribution. The dependent variable of this should be approximately normally distributed for each combination of the groups of the three independent variables. This is done with the Shapiro-Wilk test of normality. A Shapiro’s-Wilk test (p>.05) and a visual interpretation of the histograms and Normal Q-Q plots showed that the ‘healthiness of a shopping basket’ was not normally distributed for each of the combinations of the three independent variables of this study. As such, the results of this study may be slightly inaccurate. Finally, Levene’s test for homogeneity of variances was applied. Output shows a p-value of .680, indicating that the variance of the ‘healthiness of a shopping basket’ is equal across the groups, which is desirable.
4.4 Hypotheses
Below an overview of the hypotheses of this study:
H1 The introduction of a healthy food product during consumers’ shopping trip will have a more positive effect on the healthiness of a consumer’s shopping basket as compared to the control condition and to introduction of a neutral- and unhealthy food product.
H2 The effect of the healthiness of the food product introduction on the healthiness of the shopping basket depends on the timing of the introduction: an early introduction yields a less healthy shopping basket than a late introduction because of licensing effects.
H3 Health consciousness moderates the relationship between the healthiness of the food product introduction and the healthiness of a shopping basket in such a way that health conscious consumers are more likely to adopt the relatively healthy food product introduction than non health conscious consumers, resulting in a healthier shopping basket.
Table 4.1: Overview of hypotheses
A three-way between-groups analysis of variance was conducted to test the impact of the healthiness of the food product introduction on the healthiness of a shopping basket, next to the possible moderating effects of the timing of the introduction and consumer’s health consciousness. Table 4.2 below provides an overview of the test of between-subject effects.
R squared = 0.042 (Adjusted R square = -0.012)
Table 4.2: Main effects and interaction effects with the healthiness of a shopping basket as dependent variable
Variable Mean
Square
df F Sig. Partial Eta
Squared Healthiness of introduction 15.26 3 0.32 0.813 0.004
Timing 8.71 1 0.18 0.671 0.001
Health consciousness
Healthiness of introduction * Timing
304.85 16.33 1 3 6.33 0.32 0.012 0.797 0.023 0.004
Healthiness of introduction * Health consciousness
18.49 3 0.38 0.765 0.004
Timing * Health Consciousness 1.96 1 0.04 0.840 0.000
Healthiness of Introduction * Timing * Health consciousness
First, the interaction effects will be discussed. The three-way interaction effect showed no significant result as can be seen from the table. The interaction effect between the healthiness of the introduction and health consciousness showed no significant results: p>0.005 (F=0.38). In addition, there was no statistically significant result as well between the healthiness of the introduction and timing, since p>0.005 (F=0.32). Figure 4.1 below displays the interaction effect of timing and the healthiness of introduction. From this it can be seen that all the mean scores for the late introduction are slightly above the mean scores for the early introduction, except for the neutral food product introduction.
Figure 4.1 Interaction effect healthiness of introduction and timing of the introduction
Next, the main effects will be analyzed. With regard to the main effect of the healthiness of the introduction, there is no statistically significant result (i.e. p-value <0.05) as the healthiness of the introduction has a p-value of 0.813, F=0.32. However, the mean score for the healthy food product introduction is slightly higher (-2.352) than the mean scores for the control condition and the relatively unhealthy- and neutral introduction. Figure 4.2 on the next page provides an overview of the mean scores for the healthiness of the shopping basket based on the healthiness of the food product introduction. From this it can be seen that all the mean scores are negative, indicating that people buy more relatively unhealthy food products than healthy ones, regardless of the healthiness of the introduction.
Figure 4.2 Marginal mean scores for the healthiness of the shopping basket based on the healthiness of the food product introduction
When looking at the main effect timing of the introduction, there appears to be no statistically significant effect. However, results show that early introduction leads to a less healthy shopping basket on average than the late introduction. This was hypothesized, but the results are insignificant. Concerning the main effect of health consciousness, there is a significant effect (P<0.05): p=0.012, F=6.33. The partial Eta-Squared is however quite small (0.023), indicating that the actual differences in the mean values of healthiness of a shopping basket based on health conscious people and non health conscious people are very small. Only 2.3% of the between subjects variance is accounted for by health consciousness. Table 4.3 below demonstrates the marginal mean scores per variable.
Healthiness of Introduction Mean St. deviation N Control -3.381 0.853 69 Unhealthy -3.244 0.833 72 Neutral -2.750 0.848 71 Healthy -2.352 0.604 70 Timing Early -3.11 0.585 143 Late -2.75 0.604 139 Health Consciousness
Not health conscious -3.990 0.614 132 Health conscious -1.873 0.575 150
Table 4.3 Marginal mean scores healthiness of shopping basket per variable
-‐3.381 -‐3.244 -‐2.750 -‐2.352 -‐4.000 -‐3.000 -‐2.000 -‐1.000 0 Control Unhealthy
introduction introduction Neutral introduction Healthy
Mean scores healthiness shopping
Post-hoc tests on health consciousness have not been performed, as this is only relevant when the independent variable would have had more than two levels (Field, 2009). Planned contrasts, which compare the means of several groups to one other mean, have also been conducted but showed no statistically significant results.
Furthermore, a three-way ANOVA was performed with the share of healthy products as measurement for the healthiness of the shopping basket. The three-way interaction effect showed no significant results, as well as the two-way interaction effects since p>0.05. In addition, there was no statistically significant main effect for the healthiness of the introduction as well as the timing of the introduction. Nevertheless, when looking at the average scores for the healthiness of the introduction, it can be seen in figure 4.3 that the healthy food product introduction results in the highest share of healthy food products since the mean score is 30,1%. This shows that the healthy food product introduction has a more positive effect on the consumer’s purchase intentions afterwards as compared to the other conditions. However, the results are insignificant.
Figure 4.3 Marginal mean scores for share of healthy products based on the healthiness of the food product introduction
The only significant result found was the main effect health consciousness as it had a p-value below 0.05: p=0.027, F=4.97. This result is less significant as compared to the other measurement (healthiness of shopping basket), and it will therefore not be discusses in detail.
26,6 27,2 27,2 30,1 24,0 25,0 26,0 27,0 28,0 29,0 30,0 31,0 Control Unhealthy
4.5 Hypotheses overview
After performing multiple different statistical tests and analyzing the results carefully, all of the hypotheses guiding this study can now be answered. Table 4.4 below provides an overview of the hypotheses of this research and the outcome.
Hypotheses Outcome
H1 The introduction of a healthy food product during consumers’ shopping trip will have a more positive effect on the healthiness of a shopping basket than the control condition and than the introduction of a neutral- and unhealthy food product.
Unsupported
H2 The effect of the healthiness of the food product introduction on the healthiness of the shopping basket depends on the timing of the introduction: an early introduction yields a less healthy shopping basket than a late introduction because of licensing effects.
Unsupported
H3 Health consciousness moderates the relationship between the healthiness of the food product introduction and the healthiness of a shopping basket in such a way that health conscious consumers are more likely to adopt the relatively healthy food product introduction than non health conscious consumers, resulting in a healthier shopping basket.
Unsupported
5. Discussion
Following from Chapter 4, it can be stated that this study does not show any statistically significant results. This chapter will discuss the results and will answer the main research question of this study. Afterwards, some recommendations for managerial and theoretical implications will be provided. In addition, the limitations and suggestions for further research will be given.
5.1 General discussion
Worldwide obesity rates are increasing and this calls the need for measurements that encourage consumers to make more healthy food product choices. Supermarkets are seen as an institution that may be ripe for public health interventions by increasing access toward healthier food products and helping consumers to buy more healthy food products. However, several studies have demonstrated that consumers may make poor choices as a result of licensing effects and a depleted self-control resource. Since supermarkets receive more attention as possible solution to prevent obesity and overweight, research was needed to shed light on it and to prove whether an increased access to healthy food products actually leads consumers to buy more of these. This study was one of the first studies that investigated the effect of a healthy food product introduction on the healthiness of a shopping basket as compared to the effect of a neutral and unhealthy introduction. The main research question guiding this study was:
What is the effect of the introduction of a healthy food product during consumer’s shopping trip on the healthiness of a shopping basket?
Moreover, this study is also interested in whether the timing of the introduction and health consciousness will have an effect on the healthiness of a shopping basket. As such, a sub research question was developed:
Do timing and/ or health consciousness have a moderating role on the relationship between the introduction of a food and the healthiness of the shopping basket?
food categories. Each category carried three options, ranging from relatively unhealthy to relatively healthy, from which respondents could chose. To test the effects of the healthy food product introduction, one category was chosen that allowed for the introduction of a new product. The effect of the healthy food product introduction on the healthiness of the shopping basket was offset against the effects of a control condition and a relatively neutral- and unhealthy food product introduction. In total, 304 respondents participated, but only 282 completed the survey. Results of the study revealed no significant main effect for the healthiness of the introduction As such, the hypothesis ‘The introduction of a healthy food product during consumers’ shopping trip will have a more positive effect on the healthiness of a shopping basket than the control condition and than the introduction of a neutral- and unhealthy food product’ is rejected. In addition, there were no significant moderating effects found for timing and for health consciousness on the relationship between the healthiness of an introduction and the healthiness of a shopping basket. As a result, all of the hypotheses guiding this study are unsupported and the main research question for this study can now be answered:
The introduction of a healthy food product showed some more positive effect on the healthiness of a shopping basket as compared to the introduction of a relatively unhealthy and neutral food product introduction and to the control condition, but the effect was insignificant and so were the interaction effects.
Results did indicate that early introduction of the healthy food product resulted in a slightly less healthy shopping basket, but the differences were too small to support the licensing effect theory from Khan and Dhar (2006). Moreover, health consciousness did not have a statistical significant moderating effect on the relationship between the healthiness of the introduction and the healthiness of a shopping basket. However, results indicated a positive relationship between health consciousness and the healthiness of a shopping basket. Though, only 2.3% of the between subjects variance could be accounted for by health consciousness. Nevertheless, this suggests that supermarkets and public health interventions by (inter) national institutions should make attempts in making consumers more health conscious.
5.2 Limitations
accurate results, the setting was rather artificial. First of all, respondents were asked to shop from a pre-defined shopping list, which appeared in randomized order. Therefore, respondents were not able to choose freely, whereas in a normal grocery shopping setting, people are free to choose their own products. Moreover, in a real-life setting, people can substitute products with other choices or put the products back in store if they change their mind before they go the counter. This may have an impact on the healthiness of a shopping basket.
Another limitation of this study is that the introduction of one extra food item may have been too small to measure its effect on the healthiness of a shopping basket. Taking into account that this study aimed to test whether supermarkets could steer consumers to make healthier product choices, it may have been better to introduce more extra healthy options. If supermarkets are to promote healthier living amongst their customer, they will probably increase access to healthy foods not simply by just adding one relatively healthy food product to their assortment. They will rather introduce an entire new range of healthy food products or increase their healthy food product assortment by different products. Hence, it would have been more realistic if more relatively healthy products were added to certain categories or if an entire new healthy product category was introduced during respondent’s online shopping trip.
Furthermore, the results of this study may slightly be inaccurate, as the ‘healthiness of a shopping basket’ was not normally distributed for each of the combinations of the three independent variables of this study. As such, the three-way ANOVA that has been applied to test the result may not have been the appropriate method. Even though this is less likely to influence the results to a large extent.
5.3 Managerial implications
increase consumer’s health consciousness, for example by providing more information on the products, in-store nutritionists or by marketing campaigns.
5.4 Theoretical contributions
Many studies have investigated how in-store food marketing can influence food-purchase behaviors given the alarming rise in obesity. However, this study was one of the first that examined the effect of a healthy food product introduction and its timing on the healthiness of a shopping basket, thereby focusing on the role supermarkets can take to steer consumers toward making healthier food purchases. Results from this study are not significant, indicating that it is not easy for supermarkets to improve the healthiness of consumer’s shopping basket simply by increasing access to healthy foods. Nevertheless, this study brought some relevant insight into the relationship between health consciousness and the healthiness of a shopping basket as this is found to be positively related. Moreover, this study took a different perspective than previous studies conducting research on obesity, by focusing on the impact of a supermarket’s healthy food product assortment on the healthiness of a shopping basket. More research is needed to find out how supermarkets and institutions can make consumers more health consciousness and how supermarkets can actually steer consumers toward buying relatively healthier products.
5.5 Future research
5.6 Conclusion
Reference list:
Basu, S., and Lewis, K. (2014). “Reducing Added Sugars in the Food Supply Through a Cap-and-Trade Approach.” American journal of public health, 104(12), 2432-2438.
Baumeister, R. F., Bratslavsky, E., Muraven, M., and Tice, D. M. (1998). “Ego depletion: Is the active self a limited resource?” Journal of Personality and Social Psychology, 74, 1252– 1265.
Bettman, J.R., M.F. Luce, and J.W. Payne (1998), “Constructive consumer choice processes,” Journal of Consumer Research, 25(3), 187–217.
Brownell, K.D., Koplan, J.P., (2011). “Front-of-package nutrition labeling—an abuse of trust by the food industry.” New English Journal for Medicine, 364, 2373–2375.
Center For Disease Control and Prevention, 2015. Body Mass Index. Center For Disease Control and Prevention [Online] (Updated July 16, 2015) Available at: <
http://www.cdc.gov/healthyweight/assessing/bmi/> [Accessed on March 24, 2015].
Chysochou, P (2010). “Food health branding: The role of marketing mix elements and public discourse in conveying a healthy brand image,” Journal of Marketing Communications, 16(1-2), 69–85.
Dutta-Bergman, M. (2004). “Primary sources of health information: Comparison in the domain of health attitudes, health cognitions, and health behaviors.” Health Communication, 16, 273-288.
Field, A., 2009. Discovering Statistics using SPSS, 3rd
ed. London: Sage
Fischer, M., Völckner, F., Sattler, H (2010). “How Important Are Brands? A Cross- Category, Cross-Country Study,” Journal of Marketing Research, 47(5), 823-84.
Finkelstein, S.R. and Fishbach, A. (2010). “When healthy eating makes you hungry.” Journal of Consumer Research, 37 (3), 357-367.
Food Standards Agency, 2015. Signposting and traffic light labeling. Food Standards Agency [Online] (Updated January 2010) Available at: <
http://www.food.gov.uk/northern-ireland/nutritionni/niyoungpeople/survivorform/bestreadbefore/signposting> [Accessed on March 2, 2015]
Hieke S., Wilczynski P. (2012). “Color Me In - an empirical study on consumer responses to the traffic light signposting system in nutrition labeling.” Public Health Nutrion, 15(5), 773-82.
Hoaglin, D.C., Iglewicz, B., and Tukey, J.W. (1986). “Performance of some resistant rules for outlier labeling.” Journal of American Statistical Association, 81, 991-999.
Hoaglin, D. C., and Iglewicz, B. (1987). “Fine tuning some resistant rules for outlier labeling.” Journal of American Statistical Association, 82, 1147-1149.
Hollywood, L.E., Cuskelly, G.J., O’Brien, M., McConnon, A., Barnett, J., Raats, M.M., and Dean, M. (2013). “Healthful grocery shopping. Perceptions and barriers.” Appetite 70, 119-126.
Hui, S. K., Bradlow, E. T., & Fader, P. S. (2009). Testing behavioral hypotheses using an integrated model of grocery store shopping path and purchase behavior. Journal of Consumer Research, 36(October), 478–493.
Iyengar, S., & Lepper, M. (2000). “When choice is demotivating: Can one desire too much of a good thing?” Journal of Personality and Social Psychology, 79, 995–1006.
Jayanti, R. K., & Burns, A. C. (1998). “The antecedents of preventive healthcare behavior: An empirical study”. Academy of Marketing Science Journal, 26, 6–15.
Khan, U., and Dhar, R. (2006). “Licensing effects in consumer choice.” Journal of Marketing Research, 43(2), 259-266.
Malhotra, N. B. (2007). Marketing research, an applied approach (3rd European edition). Harlow: Pearson Education.
Michaelidou, N., and Hassan, L.M., (2008). “The role of health consciousness, food safethy concern and ethical identity on attitudes and intentions towards organic food.” International Journal of Consumer Studies, 32(2),163-170.
Newsom, J.T., McFarland, B.H., Kaplan, M.S., Huguet, N., and Zani, B. (2005). “The health consciousness myth: implications of the near independence of major health behaviours in the North American population.” Social Science and Medicine, 60, 433–437.
Nederlands Instituut voor onderzoek van de gezondheidszorg, 2015. Dikke dokter maakt weinig werk van obesitas, NIVEL [Online] (Updated on the 20th of February 2014) Available at: <http://www.nivel.nl/nieuws/dikke-dokter-maakt-weinig-werk-van-obesitas> [Accessed on February 12, 2015].
Palou, A., Bonet, M.L., and Pico, C. (2009). “On the role and fate of sugars in human nutrition and health. Obesity reviews, 10, 1-8.
Payne, C.R., Niculescu, M., Just, D.R., Kelly, M. (2014). “Shopper Marketing Nutrition Interventions.” Physiology and Behavior, 136, 111-120.
Roberto, C.A., Bragg, M.A., Schwartz, M.B., et al., (2012). “Facts up front versus traffic light food labels: a randomized controlled trial.” Am. J. Prev. Med. 43, 134–141.
Sonnenberg, L., Gelsomin, E., Levy, D. E., Riss, J., Barraclough, S., and Thorndike, A., (2013) “A traffic light food labeling intervention increases consumer awareness of health and healthy choices at the point-of-purchase” Preventive Medicine, 57, 253–257.
Statistics.laerd.com, (2015). How to perform a three-way ANOVA in SPSS Statistics | Laerd Statistics. [online] Available at: https://statistics.laerd.com/spss-tutorials/three-way-anova-using-spss-statistics.php [Accessed on June 1, 2015].
Thorndike, A.N., Sonnenberg, L., Riis, J., Barraclough, S., and Levy, D.E., (2012). “A 2-phase labeling and choice architecture intervention to improve healthy food and beverage choices.” Am. J. Public Health, 102, 527–533.
Togo, P., Osler, M., Sørensen, T.I.A., and Heitman, B.L., (2001). ‘Food intake patterns and body mass index in observational studies.’ Journal of Obesity, 25(12), 1751-1751.
Van Herpen, A. , and Van Trijp., H.C.M. (2011). “Front-of-pack nutrition labels. Their effect on attention and choices when consumers have varying goals and time constraints.” Appetite, 57, 148-160.
Van Trijp, H.C.M., and Van der Lans, I.A. (2007). “Consumer perceptions of nutrition and health claims,” Appetite, 48 (3), 305–24.
Vohs, K. D., and Faber, I.R. (2007). “Spent Resources: Self Regulatory Resource Availability Affects Impulse Buying,” Journal of Consumer Research, 33 (4), 537 -547.
Vohs, K, D., Baumeister, F.R., Schmeichel, B.J., Twenge, J.M., Nelson, M.N., and Tice, D.M., (2008), "Making Choices Impairs Subsequent Self-Control: A Limited-Resource Account of Decision Making, Self-Regulation, and Active Initiative," Journal of Personality and Social Psychology, 94 (5), 883.
World Health Organization, 2015. Obesity and overweight factsheet. World Health Organization[Online] (Updated January 2015) Available at:
Appendices
I. Questionnaire – cookies category
Cookies category with the introduction of a new relatively unhealhty cookie:
Cookies category with the introduction of a new neutral cookie (control variant):
II. Total Questionnaire – control condition
