T HE INFLUENCE OF IMPULSE BUYING TENDENCY ON FOOD WASTE .

T HE INFLUENCE OF IMPULSE BUYING TENDENCY ON FOOD WASTE .

by

Jorinde van Brink

University of Groningen

Faculty of Economics and Business

Msc Marketing

June, 2015

Binnenweg 29A 7384BM Wilp (06) 30 92 3000 j.van.brink@student.rug.nl

S2586908

- 1 -

A BSTRACT

Despite the growing problem of food waste and the increasing attention and interest of researchers, little research is performed on the influence of customer characteristics on food waste (Koivupuro et al, 2012). Therefore, this is the first study that investigates the influence of customer characteristics on food waste including how frequency of shopping, health motivation and the use of a shopping list impact this relationship. With a food waste diary and an additional questionnaire, the quantity and composition of food waste in 22 categories is measured and analyzed. The results of the multiple regression analysis show that a higher impulse buying tendency of a person increases food waste. Contradicting expectations, a higher frequency of shopping weakens, and planning more in advance strengthen the relationship between impulse buying tendency and food waste. Health motivation does not affect that relationship.

Key words: Food waste, Impulse buying tendency, Frequency of shopping, Health motivation, Self- determination theory, Shopping list.

- 2 -

M ANAGEMENT S UMMARY

Food waste is becoming a serious problem all around the world. In the period 2010-2012 the FAO estimated that on average 1.3 billion tons of edible food was wasted worldwide (FAO, 2009). Reducing food waste is a key element for making the food system more efficient and can have an impact on food losses in the entire food value chain (Koivupuro et al, 2012;

Beretta et al, 2013).

Although a lot of research is conducted on the influence of wasting or losing food in the food supply chain on the environment and economy, little research is performed on the influence of customer characteristics on food waste (Koivupuro et al, 2012). The findings on the influences of customer characteristics can be very useful to understand consumer attitudes and behavior in relation to (reducing) food waste. Therefore, the problem statement and aim of this study is to answer the following question: ‘Does the impulse buying tendency of a person influence food waste and how do frequency of shopping, health motivation and the use of a shopping list impact this relationship?’.

In this research, 87 Dutch respondents filled in a food waste diary and an additional questionnaire. Similar to former expectations, this research found that a higher impulse buying tendency of a person will increase food waste. Contradicting expectations, a higher frequency of shopping will weaken the relationship of impulse buying tendency on food waste meaning that a higher frequency of shopping decreases food waste. In addition, planning more in advance will strengthen the relationship which increases food waste. Health motivation does not impact this relationship. However, a higher autonomous motivation decreases food waste.

Several implications can be derived from this study to help decreasing food waste.

First, individuals with a high impulse buying tendency are more tempted to make unplanned purchases and therefore waste more food. Decreasing environmental stimuli (i.e. banners, advertisement) will decrease unplanned purchases of individuals with a high impulse buying tendency and therefore food waste (Youn & Faber, 2000; Beatty & Ferrell, 1998; and Rook &

Fisher, 1995). Secondly, individuals need to be encouraged to increase the frequency of shopping and to plan less in advance. Finally, individuals need to be motivated to create a healthier lifestyle. For example, advertisement and promotional actions can encourage individuals to start a diet or eat healthy which increase autonomous motivation and therefore decrease food waste.

- 3 -

P REFACE

The master thesis in front of you is the final product and conclusion of my Master Marketing Management at the University of Groningen. In 2013 I graduated for the HBO study Business Administration, Management, Economy and Law at Windesheim in Zwolle. After this study I realized that I hadn’t finished learning. Therefore I decided to specialize and start with the Pre-master and Master Marketing at the University of Groningen in September 2013. During my studies I worked at a Marketing and Communication department in an International organization where my interest in Marketing has grown. Because of my experiences in the past two years I can easily conclude that the master Marketing Management was a perfect study for me.

The past year I enjoyed a lot of courses which have gained me extensive knowledge on the subject of Marketing. In particular, retail marketing and brand management are courses I really enjoyed and appreciated. In February 2015 I joined the master thesis group with the topic ‘Food waste’ and decided to combine my interest in retail marketing with this topic. In addition, the course Marketing Research Methods has increasingly caught my interest and I hope to improve my skills in this area in the upcoming years. Although I enjoyed my time as a student of the master Marketing Management I am ready for my next challenge.

I would like to take the opportunity to thank the people who have supported, guided and helped me during my entire study and writing my master thesis. Firstly, I would like to thank my supervisor dr. Jenny van Doorn for her expertise, guidance, useful feedback and valuable support throughout the process. Furthermore, I would like to thank my fellow students in the master thesis group for providing ideas, support and feedback during the period of approximately six months. Thirdly, I thank my friends and family for participating in my 21- day food waste study, believing in me and cheering me up during stressful periods. Last but not least I would like to thank Gijs Koning and my parents, Albert en Irma van Brink, for motivating me, understanding that I wasn’t always available when I was awfully busy, calming me down when I felt stressed and the spellings checks during my study and writing this master thesis. You have been a great support.

Jorinde van Brink

- 4 -

T ABLE OF C ONTENT

ABSTRACT ... - 1 -

MANAGEMENT SUMMARY ... - 2 -

PREFACE ... - 3 -

TABLE OF CONTENT ... - 4 -

1. INTRODUCTION ... - 6 -

2. THEORETICAL FRAMEWORK... - 9 -

2.1 IMPULSE BUYING TENDENCY AND UNPLANNED PURCHASES ... -10-

2.1.1 The Definition of Impulse Buying Tendency ... - 10 -

2.1.2 Impulse Buying Tendency and Unplanned Purchases ... - 11 -

2.2 FOOD WASTE ... -12-

2.3 DRIVERS THAT IMPACT UNPLANNED PURCHASES ... -14-

2.3.1 Frequent and Infrequent shoppers ... - 15 -

2.3.2 Health Motivated Shoppers ... - 16 -

2.3.3 The Use of a Shopping List... - 18 -

3. METHODOLOGY ... - 20 -

3.1DATA COLLECTION AND STUDY DESIGN. ... -20-

3.2DESCRIPTIVE STATISTICS ... -26-

3.3METHOD ... -28-

4. RESULTS ... - 28 -

4.2MAIN EFFECT IMPULSE BUYING TENDENCY ... -28-

4.3IMPACT OF MODERATING VARIABLES ... -29-

4.3.1 Frequent and Infrequent Shoppers ... - 29 -

4.3.2 Health Motivated Shoppers ... - 30 -

4.3.4 The Use of a Shopping List ... - 32 -

4.3.4 Effect of Demographic Variables ... - 32 -

5. DISCUSSION ... - 33 -

- 5 -

6. CONCLUSION ... - 36 -

6.1ACADEMIC CONTRIBUTION ... -36-

6.2 MANAGERIAL CONTRIBUTION ... -37-

6.3LIMITATIONS AND FUTURE RESEARCH ... -38-

REFERENCES ... - 40 -

APPENDIX ... - 47 -

APPENDIX A:ACCOMPANYING LETTER ... -47-

APPENDIX B:FOOD WASTE DIARY ... -48-

APPENDIX C:QUESTIONNAIRE FOOD WASTE ... -51-

APPENDIX D:PRINCIPAL COMPONENT ANALYSIS TSRQSCALE ... -60-

- 6 -

T HE INFLUENCE OF IMPULSE BUYING TENDENCY ON FOOD WASTE .

1. I NTRODUCTION

Food waste is becoming a serious problem all around the world. In the period 2010-2012 the Food and Agriculture Organization of the United Nations (FAO) estimated that on average 1.3 billion tons of edible food was wasted worldwide (FAO, 2009). This amount of wasted food is approximately one third of the total amount of food produced in the same years for the entire human consumption (Munesue et al., 2015). Koivupuro et al. (2010) and Gustavsson et al. (2011) stated that wasting edible food has an impact on different factors including the environment and economy. One example is that when edible food is not consumed all the environmental impacts of producing the materials for eating and processing the food are wasted and useless. In addition, wasting edible food is not only affecting the environment but also the economic development and food protection of poor people in less developed areas (Gustavsson et al., 2011). This means that wasting or losing food takes away the opportunity for less developed areas to have access to food and results in significant reduction of resources such as land, water and fossil fuels. Another negative aspect is that food waste increases the greenhouse gas emissions which are associated with the food production processes (Farr-Wharton et al., 2014).

Currently, the interest of researchers in the issues of food losses and food waste is increasing (Munesue et al, 2015). The research of Munesue et al. (2015) shows that reductions in food waste in developed areas not only decreases the number of undernourished people in less developed areas up to 63 million but also leading to a decrease in harvesting resources such as water, utilization and gas emissions. Food waste also influences the financial resources of households. The average value of wasted food per year in a household of Austria is $616 (€581) (Baker et al, 2009), in the US $590 (€557) (Jones, 2005) and in Britain £480 (€662) (Wrap, 2009). In addition, the amount of wasted food in 2010 in the Netherlands is on average 112.5 kilogram, which is €338, per household (Monier, 2010).

Narrowing this down to one person in 2013 it can be concluded that the average amount of wasted food per person in the Netherlands is 50 kilogram which is €150 (Rijksoverheid, 2014).

- 7 -

According to the literature above it is suggested that reducing food waste can reduce the wastage of natural resources and undernourished people. In addition, reducing food waste can increase financial resources in a household (Koivupuro et al, 2012). Reducing food waste is therefore a key element for making the food system more efficient and can have an impact on food losses in the entire food value chain (Koivupuro et al, 2012; Beretta et al, 2013). In less developed countries food loss occurs in the early stages of the Food Supply Chain (FSC) due to limited financial, managerial and technical resources. In the more industrialized countries food loss occurs at the last stage of the FSC, which is the consumption phase, and is mostly caused by non-correct behaviour, such as buying or cooking too much food or storage conditions (Principato et al, 2015; Koivupuro et al, 2010; Gustavsson et al, 2011).

In the academic literature of food waste several definitions are formulated. Griffin et al. (2009), Partfitt et al. (2010), and Gustavsson et al. (2011) are stating that when food is lost at the end of the phase in the FSC, which is the distribution, retail and final consumption stage, this is defined as food waste. Another definition is coming from the research of Grolleaud (2002) which defines the concept of food waste as ‘the decrease in food quantity or quality, which makes it unfit for human consumption’. However, in this study food waste is defined according to the definition of Koivupuro et al. (2012), and Silvennoinen et al. (2014).

This definition is used because it is in line with the definition used in preceding researches (Parfitt et al., 2010: Koivupuro et al., 2012; Silvennoinen et al., 2014) and focuses on avoidable food waste and excluding unavoidable food wastes such as coffee grounds and peel.

Therefore food waste is defined as: “All kinds of eatable material and food which could be used for human consumption and could be avoidable of being disposed by a household”.

As mentioned before a lot of research is done on food waste. Parfitt et al. (2010) investigate the situation of food waste in the FSC and conclude that food waste is much higher at the post- harvest stages in developing countries. In developed countries, the post- consumer food waste accounts for the greatest food losses. In addition, the study of Gustavsson et al. (2011) estimated the food losses over the entire food value chain in seven different regions including Europe. Gustavsson et al. (2011) conclude that food waste is the highest for fruit and vegetables and that knowledge gaps throughout the FSC are increasing food waste.

Although a lot of research is done on the influence of wasting or losing food in the FSC on the environment, little research is performed on the influence of customer characteristics on food waste (Koivupuro et al, 2012). The contribution of this research is to

- 8 -

investigate the influence of the customer characteristic ‘impulse buying tendency’ on food waste. Impulse buying occurs when individuals are making unintended, unreflective and immediate purchases (Rook & Fisher, 1995). A lot of food waste begins with making choices in the grocery store that can be a result of impulse buying tendency which results in purchasing items or quantities that consumers are unlikely to consume (Gunders, 2012). The research of Schneider (2008) suggests that when an individual has a high impulse buying tendency this increases food waste. In addition, Parfitt et al. (2010) and Lyndhurst et al.

(2007) suggest that the amount of unplanned purchases is influenced by a persons’ impulse buying tendency and that therefore not only unplanned purchases but also impulse buying tendency affect the amount of food waste. However, no research has been done on the direct influence of impulse buying tendency on food waste. Therefore, the aim of this research is to investigate the influence of a person’s impulse buying tendency on food waste.

Strategies to avoid unplanned purchases. It is expected that frequency of shopping, health motivation and the use of a shopping list will impact the relationship of impulse buying tendency on food waste. The research of Stilley et al. (2010) states that when consumers decrease the frequency of shopping and increase the use a shopping list (Inman et al., 2009) the amount of unplanned purchases will decrease. In addition, Gunders (2012) suggests that when consumers are planning meals, use shopping lists and avoid unplanned purchases or marketing tactics unplanned purchases will decrease which can decrease overbuying and this will help to reduce the amount of food that is discarded in the household. Health motivation can also affect the influence of impulse buying tendency on food waste. It is suggested that a higher amount of health motivation will decrease unplanned purchases and a lower amount of health motivation will increase the amount of unplanned purchases (Finkelstein & Fishbach, 2010; Loewenstein, 1996; Verplanken et al., 2005; Wansink & Huckabee, 2005). Combining these findings it is expected that frequency of shopping, health motivation and use of a shopping list are influencing unplanned purchases which affects the influence of impulse buying tendency on food waste.

The focus of this research lies on the influence of impulse buying tendency of a person on food waste. It is suggested that a person’s impulse buying tendency increases the amount of wasted food in a household. Additional, the impact of the frequency of shopping, health motivation and the use of a shopping list on the main relationship is investigated. Therefore, the problem statement and aim of this study is to answer the following question:

- 9 -

Does the impulse buying tendency of a person influence food waste and how do frequency of shopping, health motivation and the use of a shopping list impact this relationship?

To answer the research question above four sub questions are formulated:

1. How does the impulse buying tendency of a person influence food waste?

2. How does the frequency of shopping impact the effect of impulse buying tendency on food waste?

3. How does health motivation impact the effect of impulse buying tendency on food waste?

4. How does the use of a shopping list impact the effect of impulse buying tendency on food waste?

In order to answer the research questions four hypotheses are formulated. These hypotheses will be tested by making use of a questionnaire and food waste diaries. From an academic point of view this research gives insights into the subject of food waste from a marketing perspective. In addition, it provides evidence for the relationship between the impulse buying tendency and the amount of wasted food and the influence of frequency of shopping, health motivation and the use of a shopping list on this relationship. From a managerial point of view this research contributes and can offer solutions about how to reduce food waste.

Findings from this study can help marketers to understand the impact of impulse buying tendency which can be used to develop consumer profiles, programs and communicate effectively how people can reduce food waste.

The research is structured as follows; first, the theoretical background is giving a clear and detailed explanation about impulse buying tendency, food waste, frequency of shopping, health motivation and use of a shopping list. Secondly, the methodology part describes the research design. Thirdly, the results of the questionnaire and food-waste diary are presented followed by the conclusion and discussion. Finally, the limitations, implications for further research and marketing implications are given.

2. T HEORETICAL F RAMEWORK

- 10 -

2.1 I

B

T

U

P

Researchers believe that consumers not only differ in the behavior to act impulsive but also in the degree of impulse buying tendency (Beatty & Ferrell, 1998; Puri, 1996; Rook, 1987; Rook

& Fisher, 1995; Rook & Gardner, 1993; Weun et al., 1998). Research states that impulse buying tendency is a personality trait indicating that consumers can be distinguished from other consumers based on this trait (Verplanken & Herabadi, 2001; Beatty & Ferrell, 1998;

Puri, 1996; Rook & Fisher, 1995; and Rook & Gardner, 1993). Consistent with existing research impulse buying tendency can be defined as ‘the degree to which an individual is likely to make unintended, immediate, and unreflective purchases (i.e. unplanned purchases) (Rook & Fisher, 1995; Rook, 1987). The research of Rook & Fisher (1995) and Rook (1987) concludes that there is a relationship between the amount of impulse buying tendency and unplanned purchases meaning that when a person has an impulse buying tendency these results in a higher amount of unplanned purchases.

A lot of research is done regarding the concept of impulse buying tendency.

Researchers have provided a theoretical framework for examining the relationship between impulse buying tendency and psychological variables (Puri, 1996), personality traits and cues (Youn & Faber, 2000) and situational variables in a shopping context (Beatty & Ferrell, 1998;

Rook & Fisher, 1995). Puri (1996) found that when the benefits of impulse buying tendency are the pleasure of yielding a temptation, hedonic individuals responded to appeals that decreases this desire. In addition, they found that when the costs of yielding to impulse buying tendency are probable individuals are more likely to act that way and have more unplanned purchases. The research of Youn & Faber (2000) concluded that several types of an individual’s internal states and environmental stimuli can serve as cues for triggering unplanned purchases because of the presence of impulse buying tendency. Internal cues include an individual’s positive and negative feelings and environmental cues consist of atmospheric cues such as a retail setting, marketer-controlled cues and marketing mix stimuli.

In addition, the research of Beatty & Ferrell (1998) and Rook & Fisher (1995) found that situational variables, such as time and money availability, and individual differences, such as shopping pleasure and impulse buying tendency, are influencing a persons’ felt urge to buy on impulse and make unplanned purchases. Regarding the demographic determinants of impulse buying tendency the research of Bashar et al. (2013) concludes that individuals who have enough disposable income and are medium in age are more likely to buy unplanned purchases

- 11 -

and have a higher impulse buying tendency. However, because of the fact that there is no research conducted on the influence of the internal trait, impulse buying tendency, of a person on the amount of wasted food this will be the key topic of this research.

2.1.2IMPULSE BUYING TENDENCY AND UNPLANNED PURCHASES

Impulse buying tendency influences the amount of unplanned purchases which are extremely important for supermarkets due to the fact that 38.7% of purchases in a supermarket are bought on impulse (Bellenger et al, 1978; Rook & Fisher, 1995; Rook, 1987). This high percentage of unplanned purchases can result from impulse buying tendencies which increase the focus of retailers and academic research on the concept of unplanned purchases (Jones et al, 2003).

Wu et al. (2013) are defining impulse buying behavior as ‘an emotional, irrational purchasing behavior in which the consumer responds immediately and impulsively without thinking of consequences’. This research concludes that consumers are buying products unplanned when being reminded of a need when seeing a product through promotional activities, advertisements or related information. In addition, Bucklin and Lattin (1991) conclude that unplanned purchasing is the opposite of planned purchases which are decisions made before entering a store. Another definition of unplanned purchases which is largely supported by academic researchers and therefore used in this research is given by Rook and Fisher (1995). According to this definition unplanned purchases can be defined as ‘an individual is making unintended, unreflective and immediate purchases’ (Rook and Fisher, 1995; Rook, 1987). In this definition, ‘unintended’ refers to the situation that a consumer is making a purchase while shopping and was not actively looking for that item, had no plans to buy it and is not engaged in the shopping task. This situation can arise when a consumer has a sudden urge for buying a specific item during a shopping trip. Unreflective refers to a purchase that is made by the consumer without evaluating the purchase. In this situation the consumer is buying on impulse and is less likely to think about the consequences before purchasing meaning that the consumer is only focusing on the immediate fulfillment of the urge. The last aspect of this definition, immediate, refers to the short time between seeing the item and purchasing it.

The definition of Rook & Fisher (1995) is supported by the research of Karbasiver et al. (2011) which also formulated five elements of the process of unplanned purchases which are: 1) a spontaneous and sudden desire in store to act in a certain way 2) the state of psychological disequilibrium, 3) the onset of psychological struggle and conflict, 4) a

- 12 -

decrease in cognitive evaluation and 5) a lack of regard for the consequences of the unplanned purchases.

When an individual with a high impulse buying tendency is buying products without planning it in advance, it is most likely that this products are categorized as vice instead of virtue (Wertenbroch, 1998). The categorization of vice and virtue products is important because it can influence the amount of food waste for these individuals. Vices, which are also known as ‘wants’, are products that provide an immediate satisfying experience for the individual but are contributing to a negative long- term outcome. An example is a good taste of chocolate cake which results in weight gain or health issues. Virtue products are also known as ‘should’ products and less pleasurable and satisfying in the short term. However, these products have a less negative long- term outcome as compared to vice products (Milkman et al., 2008; Okada, 2005; Wertenbroch, 1998). The findings of Wertenbroch (1998) are consistent with the research of Milkman et al. (2008) which concludes that unplanned purchases for individuals with a high impulse buying tendency are mostly made in the category vices.

2.2 F

W

Food waste occurs through all stages of the food life cycle from harvesting, through processing and production via trade and lastly consumption (Schneider, 2008). On an annual basis, the average amount of wasted food in Europe is 90 million tons (European Commission, 2014). The highest proportion of food waste occurs in the category of fresh, fruits and vegetables followed by bakery and dairy products, meat and fish. (Pekcan et al, 2006).

Food waste, sometimes defined as food disposal, is the last stage of the food chain after the stages production and consumption (Cappellini, 2009). Wasting or disposing food is an important part of the consumption chain because the chain is a continuous cycle. This means that food disposal influences a consumer’s pre and post consumption behavior and decisions (Munro, 1995; Gregson et al, 2006; Cappellini, 2009). Some of the food that is wasted is not appropriate for human consumption (Schneider, 2008) meaning that incorrectly labelled or damaged food during transport can also increase the amount of food waste. The research of Schneider & Lebenorger (2011) divided food waste into avoidable and unavoidable categories based on the edibility of the food. Avoidable food waste can be defined as discarded food that has been edible in the past. Unavoidable food waste refers to

- 13 -

for example vegetable peel, bones and coffee grounds that is inedible. Within the literature food waste is also defined as ‘the decrease in food quantity or quality, which makes it unfit for human consumption’ (Grolleaud, 2002). However, in this study food waste is defined according to the commonly used definition of Koivupuro et al. (2012), and Silvennoinen et al.

(2014) as: “All kinds of eatable material and food which could be used for human consumption and could be avoidable of being disposed by a household”. This definition is used because it focuses on avoidable food waste and exclude unavoidable food waste such coffee grounds and peel.

It is mentioned that in the later stages of the food supply chain, food waste is mostly related to behavioral issues (Grolleaud, 2002). The European Commission (2014) stated that the customer is responsible for approximately 40% of the food waste. However, in the research of Gustavsson et al. (2011) this percentage is even higher with 50% of food losses and in the research of Griffin et al. (2009) it is almost 60% of the disposed food. This high percentage results from the fact that households are purchasing too much food which results in a surplus and has to be disposed (Evans, 2011). However, several other factors are also important for the explanation regarding the high amount of food waste in a household.

A lot of research investigated the impact of determinants on the amount of wasted food in the FSC (Parfitt et al, 2010; Grolleaud, 2002; Gunders, 2012; Koivupuro et al, 2010) and how to quantify food waste in different units (kilogram, kcal and financial resources) (Koivupuro et al, 2012). The studies from Wenlock & Buss (1977) and Garde & Woodburn (1987) are showing that the amount of wasted food is influenced by the structure of a family indicating that adults are wasting more food compared to children. The studies also concludes that the larger the size of the household (versus smaller households) the less wasting per person and that younger people are wasting more compared to older people. Another aspect that influences the amount of food waste is household income in which it is suggested that lower-income households are having a lower amount of wasted food (Osner, 1982). Since the increasing concern about food waste a lot of studies investigate factors that influence food behaviours such as social norms, attitudes, cultural upbringing (Farr- Wharton et al, 2014), age, income and time spent at home (Schneider & Obersteiner, 2007). The research of Parfitt et al. (2010) investigated that, based on the participants in this research, promotional activities are accountable for almost 50 percent of the unplanned. As a consequence of this behavior it is likely that those customers are wasting more food (Lyndhurst et al, 2007).

- 14 -

Regarding the relationship of impulse buying tendency and the amount of food waste the research of Stefan et al. (2013) used the theory of planned behavior to predict household decisions that are made to minimize food waste. Following the theory of planned behavior it is suggested that planning purchases can help to control impulse buying tendency and therefore food waste. It is well documented in scientific studies that individuals with a high impulse buying tendency are, despite the awareness of potential negative consequences, making more unplanned purchases which are not actually needed (Baumeister et al., 1998;

Parfitt et al., 2010; Lyndhurst et al., 2007; Rook, 1987). In addition, the main reason for food waste in a household is ‘Food is not used in time’ meaning that more unplanned purchases results in food that is not eaten on time and cannot be consumed (Parfitt et al., 2010). It is therefore expected that a high impulse buying tendency results in more unplanned (food) purchases which increases the amount of food waste in a household Therefore, the following hypothesis is formulated:

H1: The higher the impulse buying tendency of a person the higher the food waste.

Figure 1 Conceptual model

2.3 D

I

U

P

It is expected that the frequency of shopping, health motivation and the use of a shopping list are influencing the amount of unplanned purchases of an individual which affect the relationship of impulse buying tendency on food waste. The research of Inman et al. (2009) concludes that when consumers are shopping more frequently and use shopping lists the

- 15 -

amount of unplanned purchases of a person will decrease. In addition, Finkelstein & Fishbach (2010), Loewenstein (1996), Verplanken et al. (2005) and Wansink & Huckabee (2005) are stating that a person’s amount of health motivation has a great influence on the amount of unplanned purchases. It is concluded that a high amount of health motivation will decrease unplanned purchases and a low amount of health motivation will increase unplanned purchases. Because of findings from the research of Inman et al. (2009), Finkelstein &

Fishbach (2010), Loewenstein (1996), Verplanken et al. (2005) and Wansink & Huckabee (2005), it is suggested that the frequency of shopping, health motivation and use of a shopping list will impact unplanned purchases and strengthen or weaken the influence of impulse buying tendency on food waste.

2.3.1 FREQUENT AND INFREQUENT SHOPPERS

Previous researchers have shown that there are frequent shoppers and infrequent shoppers (Byung-Do & Kyungdo, 1997) and that one characteristic of a consumers shopping experience is the repetition at regular time intervals (e.g. once a weak) (Park et al, 1989). The frequency of shopping can differ per person and is defined as the number of purchase occasions from an individual in one particular store (Inman et al, 2009).

The amount of unplanned purchases should depend on the shopping frequency of an individual (Inman et al., 2009). When impulsive buyers are shopping more frequently this decreases the amount of unplanned purchases on a certain store visit because these buyers only buy the items he or she needs (Inman et al., 2009). In contrast to this finding, Park et al.

(1989) suggest that when an individual with a high impulse buying tendency is shopping more frequently in stores the search process is primarily guided by internal memory which requires a low effort and increases the ability of in-store decision making activities and therefore unplanned purchases. However, Gutierrez (2004) stated that individuals with a high impulse buying tendency, who have a regular pattern of shopping, have an easier time in searching products in the store and tend to find the preferred products faster. In addition, these individuals who are shopping in stores more frequently are tending to plan desired purchases which decrease unplanned purchases.

There are individual differences in the amount of unplanned purchases (Verplanken &

Herabadi 2001) meaning that individuals with a high impulse buying tendency appears to be influenced by store visit frequencies. It is concluded that the emphasis on the frequency of shopping in a grocery store is an important driver of whether impulsive buyers are making unplanned purchases (Mario, 2008). When individuals with a high impulsive buying tendency

- 16 -

are shopping more frequently, it is more likely that they are making more unplanned purchases (Mattila & Wirtz, 2008; DeSarbo & Edwards, 1996; Bellenger et al., 1978;

Skallerud et al., 2009)

There is a lot of research stating that a higher frequency of shopping will increase unplanned purchases. However, it is also suggested that there are ways to control an individual’s impulse buying tendency and thus reduce the amount of unplanned purchases (Verplanken & Sato, 2011; Rook & Fisher, 1995). Vohs & Faber (2007) stated that a high self- control, such as the willpower to decrease the frequency of shopping and therefore the unplanned purchases, helps individuals to actually resist the desire to make unplanned purchases (Vohs & Faber, 2007). It is well documented in the literature that individuals employ techniques, such as lowering the frequency of shopping, to help resist the urge and desire of making unplanned purchases (Baumeister et al., 1998; Loewenstein, 1996;

Wertenbroch, 1998). Stilley et al. (2010) also show that a routinized activity and therefore a lower frequency of shopping can control the unplanned purchases of an impulse buyer. It is concluded that a lower amount of store visits (i.e. frequency of shopping) reduces the urge and desire of an individual with a high impulse buying tendency to make unplanned purchases. Therefore, if this individual is shopping on a routinized nature and has a lower frequency of shopping, which mostly happens with grocery shopping, past trips are used to decide which products are needed and unplanned purchases are decreasing (Cyert & March, 1963; Wildawsky, 1964).

According to the previous literature it is suggested that when an individual with a high impulse buying tendency has a lower frequency of shopping (vs. high frequency) this will decrease (vs. increase) the amount of unplanned purchases (Desarbo & Edwards, 1996; Rook

& Hoch, 1985; Bellenger et al, 1978; Lyndhurst et al, 2007). Hence, it is expected that a lower frequency of shopping will decreases unplanned purchases which will decrease food waste.

Therefore the following hypothesis is formulated:

H2: The influence of impulse buying tendency on food waste is weaker when a person has a lower frequency of shopping.

2.3.2 HEALTH MOTIVATED SHOPPERS

- 17 -

Health motivation is one of the most important factors that influence people’s food choice (Brug, 2008). The aspect of health involvement or a person’s interest in eating healthy are important factors in the decision making process of buying and eating food (Saba & Vassallo, 2012). In the research of Moorman & Matulisch (1993) the concept of health motivation is defined as ‘consumer’s goal-directed arousal to engage in health behaviours’. This research state that this characteristic of a consumer focuses on the willingness to be interested in or perform healthy behaviour. It is suggested that health conscious individuals will seek to engage in activities that encourage and maintain a healthy life. This means that these individuals are more likely to engage in an intensive searching process for products that contain the ingredients that are seen as beneficial themselves as compared to individuals without a health motivation (Kim & Chung, 2011). In addition, Lindsay (2010) argued that eating healthy is a moral obligation and a responsibility for individuals.

A person’s health motivation may be influenced by self- efficacy. Self- efficacy refers to a person’s perception or a confidence in a person’s ability and skills to engage in healthy behaviour (Brug, 2008). In addition, health motivation is effecting and directly satisfying psychological needs and therefore often described as an autonomous goal (Vansteenkiste et al., 2006). The research of Vansteenkiste et al. (2006) shows that an individuals’ health motivation can be explained by the self- determination theory (SDT) which is divided in three different states of health motivations that influence consumer’s behaviour. SDT means that

‘an individual will experience a sense of choice in regulating and initiating an individual’s own actions’ (Aryee et al., 2015). According to this theory individuals can engage in extrinsic motivation and amotivation where extrinsic motivation is divided in controlled and autonomous motivation. Controlled motivation refers to the degree in which individuals are responsible or obligated to behave in a certain way (Williams et al., 1996) and mostly resulting from external rewards like salary or promotion (Deci & Ryan, 2000). Autonomous motivation is driven by the interest in an action for its own sake such as intrinsic motivation (Aryee et al., 2015). Regarding a person’s amotivation it is suggested that within this state an individual is not showing any motivation to conduct certain behaviour (Deci & Ryan, 2000).

Autonomous motivations can result in higher self- esteem and an improved performance (Deci & Ryan, 2000). In addition, when an individual is taking actions based on autonomous motivation this results in a higher well- being and an increase in social function which will therefore increase the overall improved psychological health for this individual.

The research of Williams et al. (1996) showed that autonomously motivated people had a

- 18 -

higher success rate regarding weight losses and a higher likelihood of maintaining this behaviour in the future. This research also concludes that autonomously motivated participants are more motivated to think about the quality of food and are eating healthier because these individuals make a habit of eating healthy food. Controlled motivated individuals are less motivated to eat healthy and the quality of the food and more focused on the quantity of it (Koestner et al., 2008).

Debabrata & Charles (2013) conclude that the amount of unplanned purchases of an individual should depend on the degree of health motivation of that individual. It is concluded that individuals with a high impulse buying tendency and a low health motivation (i.e.

controlled motivation) have a higher likelihood of making unplanned purchases. In addition, individuals with a high impulse buying tendency and a high health motivation (i.e.

autonomous motivation) make less unplanned purchases (Finkelstein & Fishbach, 2010;

Loewenstein, 1996; Verplanken et al., 2005; Wansink & Huckabee, 2005).

According to the previous literature it is suggested that when an individual has a high (vs. low) health motivation the amount unplanned purchases is lower (vs. higher). Hence, it is expected that when a person has a high health motivation this decreases unplanned purchases which will decrease food waste. Therefore, the following hypothesis is formulated:

H3: The influence of impulse buying tendency on food waste is weaker when a person has a high health motivation.

2.3.3 THE USE OF A SHOPPING LIST

When consumers are facing numerous distractions and multiple decisions, they might rely on aids, such as the shopping list, to reduce the complexity and unplanned purchases during the decision making process. Previous research defined the shopping list as a memory aid for grocery shopping and states that this shopping list can be a useful tool for helping the consumer making planned purchases. In addition, shopping lists are a specific form of tools that are defined as ‘memory aids’ which is any tool or device used to enhance remembering (Block & Morwitz, 1999).

In order to reduce the sensitivity of the impulse buying tendency a consumer may want to use pre-shopping plans such as a shopping list (Lyndhurst et al, 2007). Spiggle (1987) found that using a shopping list is associated with ex ante planning (e.g. planning before the shopping trip) which implies that without using a shopping list as memory aid consumers are

- 19 -

more likely to make unplanned purchases. Similar to this finding, it is concluded that consumers who are using a shopping list have a higher discipline in stores to buy the products that are actually needed and buy a lower amount of unplanned purchases (Parfitt et al, 2010;

Lyndhurst et al, 2007). In addition, Thomas & Garland (1993) find that consumers with shopping list are still buying fewer items and spent less money in stores as compared to shoppers without a shopping list.

Shoham & Makovec (2003) stated that the extent of consumer’s engagement in in- store decision making (i.e. spontaneous and unplanned purchases) can be used to determine the impulse buying tendencies of this person. This research concludes that the amount of unplanned purchases should depend on the degree to which consumers are pre-planning and stick to the plans or shopping lists. When high impulsive buyers are not using a shopping list and are shopping based on spontaneous buying stimuli, this means that these buyers are more open and accessible for buying unplanned products (Rook & Fisher, 1995). In addition, it is more likely that this consumers are shopping unreflective, provoked by physical closeness, under control of emotional attraction and in the need of an immediate gratification (Hoch &

Loewenstein, 1991; Thompson et al., 1990). As a result, without the use of a shopping list the individuals are more likely to actually act on impulse which means that the amount of unplanned purchases is increasing.

On the other hand, making a shopping list can help to decrease unplanned purchases for individuals with a high impulse buying tendency (Bell et al., 2011). Laran (2009) concludes that when individuals with a high impulse buying tendency are making a planning for reaching a goal (i.e. use of a shopping list) this makes achieving this goal easier.

Therefore, when these individuals use a shopping list it will decrease the likelihood of considering additional unplanned purchases (Bell et al., 2011; Laran, 2009). In addition, Ganglbauer et al. (2013) concludes that individuals with a high impulse buying tendency, who are using a shopping list, are more disciplined and therefore make less unplanned purchases in stores. In addition, these individuals only buy the products that are actually needed, stick to lists and are therefore making less unplanned purchases (Parfitt et al, 2010; Lyndhurst et al, 2007; Doron, 2013).

According to the previous literature it is suggested that when people are using (vs. not using) a shopping list the amount of unplanned purchases is lower (vs. higher) (Ganglbauer et al, 2013; Lyndhurst et al, 2007; Parfitt et al, 2010). Hence, it is expected that when a person

- 20 -

makes use of a shopping list this decreases unplanned purchases which will decrease food waste. Therefore, the following hypothesis is formulated:

H4: The influence of impulse buying tendency on food waste is weaker when a person uses a shopping list.

In figure 1 the conceptual model for this research is defined based on the previous literature review. This conceptual model gives an overview of the relationship between the dependent and independent variables including the moderators.

3. M ETHODOLOGY

3.1 D

C

S

D

.

In this research two study approaches are used. The first approach is the food waste diary and the second approach is an additional questionnaire. This study is focusing on food waste in the Netherlands and therefore only conducted among persons who are speaking Dutch. The respondents are selected via the social network of four researchers living in different parts of the Netherlands. In addition, the broad target group provides the highest probability of attracting respondents that differ in terms of food waste, impulse buying tendency, frequency of shopping, health motivation and use of a shopping list. In order to ensure that the results of the study consist of a broad variety of different respondents the first questions are about the respondent’s age, gender, education, occupation, household size, income, number of pets and time spent in store. The data collection of both approaches is collected in April and May 2015. This offers the opportunity to analyze food waste behavior three weeks in a row which makes the results more reliable. Another reason for choosing this period is the fact that it reduced the possibility that respondents are going on a major vacation (Koivupuro et al., 2012). To minimize the bias of the results, an accompanying letter is added that explains what is expected of the respondent and how the diary can be filled in (appendix A). In addition, each household can only fill in the food waste diary and questionnaire once. Due to privacy issues, personal information (i.e. name) of each respondent is deleted and the results are analyzed completely anonymous.

Food waste diary. Respondents are asked to fill in a food waste diary every day for twenty- one days (i.e. three weeks) in a row. By using food waste diaries the amount of food waste per

- 21 -

household can be measured. As mentioned by Koivupuro et al. (2012) a food waste diary is organized in such a way that it let respondents measure own food waste which results in detailed information on the composition and the reasons for food waste. This measurement method results in a detailed dataset that gives evidence for the reasons for each single food disposal.

The food waste diary of Koivupuro et al. (2012) is central during this research and divided in two main categories which are: (1) Products that you bought and (2) Products that you have thrown away. The first category is measuring how often a respondent is eating a take- away meal or eating in a restaurant. In addition, this category also measures how often and in which supermarket the respondent is buying products followed by how often and which promotional products a respondent buys. The second category, products that you have thrown away, determines the amount of food waste per milliliter or gram. It is important to note that unavoidable food waste, such as vegetable peel, fruit peel, bones and coffee ground are not taken into account in this study.

The food waste diary has also some limitations. First of all, respondents are requested to fill in the diary at a daily basis meaning that this can be very demanding and some respondents may forget or actively choose not to fill in a part of the wasted food during the research period. Another limitation can be that respondents become aware of the (high) amount of food waste per day which decreases the amount of food waste in the research period as compared to normal, unmonitored periods. It is therefore suggested that because of these limitations the amount of food waste in the research period is slightly underestimated (Koivupuro et al, 2010; WRAP, 2010).

To make sure that the food waste diary was appropriate for this study a pretest is conducted before the actual data collection. In this pretest several problems regarding the format, spelling, completeness and formulation are solved. After solving these problems the final food waste diary was validated and distributed (appendix B). As mentioned before, the respondents had to fill in the food waste diary each day for twenty-one days. In the last category of the food waste diary the respondent can fill in the amount of wasted food in milliliter or gram. The total amount of product categories included in the food waste diary in this study is twenty two. It is important to note that these product categories are responsible for 80% of the food purchases of the Dutch households (van Doorn & Verhoef, 2014) and therefore used in this diary. In addition to these product categories another label is added in which the respondent can fill in a product that is wasted but missing on the list. According to

- 22 -

the measurement scales in this study, the first seven products are measured in milliliter and the remaining sixteen product categories are measured in gram. Despite the fact that respondents are asked to weight the amount of food waste some guidelines are given (for example, 1 glass is 150ml) to get more precise results.

Questionnaire: In addition to the food waste diary an additional questionnaire is used (appendix C). The questionnaire is made by four researchers and will serve different studies.

In this research all questions, including those which are not used in this study but collected for the other researchers, are discussed in table 1. However, only the questions used in this study are explained in more detail later on. The first part of this questionnaire consists of questions regarding the demographics of a respondent. The following five questions collect data about the moderating variables, frequency of shopping and the use of a shopping list. After these questions the respondent’s attitude and propensity towards convenience products, involvement with food and cooking and impulse buying tendency is measured.

The next part collects data about how a respondent reacts on promotions, how often convenience products are purchased and what the main reason is regarding food waste and buying too much. Finally, the respondents’ health motivation, intention to maintain a healthy diet and perceptions about supermarket assortments are measured. Table 1 shows an overview of the questions, variables and sources of the questionnaire. Additionally, if the respondent wants to win a voucher for a dinner or check how they score based on food waste relative to the other respondents they can put a cross in the appropriate box at the end of the questionnaire.

After measuring the variables the reliability and descriptive statistics are tested. These results are shown in table 3. It is important to note that, due to the availability of the data, the amount of wasted food, the take- away and dining in the restaurant is translated to a three- week basis. Similar to the research of Koivupuro et al. (2010) and Lyndhurst et al. (2007) the frequency of shopping, main shopping activities and extra shopping activities are measured in the food waste diary. In addition, the respondent is asked in the questionnaire to fill in the frequency of shopping on a weekly basis. This question is measured on a five point Likert scale from ‘Less than once a week’ to ‘four times a week or more’.

- 23 - Table 1

Overview questionnaire food waste.

Question Variables Examples Sources

1 – 9 Demographics Age, gender, education, occupation, household size, income, number of pets and time spent in store

Bos- Browers et al. (2013);

Lyndhurst et al. (2007);

Silvennoinen et al. (2014);

Bell et al. (2011).

10 – 11 Frequency of shopping and responsibility for shopping

How often are you on average going to the store per week?

Bell et al. (2011);

Koivupuro et al. (2012).

12- 14 Planning in Advance, Use of shopping list, Discipline using a shopping list

How often are u using a shopping list?

Koivupuro et al. (2012);

Lyndhurst et al. (2007).

15 – 19, 24

Attitude and propensity towards convenience products Involvement with food and cooking

I choose easy, quick-to- prepare food for weekday evening meals.

Buckley et al. (2008);

Scholderer & Grunert (2005).

20 Impulse buying

tendency

I often buy things spontaneously

Rook & Fisher (1995)

21 – 22 Sales promotion effectiveness and promotions

With this type of promotion I really save money

Chandon et al. (2000)

23, 25 Main reason for food waste and buying too much

What is your main reason for wasting food?

Bos- Browers et al. (2013);

Lyndhurst et al. (2007);

Silvennoinen et al. (2014)

26 Motivation to

maintain healthy diet.

I feel that I want to take responsibility for my own health.

Treatment Self-Regulation Questionnaire (TSRQ) e.g., Ryan & Connell (1989) ; Levesque et al. (2007) 27 Perceived assortment

variety

This assortment offers a large number of items to choose from

Broniarczyk, Hoyer and McAlister (1998)

- 24 -

Regarding the variable use of a shopping list three questions are asked based on the research of Lyndhurst et al. (2007). The respondents are asked if shopping activities are planned (measured with a four point Likert scale from ‘I am getting inspired at the store’ to ‘I am planning for more than two days ahead’) how often they are using a shopping list (measured with a five point Likert scale from ‘never’ to ‘always’) and how high the respondents discipline is to stick to the shopping list (measured with a five point Likert scale from ‘very strictly’ to ‘not strictly at all’). The outcomes regarding the mean and standard deviation of these questions are summarized in table 2.

Table 2

Measures & Reliability

Variable Cronbach’s alpha Mean SD

Frequency of shopping (Bell et al., 2011; Koivupuro et al., 2012)

Frequency of shopping n.a. 3.43 1.10

Shopping list (Koivupuro et al., 2012; Lyndhurst et al., 2007)

Planning in advance n.a. 2.57 .90

Use of shopping list n.a. 3.48 1.35

Discipline using shopping list n.a. 2.79 .88

Impulse buying tendency (Rook & Fisher, 1995)

Impulse buying tendency .841 28.48 9.92

Health motivation (TRSQ) (Ryan & Connell, 1989; Levesque et al., 2007)

Autonomous motivation .743 33.13 5.69

Controlled motivation .769 7.90 4.54

Amotivation .644 5.95 3.43

After these questions the validated scale developed by Rook & Fisher (1995) measures a respondent’s impulse buying tendency. This scale is based on a seven point Likert scale from 1 ‘Totally disagree’ over 4 ‘Neither agree, neither disagree’ to 7 ‘Totally agree’. This scale measures a respondent’s impulse buying tendency with nine statements. In the research of Rook & Fisher (1995) a confirmatory factor analysis on the nine- item scale of impulse

- 25 -

buying tendency is conducted. The results show an acceptable model in which all coefficients are large and significant (P < .001). The scale’s mean is 25.1 with a standard deviation of 7.4 and Cronbach’s alpha α = .88. In this study a factor analysis and reliability analysis is conducted to find whether the variables in this study are internally consistent and can be combined into one single variable. The items are internally consistent when Cronbach’s Alpha is at least 0.6. In the nine- item scale of Rook & Fisher one item, which is item eight, is reversed. Hence, this item is recoded in order to correspond with the remaining eight items.

The scale’s mean in this study is 28.48 with a standard deviation of 9.92 and Cronbach’s alpha of α = .841. The internal consistency cannot be improved after deleting an item. Hence, all nine items are internally consistent and can be combined into one single item which is called impulse buying tendency.

According to the health motivation of the respondent the questionnaire of the Health- Care- Self- Determination Theory is used (Ryan & Connell, 1989; Levesque et al., 2007).

This scale is also measured with a seven point Likert scale from 1 ‘Totally disagree’ over 4

‘Neither agree, neither disagree’ to 7 ‘Totally agree’. This scale measures a respondent’s health motivation with fifteen statements and differentiates between three motivational states which are: autonomous motivation, controlled motivation and amotivation. In the research of Levesque et al. (2007) the validity of the TSRQ is supported because of the fact that the internal consistency for each sub item was acceptable with most α > 0.644. Before testing the internal consistency a factor analysis is conducted to test if the items can be combined into one single factor because of common variance. The factor analysis shows a KMO measure of sampling adequacy of .769 meaning that the items are likely to factor well. In addition, Bartlett’s test for sphericity is significant (p < 0.05) meaning that the factor analysis is an appropriate method and can be continued. Based on the eigenvalues, three factors are identified. In addition, after running the factor analysis based on three factors the outcome shows that dividing the items into three new factors is appropriate (Appendix D).

The scale’s mean in this study for the items that measure ‘autonomous motivation’ is 33.13 with a standard deviation of 5.69 and Cronbach’s alpha of α = .743. The internal consistency cannot be improved after deleting an item. Hence, all six items are internally consistent and can be combined into one single item which is called autonomous motivation.

Regarding the items that measure ‘controlled motivation’ the mean is 17.69 with a standard deviation of 5.48 and Cronbach’s alpha of α = .625. In contrast to the items of autonomous motivation, Cronbach’s alpha can be improved by deleting two items. Hence, statement 2

- 26 -

‘Because I would feel guilty or ashamed of myself if I did not eat a healthy diet’ and 7

‘Because I would feel bad about myself if I did not eat a healthy diet’ are deleted. The remaining four statements are internally consistent and can be combined into one single item which is called controlled motivation. The new mean for this scale is 7.90 with a standard deviation of 4.54 and Cronbach’s alpha of α = .769. Finally, the reliability analysis of amotivation is performed which shows a mean of 5.95 with a standard deviation of 3.43 and a Cronbach’s alpha of α = .644. The internal consistency cannot be improved after deleting an item. Hence, all three items are internally consistent and can be combined into one single item which is called amotivation. A summary of the outcomes is shown in table 2.

The items that measure impulse buying tendency, autonomous motivation and controlled motivation are showing a high reliability. However, the items that measure amotivation demonstrate a somewhat lower reliability than expected. An explanation might be the small size of respondents. Because of the fact that both scales are reliable in this study and other studies this study will maintain using them.

3.2 D

S

The total number of households that participated in this study is 87 which are 148 adults and 52 children. The persons in the household that is mainly responsible for shopping groceries and therefore filled in the questionnaire are mostly females with 83.9%. Regarding the age of the respondents 31% was younger as 30 years, 6.9% was between 31 and 40 and 12.7% was between 41 and 50. The largest group of the respondents was between 51 and 60 with 37.9%.

The remaining 11.5% was 61 or older. The average age of the respondents was close to 45 years.

Regarding the education of the respondents, the largest group of respondents (39.1%) stated HBO as the highest completed education, followed by MBO (25.3%) and a university degree (18.4%). In addition, 62.1% of the respondents were employed whereas 10.3% were students and 9.2% are self- employed. According to the amount of working hours nearly 50.6% of the respondents works less than 25 hours per week. In addition, 16.9% of the respondent’s works between 20 and 30 hours a week and the remaining 32.5% works more than 30 hours a week. The average is nearly 22.5 working hours per week. In total, 60.9% of the households included in this research did not have any children living in their households while 20.7% had one child. In addition, 16.1% had two children and 2.3% had three children.

The average age of the children is nearly 15 years. Consistently, the average household size in this study is 2.3 persons per household. The main category of monthly income per household

- 27 -

is €3000 - €3999 with 21.8%. Moreover, 8% had a monthly income lower than €999, 13.8%

between €1000- €1499 and 8.0% between €1500- €1999. In addition, 16.1% between €2000-

€2999 and 16.1% has a monthly income of €4000 or more. The remaining 16.1% of the respondents prefer not to say the amount of monthly income. Focusing on the number of pets 57.5% does not have any pets and 42.5% of the respondents do have one pet or more.

Regarding the time spent in the supermarket 46% of the respondents are shopping 16-30 minutes for groceries. In addition, almost 23% of the respondents are spending 31- 45 minutes, 17.2% are spending 46-60 minutes and 13.8% are spending 0-15 minutes in the supermarket when shopping for groceries. Focusing on dining in a restaurant or ordering meals the majority (72.4%) eat in a restaurant or order food once or twice a month. The second largest group (18.8%) is eating in a restaurant or ordering food three or four times a month followed by 18.4% that with less than once a month. Finally, 9.2% indicated to eat in a restaurant or order meals more than four times per month.

Table 3

Overview categories of disposed food Category of

waste

Waste in units

Mean Min Max SD Percentage

of total

Total liquids 92157.0 1059.3 0 7225.0 1391.9 38.0%

Total groceries 14910.0 1713.8 0 8705.0 1781.0 62.0%

Total waste 238998.0 2747.1 0 13810.0 13810.0 100%

Table 3 gives an overview of the disposed products in three different categories. The outcome indicates the total amount of wasted food and divides the results in milliliter and gram. The total amount of avoidable food waste in this study is 238998 units with a mean of 2747.1 units. The total food waste in units for the category liquids (i.e. products measured in milliliter) is 92157 units with a mean of 1059.3 units. The total amount of food waste in the category groceries (i.e. products measured in gram) is 149101 units with a mean of 1713.8 units. This means that liquids are responsible for 38% of the avoidable food waste and groceries for 62%.

- 28 -

3.3 M

In this research the dependent variable is defined as the amount of wasted food regardless the category of the products (i.e. milliliter and gram). Since this research wants to investigate the conceptual model with variables that are measured using an interval scale, a multiple regression analysis is performed to test the hypothesis and overall model. In this study the following formula is used:

y = a + b₁x₁ + b₂x₂ + b₃x₃ + ... + b_zx_z

However, before conducting the multiple regression analysis, the independent variables are tested on multicollinearity.

Multicollinearity exist when there are high correlations between the variables and can be observed via the VIF- score (Variance Inflation Factor). If the VIF- score is higher than 5 and the tolerance lower than .20 the variables are highly correlated which means that the there is a high degree of correlation between explanatory variables (Sinan & Alkan, 2015). Therefore, the basic assumption of multicollinearity is that the better an independent variable can be explained by the other independent variables (i.e. the moderators), the higher the multicollinearity, the higher the R² and the higher the VIF- score. If there is a high VIF- score this may lead to excluding the variables from further analysis because it may cause inaccurate analysis about the effect of the independent variables. Table 4 shows that all the tolerance (T) and VIF scores are higher than .20 and 5 meaning that multicollinearity is not an issue.

Hence, no variables are excluded for further analysis in this study.

4. R ESULTS

In this section the hypotheses formulated in the theoretical framework are tested within SPSS 20. The main effect of impulse buying tendency on food waste, as well as moderating effects of the three moderators which are frequency of shopping, health motivation and use of a shopping list are tested by using a multiple regression analysis.

4.2 M

E

I

B

T

The first hypothesis in this research is defined as ‘the higher the impulse buying tendency of a person the higher the food waste’. In the multiple regression analysis the influence of impulse buying tendency on food waste is tested including interaction effect of the moderating variables in the conceptual model. The results of the multiple regression analysis are depicted