An investigation into the role of socio-economic and educational factors on customer purchasing behaviours in the health-food market : a cross-cultural analysis into the fruit and vegetable industry

(1)

Tom Farquharson

10965742

26-06-2018

Economics and Business

Business Administration

Drs.Ing.A.C.Y Meulmans

An Investigation into the Role of Socio-Economic and

Educational Factors on Customer Purchasing

Behaviours in the Health-Food Market : A Cross-Cultural

Analysis into the Fruit and Vegetable Industry

Abstract

This paper aims to investigate the extent to which socio-economic and educational factors influence consumer willingness to purchase fruit and vegetables in the health-food industry, and how gender moderates this relationship. A survey data collection analysis of 99 surveys was distributed via the Internet and the results subsequently quantified. The results suggest that socio-economic and education status did not affect female willingness to purchase fruit and vegetables. Further still, men of a higher socio-economic status and education were also not found to have a higher willingness to purchase fruit and vegetables, compared to those of a lower status. This paper provides managers and marketers with evidence that gender plays a significant role in consumer behaviours and their willingness to purchase.

(2)

Statement of Originality

This document is written by Student Tom Farquharson who

declares to take full responsibility for the contents of this

document.

I declare that the text and the work presented in this

document are original and that no sources other than those

mentioned in the text and its references have been used in

creating it.

The Faculty of Economics and Business is responsible

solely for the supervision of completion of the work, not for

the contents.

(3)

One of the most influential mega trends set to dominate 2018 is the concept of healthy eating, particularly organic, raw and non-sugary foods (Yu, 2018). According to Statista, the value of the health food industry is set to hit 1.1 trillion dollars in 2019 (2018) and this provides firms with challenges to maximise and understand consumer

purchasing behaviours in this industry. One industry which is predicted to grow

substantially is the fruit and vegetable industry which generated 269 billion US dollars in 2015, and expects to generate 317 billion dollars in 2021 (Statista, 2018). As consumers become more health conscious, their dietary choices begin to reflect this trend as they become more educated about the added health benefits of healthy food. It has been discovered that consuming 7 portions of fruit and vegetables a day can lower risk of death from any cause by 42% (Sedghi, 2014). Along with the considerable health benefits that influence consumer purchases of health foods, the role of gender is also an important factor. According to Sedghi, gender plays a significant role in the varied consumptions of fruit and vegetables with noticeable fluctuations in consumption over the last decade (2014). A study conducted by the Private Label Manufacturers Association suggested that women still dominate the food retail industry and despite their career advancements, do not spend less time shopping for food products (2013). However, other studies have actually found that in recent years, men are beginning to enter the food industry more consistently and regularly. According to Wells, in recent years, eighty four percent of men consider themselves the primary grocery shopper in the household (2017). This is due to changing gender roles in society as more men are expected to contribute to household activities such as cooking and shopping (Stevens, 2017). Thus, men are beginning to take a more active role in the foods they consume. According to Stevens, another reason for a higher percentage of men entering the food industry is men are waiting longer to settle down with a partner (2017). Whilst there are a higher percentage of men in the food

industry, the actual consumer behaviours of men and women differ considerably. Men are more likely to make impulse decisions and choose products which catch their immediate attention whereas women plan ahead and think more long term when it comes to food shopping (Bhattarai, 2017). However, these findings are not consistently backed up, which highlight the complexity of consumer buying behaviour and the role of gender in these purchasing decisions. In other studies, women have in fact been found to be

(5)

2003). Therefore, there is a clear conflict of evidence when it comes to the role of gender differences in relation to customer buying behaviour decisions. These differences can be put down to physiological gender differences but also cultural values in a society.

Numerous studies have shown that education is a pivotal factor in fruit and vegetable intake (Stornes, Bere, Eikemo and Stea, 2017) as well as socio-economic factors (Van Lenthe, Jansen and Kamphuis, 2015). This paper aims to investigate the role

demographic factors play in its relationship with consumer’s willingness to purchase fruit and vegetables. Thus, the following research question will be answered: To what extent does the role of socio-economic and educational factors affect consumers’ willingness to purchase fruit and vegetables in the health food industry? This paper will follow the

outlined structure; firstly, a literature review will be conducted to give a theoretical context to the study; secondly, a theoretical framework will be outlined with this paper’s

theoretical model and hypotheses this study will be analysing; thirdly, a methodology will be included which will highlight the overall research design; fourthly, the results will be drawn up and analyses; penultimately, the discussion section will be written to discuss the findings and discuss the limitations of the study and lastly, the conclusion will summarise the findings and offer ideas for future research of the topic.

(6)

2.Literature Review

2.1.Role of Gender

The current literature points to the different roles men and women play when it comes to purchasing behaviour in the health food industry. Men are generally less enthusiastic about the benefits of healthy eating (Wardle, Haase, Steptoe, Nillapun,

Jonwutiwes, & Bellisie, 2004). Paul and Rana discovered that women, on the other hand, considered organic and healthy foods more important than men (2012). This might help to explain why women were found to be more health conscious and avoid foods which might negatively affect their appearance (Chambers, Lobb, Butler & Traill, 2008). Women were also strongly influenced by social influences when it came to purchasing intentions whereas men were less influenced by social pressures (Pascual-Miguel, Agudo-Peregrina & Chaparro-Pelaez, 2015). There is also a distinct difference in terms of eating restraint and self-control when it comes to food consumption between men and women. A recent study discovered that when the use of ‘priming’ was utilised- cueing key words to

respondents before they ate which resonates with them such as ‘healthy’- women showed stronger signs of resonation and as a result food restraint, whereas men did not (Minas, Poor, Dennis & Bartelt, 2016). In other words, women are more tuned into their emotional drivers for consumption whereas men seem to exhibit less self-knowledge, and this could lead to assumptions about gender differences in healthy eating, such as fruit and vegetables. Little, Ilbery and Watts refer to this as ‘gender identity’ and suggest that the reason women are so different to men in terms of food consumption is that they have been socially conditioned to pursue ideologies of ‘weight and womanhood’ (2009). These can be classed as emotional triggers rooted from societal pressures that influence a consumer’s willingness to purchase. According to Nieves-Rodriguez, Perez-Rivera, Longobardi and Davis-Pellot, specifically women’s purchasing decisions are used to build their self-esteem (2017). However, societal pressures are not just aimed at women. The increasingly powerful and imposing role of the media is a reason for greater consumer emphasis on self-image and therefore healthy food consumption for men and women. According to Islam, Wei, Sheikh, Hameed and Azam the media present this idyllic image that people wish to attain and this especially influences younger people, including men (2017).

(7)

Education, socio-economic factors and gender roles can also influence consumer buying behaviours. Brennan and Singh suggest that social status is linked to a wide array of health-related factors such as fruit and vegetable consumption (2010). According to Van Lenthe, Jansen and Kamphuis, “being higher in the pyramid of needs” was

connected to a more healthy diet (2015). The concept of image-congruence theory suggests that consumers purchase products which are aligned with their self-concept (Jones, Reilly, Cox & Cole, 2017) and this can relate back to one’s socio-economic status as by purchasing healthier, higher quality food, one is self-reinforcing their own image to society. Paul and Rana suggest that income is one of the most important factors when it comes to purchasing organic and healthy food (2012) and therefore emphasise the important demographic influences that surround health food consumption. Jones et al., on the other hand present the social identity theory which implies gender is the major influence of self-concept and therefore purchasing decisions (2017) as opposed to socio-economic factors. However, this is highly contested by researchers who believe

willingness to purchase is linked largely to money availability and therefore socio-economic factors (Badgaiyan and Verma, 2015).

Along with socio-economic factors, education also has an influence on consumer willingness to purchase. Paul and Rana discovered that higher educated people were more likely to purchase organic health food compared to those with less education (2012). This was reinforced by Lallukka, Laaksonen, Rahkonen, Roos and Lahelma who concluded that education has a positive relationship to healthy eating (2007). This emphasis on education can also be translated to not only the concept of healthy eating, but specifically consume intake of fruit and vegetables. Wardle, Parmenter and Waller discovered a strong link between nutritional education and intake of fruit and vegetables (2000).

(8)

3.Theoretical Framework

Men and women have different consumer buying behaviours especially when it comes to their willingness to purchase and also their brand loyalty (Radojka and Filipovic, 2017). The theoretical model presented in this paper builds on the Howard Sheth Model

proposed by Farley and Ring (1970) and examines these differences. As observed in Figure 1, the model analyses consumers’ motivations, attitudes and intentions when it comes to their willingness to purchase fruit and vegetable products. Essentially, these three concepts; motivation, attitude and intention; will subsequently give an accurate insight into the psychological thought-processes consumers’ go through when it comes to their willingness to purchase fruit and vegetables. Paul and Rana discovered that health was the biggest influence on consumer willingness to purchase heathy foods (2012) whilst Dean, Raats and Shepherd point to the role of ‘self-identity’ and more

FIGURE 1, THE HOWARD SHETH MODEL AS PROPOSED BY FARLEY AND RING (1970) DESCRIBES THE PSYCHOLOGICAL FACTORS BEHIND CONSUMERS’ PURCHASING DECISIONS.

(9)

analysis, it will equally consider the key psychological drivers behind consumers’ willingness to purchase fruit and vegetables.

Women’s attitudes towards healthy eating are linked to different sources than men, for example women are largely influenced by social factors and pressures

(Pascual-Miguel, Agudo-Peregrina & Chaparro-Pelaez, 2015). According to Pascual-Miguel et al, women as a result, are more concerned with eating foods low on salt and avoiding fatty foods (2015). Furthermore, Nieves-Rodriguez et al discovered that women decided on the type of foods they consumed largely due to the effect it will have on their self-image (2017). Thus, the overall effects healthy eating has on women’s bodies, health and

appearance seem to be the main drivers of female fruit and vegetables consumption. This pattern seems to be driven by social and media-related factors that have an impact on female willingness to purchase. Therefore, this paper will analyse the extent to which women are willing to purchase fruit and vegetables, and consequently leads to the first hypothesis:

H1: Women have a higher willingness to purchase fruit and vegetable products compared to men.

In order to analyse the socio-economic factors that will influence the relationship between gender and willingness to purchase in this paper, ‘Maslow’s Pyramid of Needs’ will be extracted from the work of Van Lenthe, Jansen and Kamphuis (2015) and

subsequently measured against, to make assumptions from the data results. According to Van Lenthe, Jansen and Kamphuis, the position a consumer is situated in the pyramid directly influences their consumption of health foods and the suggest the higher up in the pyramid one is, the healthier their food choices are regardless of gender (2015). Maslow’s Hierarchy of Needs, however, is not without criticism; the pyramid does not work

systemically as an ascension as some needs are satisfied before other needs below them. The model is simply a guideline into how human behaviour functions and highlights the human’s psychological needs function, using a clear diagram relating to assumptions about human behaviour. The easy understanding of such a complex psychological

phenomena makes Maslow’s pyramid an ideal model to use in this paper.

The assumptions mentioned in Maslow’s model are backed up by other researchers such as Brennan and Singh who also believed that there was a direct link between social-economic levels and fruit and vegetable intake (2011). Other researchers point to

(10)

economic logic such as Maguire and Monsivais who suggest that as food prices increase, only those in higher socio-economic circles can afford sufficient fruit and vegetable diets (2015). Nonetheless, fruit and vegetable products are not on the whole expensive,

depending on the type of fruit or vegetable one is consuming, and therefore is not

necessarily a product which requires large amounts of spending power. Despite this, food shopping is always price-sensitive and if fruit or vegetable prices increase, this would affect consumer willingness to purchase. Whilst price is inevitably a factor to fruit and vegetable intake, the role of gender on perception of fruit and vegetable intake is also a influential part of this paper’s analysis. For women in particular, their self-image has been discovered to be a factor when they make food purchasing decisions. According to Vila-Lopez and Kuster-Boluda, women’s diets are linked inextricably to their weight control and overall health values (2016). The internet and social media are bombarding

consumers with ideal physiques and also idyllic eating regimes that it is almost inevitable that consumers will alter their diets and try to reach this perfect self-image. Further still, Jones et al, point to the other factors that may affect women’s food choices such as the concept of social identity theory and suggest that this may drive women to adopt food choices not necessarily rationally economic (2017). Thus, this leads to this paper’s second hypothesis:

H2: The level of socio-economic factors in women does not affect their willingness to purchase fruit and vegetables

Figure 2, Maslow’s Hierarchy of Needs illustrates the different stages a person goes through

(11)

It is suggested that more highly educated people are more likely to reach the upper limits of Maslow’s pyramid of needs and therefore have the ability to control their healthy food choices ( Van Lenthe, Jansen & Kamphuis, 2015). Whilst education can be classed as an indicator of one’s willingness to purchase fruit and vegetables, the type of education is a significant factor that could be investigated. For example, nutritional education is different to academic education and therefore should be considered in the analysis. Nonetheless, it has been suggested there is in fact a direct link between higher education and healthier lifestyle behaviours such as exercise and healthy eating (Maguire and Monsivais, 2015). Furthermore, plant-based diets, such as an ample intake of fruit and vegetables, are more likely to occur in those educated from a collegiate level (Kell, Judd, Pearson, Shikany and Fernandez, 2015). However, the role of gender plays a critical role in navigating this intricate web of consumer behaviours. According to Stornes, Bere, Eikemo and Stea lower educated women consume just as much fruit as highly educated men (2017). Thus, despite certain financial constraints, women are still driven in some cultures to eat more healthily. Women have been found to make purchase decisions from a more emotional perspective (Jamil, Hassan, Farid & Ahmad, 2017) and suggests that women have a greater, inner desire to consumer fruit and vegetables than men, despite any seemingly irrational economic factors. This leads to the third hypothesis of this paper:

H3: The level of education in women does not affect their willingness to purchase fruit and vegetables

According to Wardle, Haase, Steptoe, Nillapun, Jonwutiwes and Bellisie, men have been found to have in general a higher tendency to pursue unhealthy lifestyle behaviours (2004). In Parson’s book, the author refers to men’s ability to reject cultural rules

surrounding healthy eating norms due to their more elevated position in society (2016). This culminates in a general ‘less enthusiastic’ approach to healthy lifestyle and eating choices amongst men (Wardle et al., 2004). However, the role of gender and healthy eating is a relationship that is changing in this modern era. Whilst women are

predominantly more health conscious, men are not completely ignorant of the benefits fruit and vegetable intake can have on their physical fitness and well-being ( Dumbrell and Mathai, 2008). When it comes to purchasing decisions, men are more concerned with the level of the pleasure they will derive from the purchase decision (Jamil et al., 2017) which

(12)

suggests that they might be more concerned with the taste of food, as opposed to the benefits of eating such foods on physical health, for example. Men have also been discovered to be more ‘objective’ and ‘realistic’ when making purchase decisions compared to women (Łatuszyńska, Wawrzyniak, Wąsikowska and Furaji, 2013). Thus, with greater socio-economic levels and a higher education of the products they consume, one could suggest men will consume a higher amount of fruit and vegetables if they possess greater knowledge and economic means. This leads to the paper’s final hypotheses:

H4: Men from a higher educational background have a greater willingness to purchase fruit and vegetables.

H5: Men from a higher socio-economic background have a greater willingness to purchase fruit and vegetables.

Figure 3, this paper’s theoretical model. Socio-economic and educational factors make up part of the independent variable. Willingness to purchase is broken down

Demographic factors Socio-Economic

Education

Willingness to Purchase Motivation-Attitude-Intention Price Price Gender Customer Buying Behaviours Moderator

(13)

4.Methodology 4.1.Research Design

Using a sequential exploratory mixed method research approach this research will involve an initial phase of qualitative data collection and then quantitative analysis

(Saunders, Lewis and Thornhill, 2012, pp.171). Using cross-sectional research, a series of 15 question surveys will be created and qualitative data will be collected from them

accordingly (Appendix 9). The language of the surveys will be analysed to make sure the questions are clear and concise. Furthermore, prior to distribution the questions will be analysed for applicability and validity to ensure the hypotheses will be answered

accurately; the notion of ‘willingness to purchase’ is a slightly intangible concept and therefore the questions of the survey were carefully structured. Instead of asking simply what a respondents’ willingness to purchase was, a scenario question was constructed to allow the respondent to think about their own shopping tendencies. This was to prevent respondents answering without thought and was aimed to provide meaningful data. This data will then be transferred to Microsoft Excel which will be used to analyse the

descriptive statistics. This is particularly important because it will allow the creation of various pie charts and graphs which allows for visual comparisons (Appendix 6 and Appendix 7). After this procedure, the data will then be analysed quantitatively using statistical analysis software: SPSS.

One of the strengths of using surveys is the low cost requirements and high speed of data collection (Saunders, Lewis and Thornhill, 2012, pp.182) which leads to greater

repeatability. They can also be administrated through the internet and via online devices such as iPads or iPhones. Combined with the fact this survey took very little time to complete, it makes for rapid and convenient data collection for respondents.

Furthermore, a wide range of respondents could be targeted in the Netherlands and also in France. However, one of the major weaknesses is potential for response bias and a tendency for respondents to provide dishonest answers. A further disadvantage is that respondents might not feel comfortable answering certain questions which entail sensitive information.

These disadvantages can thus culminate in a lowering of this investigation’s internal validity.

(14)

4.2.Distribution and Sample

The surveys will be distributed using a combination of the Internet (SurveyMonkey) and also face-to-face. The benefit of using the Internet as a method of distribution is the wide potential range of the population that can be accessed and speed of data collection. However, one of the biggest drawbacks of using the internet is the tendency for surveys to be ignored or unanswered completely. Therefore, this paper will also interact in-person with respondents to ensure they fill in the survey completely and correctly. A further factor that was considered was the length of the survey. The questions were concisely

structured so the data collected could be thorough and specific. But perhaps more importantly, the surveys were able to be completed within a couple of minutes which encourages respondents’ response rates.

In order to eliminate any forms of researcher bias, the survey was not distributed on Facebook or to email addresses of other students at the University of Amsterdam (UvA). Although this would provide convenient forms of data, the population sample would be too biased and would subsequently eliminate true effects, such as ‘education,’ if the majority of the sample were bachelor students. This would pose a threat to the external validity of the study. On top of this, to further increase reliability, the size of the sample would be aimed to be at least 100 respondents which was considered an acceptable requirement to ensure reliability of results and improve external validity.

The response rate was an important factor in this research design that would ensure relevant data collection. According to Fryrear, an ideal response rate is around 80% from a small, random sample (2015) and with this paper sending personal email invitations to a large proportion of the emails sent, a high response rate was expected. Nevertheless, a number of actions were taken to encourage a high response rate. Firstly, the design of the survey was easy to complete and quick to fill in. Secondly, the value of the study was communicated to the respondents explaining their response was to be part of a bachelor thesis and was involved in a study about the health and food industry. Lastly, reminders were distributed to respondents to fill in the survey with the first two actions emphasised.

(15)

4.3.Variables

As illustrated in Figure 3, the study contained two independent variable, one dependent variable and one moderating variable. The independent variables- socio-economic and education- was related to society and its role in consumer behaviour. The dependent variable- willingness to purchase- was categorised into two sections. The first referenced the Howard Sheth Model, Figure 1, which included consumer motivation, attitude and intention behind purchasing fruit and vegetables.The second dependent variable was price and its relation to a consumer’s willingness to purchase fruit and vegetables. The moderating variable was gender and was categorised into men and women, despite the fact there are of course several other forms of gender. This decision was made to specifically target these two forms of gender in this study, and therefore will produce conclusions that cannot be generalised to all forms of ‘gender.’

4.4.Internal Validity

Internal validity was an important consideration in this study because there were two independent variables: socio-economic and education. Thus, there was a requirement to avoid confounding variables and to prevent observations in the study being manipulated by a factor other than the independent variable. The time at which the surveys were distributed were aimed to be the same so as improve internal validity. Furthermore, to ensure all variables in the model were operationalised, and to improve internal validity, a pilot study was conducted on a small scale to evaluate feasibility and assess the survey’s face validity. According to Saunders, Lewis and Thornhill, a pilot test is designed to minimise any problems whilst recording data (2012, pp.473). Further still, a minimum of number for a sufficient pilot test is 10 surveys (Saunders, Lewis and Thornhill, 2012,pp.473), and therefore this was conducted accordingly. As a result, several of the questions were either adapted or removed due to their lack of conciseness or value. This pilot test was conducted to improve the accuracy of the data being collected but also to maximise time efficiency when it came to filling out the survey for respondents. This was particularly relevant for some of the motivational and price questions related to

willingness to purchase fruit and vegetables. Initially in the pilot test, a few of the questions were too ‘closed’ and did not allow the respondent to fully interact with the questions. Thus, a series of scenario questions were created. Once finalised, the survey remained the same for the entire data collection phase so as to rule out any threat to internal validity such as instrumentation.

(16)

4.5.External Validity

External validity is how well data and theories from one setting apply to another

setting (Saunders, Lewis & Thornhill, 2012,pp.204) and is essentially the generalisability of the study to the wider population. The external validity of a study can be broken down into two sub-sections: population validity and ecological validity (Shuttleworth, 2018). Population validity is the extent to which the sample in the study is applicable to the wider population. In this particular study, population validity was ensured by trying to avoid a sample of students who filled out the surveys because this was not reflective of the educational background of the wider population. With education being one of the independent variables in this study’s model, this was important not only to external validity but also to internal validity. Ecological validity is the generalisability of results to real-life settings (Shuttleworth, 2018) Thus, it measures whether the study can be repeated in different settings and in order to ensure ecological validity, the survey’s questions were phrased in simple, concise English so as to allow this study to be repeated in any English speaking countries.

4.6.Limitations

There are a number of impact limitations due to time constraints, the sample size is relatively small and could arguably be a factor which might limit generalisability. As

mentioned above, due to a small sample size, the field of gender was narrowed down to just male and female which could arguably limit marginal generalisability. A further limitation to the study is the potential for self-reporting error such as participant bias which is ‘any factor that induces a false response’ (Saunders, Lewis and Thornhill, 2012). For example, respondents might know the importance of healthy eating and therefore report what they believe their eating habits should be instead of their actual eating habits.

There are also a number of statistical data limitations this study encountered. Although the data collected was relatively proportionate in terms of males to females, the age range was focused quite narrowly on an older generation of adults which may

indicate certain limitations of generalisability. Nonetheless, the data collected was of good quality which added extra credibility to the results.

(17)

4.7.Tests and Measurements

In order to test the difference between two means, such as those related to socio-economic and education variables, an analysis of variance (ANOVA) was carried out. This particular test is used to test hypotheses and whether there is a significant difference in variance or not. The ANOVA test produces an F-test statistic which then leads to a p-value. A significant p-value (p<0.05) suggests that at least one variance differs from other groups, and therefore rejects the null hypothesis.

The first hypothesis was analysed using a T-test to compare sample populations and to then discover if there is a significant difference between their means. The T-test produces a relevant t-value which subsequently leads to a p-value. From this p-value, it signifies whether the t-value falls into a significant range and whether it is significant (p<0.05).

In order to conduct the T-Test and transfer the data from SurveyMonkey to SPSS, the variables needed to be coded before analysis was possible. For gender, the values were coded as either 1 or 0, which allowed for a T-Test to be conducted.

For the other questions, the values were coded from 0-4 which allowed SPSS to then use an ANOVA one way test. The questions of the survey were all closed questions which allowed for conclusive and specific data results, and also allowed for more data to be efficiently quantifiable. Open ended questions were considered but when time was considered and also the speed at which data collection and analysis was required, the exploratory benefits of including such questions did not outweigh the advantages of using closed questions.

(18)

5.Results

5.1.Descriptive Statistics: Demographic Analysis

The survey was distributed using a combination of face-to-face interviews and over the Internet. The survey was emailed to 75 respondents who could then follow the link provided and fill in the survey online. The response rate was fairly high at 65 (87%) while the remaining surveys were completed using face-to-face data collection analysis in various locations across the Netherlands and in France. According to the SurveyMonkey analysis, on average the survey took 3 minutes to complete and had an 89% completion rate. In total, 99 surveys were collected which consisted of 48 (48.5%) females and 51 males (51.5%). This was one survey short of the expected response rate but was

considered sufficient to provide acceptable data to processed. The highest percentage of the sample age range was between 25-35 (29.3%) and 35-50 (33.3%), with the lowest percentage age group being 70+ (2.02%). In terms of socio-economic analysis, the sample’s working hours centred around 30-40 (32.3%) and 40+ (39.4%), whilst 57.9% owned their home and 26.3% were renting.

5.2.Gender and Socio-Economic and Education Status

There were apparent differences in the socio-economic statuses between men and women in the study. One of the socio-economic factors considered in the survey was housing status and with this criteria, the male sample appeared to be of a higher socio-economic status; 45.8% of women owned their own home compared to 68.6% of men. In ProsperityScorecard, research showed that women owned 5% more homes than men (2018). According to Martin, twice as many single women are buying homes than single men in the modern era (2018). Thus, this study shows somewhat conflicting data results in the home ownership of gender.

Education status was measured from an academic qualification acquisition and men appeared to show higher levels of education in the population sample; 74.5% of men obtained a university or higher university qualification compared to 58.3% of women. Thus, in total, the data seems to illustrate a higher socio-economic status for men

(19)

5.3.Gender and Willingness to Purchase: Price and Motivation

According to the data results from Q14, females were willing to pay a higher price to buy fruit. The data shows 91.7% of women were willing to pay equal to, and above, the typical price of a fruit salad whereas only 76.5% of men were willing to do the same. When combined with the fact men’s socio-economic status was relatively higher than women’s, it suggests that women had a higher willingness to purchase fruit and vegetables than male respondents when it comes to price sensitivity.

From the data in Q15, a similar trend can be observed from the results when it comes to willingness to purchase a green salad; 83% of females were willing to pay equal to, and above, the standard price of a green salad compared to 67% of men. Thus,

clearly women are willing to pay a higher price for both fruit and vegetables. Whilst this data is driven purely by price in these questions, they indicate that women’s willingness to pay.

The intrinsic motivations behind eating fruit and vegetables were also compared between genders and there appeared to be differences when it came to the desire for the health benefits of eating fruit and vegetables. 57% of men were interested in receiving the health benefits of eating fruit and vegetables compared to 50% of women. This pattern was reversed when it came to the appearance and well-being benefits derived from fruit

(20)

and vegetable consumption. 35% of women wished to eat fruit and vegetables to improve their appearance compared to 19% of men.

5.4.Analytical Statistics

Internal Consistency

In this section, the internal consistency of the data was conducted under a

reliability analysis framework on SPSS. In order to measure and indicate the purchase of fruit and vegetables, three different items were part of a reliability analysis and labeled: ‘fruitS.’ The results (Appendix 19) showed a relatively good internal consistency according to the Cronbach’s Alpha (α=0.665) for the following three items of the survey: daily intake of fruit and vegetables, percentage of fruit and vegetables in shopping cart and

willingness to purchase a fruit salad. This was organised into a sum score from 0-9 where a combination of scores are calculated; therefore the higher the sum, the higher the willingness to purchase fruit.

(21)

Gender Differences in Willingness to Purchase Fruit and Vegetables

In order to conduct an independent T-Test on fruitS and gender, a number of assumptions must be assumed (Maverick, 2018);; firstly, the scale of measurement is continuous and uses ordinal data;; secondly, the data was collected from a random sample from the total population;; thirdly when plotted, the graph is shaped in a bell-shape, normal distribution;; fourthly, a reasonably large sample is used so as to produce a normal

distribution;; fifthly, there is homogeneity of variance which implies samples have equal standard deviations. From the SPSS output, in particular the skewness and kurtosis were analysed and found to be within the boundaries of -1.96<1.96 after being divided by their standard error and therefore can be assumed as normally distributed (Rose, Spinks & Canhoto, 2015). Further still, from a visual perspective, one can observe from the stem and leaf diagram (Appendix 3), a distinct bell-shaped curve that denotes a normal distribution. To answer the first hypothesis, a comparison of means between fruitS and gender was tested using an independent samples T-Test (Appendix 20). Assuming equal variances, the results from the test, showed there was a significant statistical difference between the two means (t(97, 3.54), p=0.001). This therefore shows that females had a higher mean intake of fruit and vegetables than their male counterparts

(Female=6.04:Male= 4.72). Therefore, H1 can be accepted as women appear to have a higher willingness to purchase fruit and vegetables than men.

Figure 1 illustrates a sum score of fruitS. The figure shows the

highest scores display the highest willingness to purchase fruitS.

(22)

Female Socio-Economic Factors on Willingness to Purchase Fruit and Vegetables The variable ‘housing’ was chosen as a suitable measurement of socio-economic factors and was separated into 4 distinguished categories. However, when it came to the analysis of the variable, it was recoded into a relevant order and the singular

uncategorised variable ‘other’ was removed as an outlier. Using these three groups, a one- way ANOVA test was conducted to test the differences of the means of these different groups (Appendix 22). The result showed a non-significant difference (p=0.552) between women’s housing and their willingness to purchase fruit and vegetables. Thus, H2 was subsequently accepted. The variable education was also tested under a one-way ANOVA (Appendix 23) test and produced a non-significant p value (p=0.372). Thus, H3 can be accepted and education does not have a significant effect on female willingness to purchase fruit and vegetables.

Male Socio-Economic Factors on Willingness to Purchase Fruit and Vegetables Similar to the female education variable, the male education variable was

separated into three categories: high school, university education and higher university education. A one-way ANOVA test produced a non-significant p value (p=0.493) which subsequently rejects H4 and denotes that men with higher educational status do not have a higher willingness to purchase fruit and vegetables compared to lower educated men (Appendix 25). The male housing variable as an indicator of socio-economic factors was also tested with a one-way ANOVA test and came back with a non-significant p value (p=0.169) and therefore also rejects H5 (Appendix 24). Therefore, we found that there was not a significant difference in willingness to purchase fruit and vegetables in men from a higher socio-economic and educational background.

Moderation Effect: Regression Analysis

To predict willingness to purchase fruit and vegetables through education factors, whilst testing gender as a moderation effect, a regression analysis was performed (Appendix 26). The beta coefficient was focused on as the degree of change in the outcome variable. The negative coefficient beta (b=-1.41) for gender had a significant p value (p=0.0004) and illustrates that males correlate to having a lower willingness to purchase fruit and vegetables than women. However, using housing in the regression analysis instead of education, there is a similarly significant trend for moderation. The

(23)

(highlighted in red) who had higher housing standards, the willingness to purchase fruit and vegetables increased whilst for women (highlighted blue) there was a slight decrease.

6.Discussion

An interpretation of the results will be discussed in this section in the context of this paper’s theoretical framework, in an attempt to answer this paper’s research question. It will begin by discussing specific results and offering contributions to current literature, and then move toward more general explanations and possible applicable frameworks.

Following on from this, the limitations of this research will be analysed and then subsequently followed by suggestions for future research.

6.1.Gender’s Role in Willingness to Purchase Fruit and Vegetables

The extent to which the role of socio-economic and education factors affects consumers’ willingness to purchase fruit and vegetables was moderated by gender in this model. The results show there is a significant difference between the willingness to

purchase fruit and vegetables between males and females. Women on the whole have a higher willingness to purchase fruit and vegetables than men. The reasons for this can be listed accordingly;; firstly, women seemed to be more concerned with their appearance and weight control (Vila-Lopez and Kuster-Boluda, 2016) and therefore, would be expected to have a higher willingness to purchase fruit and vegetables;; secondly, women were

motivated to purchase fruit and vegetables due to the positive effects it had on their self- image due to the effect of the media on female self-esteem (Pascual-Miguel, Agudo- Peregrina & Chaparro-Pelaez, 2015);; thirdly, because of this deep emotional impact appearance has on female purchasing decisions, females will base their decisions from this emotional perspective (Jamil, Hassan, Farid & Ahmad, 2017) and therefore have a stronger, more emotionally-driven willingness to purchase fruit and vegetables. This emotional aspect of the purchase decision process cannot be underestimated as it is this mental fundamental factor that helps to explain why women have a consistently stronger willingness to purchase fruit and vegetables than men despite socio-economic or

educational factors.

Observing Maslow’s Pyramid of Needs, one can identify the different positions men and women generally take up on this model when it comes to eating fruit and vegetables. This paper’s results seem to suggest men may take up a lower position in the pyramid, such as satisfying physiological needs, which emphasises the ‘health benefits’ of eating fruit and vegetables. On the other hand, the importance of fruit and vegetable consumption

(24)

for women is not only physiological but also affects higher areas of their pyramid of needs such as esteem. According to McLeod, the desire for respect from others and one’s own status is one of the strongest needs in this model (2018) and it’s clear to see men and women have different interpretations of status. Whilst women have been programmed to project an idyllic feminine image to the world through the media (Islam et al., 2017), men have not had such a physical presence to live up to. The results reflect this phenomenon as women identified their motivation behind eating fruit and vegetables as having a

positive influence on their appearance whereas men largely focused on the health benefits they provided. Thus, managers and marketers should consider this gender difference when it comes to marketing fruit and vegetable products. For example, health food products for women should emphasise the positive benefits they have on their skin, hair and body figure. On the other hand, products for men should highlight the positive effects they have on their body such as heart health and brain functioning.

However, this pattern may not be a permanent approach to marketing fruit and vegetables for different genders as their consumption is changing. As Sedghi notes, consumption of fruit and vegetables is fluctuating (2014) and men are beginning to

consume higher amounts. This could arguably be because of men’s changing values and society’s expectations for their self-image. The results of this study show that even though over half of men do indeed focus on the health benefits of fruit and vegetables

consumption, almost a quarter of men also bought fruit and vegetables for the effect it had on their appearance. This is a male trend that is predicted to merge more closely to female values as recent studies have shown that men are just as image-conscious as women (Moss, 2016). Over the next decades, as men become more aware of their appearance and the benefits of healthy eating, marketers and managers must be aware of the changing diet trends of each gender.

A further factor involving gender in the willingness to purchase fruit and vegetables is the fact the data was almost entirely collected in the Netherlands, where it was once ranked 13th in the world when it came to gender equality (Solanki, 2017). Thus, it is important to note that in perhaps less equal societies or societies with different values might produce different conclusions. In the Netherlands, for example, Western values of beauty and status are inherent in the system and therefore, women might be more

(25)

women are fairly equal, independent in society and when it comes to their food shopping habits.

6.2.Socio-Economic Factors and their Affect on Willingness to Purchase Fruit and Vegetables: Motivation

When it comes to willingness to purchase fruit and vegetables, the most common factor predicted to influence this relationship is socio-economic factors and the importance of financial ability to purchase healthy food (Brennan and Singh, 2011). However, the results seem to suggest socio-economic factors should not be emphasised so strongly as regardless of socio-economic status, females had a consistently high willingness to purchase fruit and vegetables. This is an important finding because it links back to the previous discussion involving female identity and the fact that financial status is not such a defining factor when it comes to their decision to purchase fruit and vegetables. Women appear to be driven to buy fruit and vegetables because they value the appearance and health benefits they provide at a fundamental level that goes beyond economic rationality. This does not suggest that females are not price sensitive when it comes to purchasing fruit and vegetables but it does mean that they are not wholly price focused. Males on the other hand, from higher socio-economic backgrounds, were not found to have a higher willingness to purchase fruit and vegetables which was a slightly surprising result. On the one hand, this seems to suggest that males in fact are also not so influenced by the financial constraints many previous researchers have pointed to and are perhaps more closely aligned with female values. On the other hand, this could also mean that men with higher financial standing do not value healthy food consumption as highly as women and therefore, are not willing to purchase such foods, even with higher financial capabilities. Using the Howard Sheth Model, Figure 1, consumer decisions were broken down into motivation, attitude and intention. Whilst this study focused on the motivational side of consumer behaviour, the concepts of attitude and intention are also relevant here. The concept of consumer attitude is an important area to discuss because it refers to the influence of one’s attitude towards fruit and vegetables and to what extent this affects willingness to purchase. On the surface, it might seem that a positive attitude would mean immediate higher willingness to purchase fruit and vegetables but in fact there are also a multitude of attitudes that can be investigated. According to McLeod, there are three main types of attitudes: affective, behavioural and cognitive (2009). Affective attitude refers to one’s feelings and involves emotions. Behavioural attitudes involve the patterns of one’s behaviour. Cognitive attitudes refer to the person’s knowledge and belief system. This

(26)

study looked closely at the behavioural aspect of consumer behaviour and the habits consumers have created when it comes to their consumption of fruit and vegetables. However, further studies could investigate the affective and cognitive attitudes that affect consumer behaviour as these two are linked primarily to one’s beliefs system which strongly affects willingness to purchase.

The concept of taste was a significant difference between males and females in the study. Men were considerably more motivated to eat fruit and vegetables due to the

positive taste they experienced whereas women were more concerned with the outcomes of such dietary consumption. It is important to point out that even though women had a higher willingness to purchase fruit and vegetables in general, this motivation was linked to, on the whole, their desire for an outcome: more positive appearance and general well- being. Men, on the other hand, were also motivated by the outcome of the health benefits of eating fruit and vegetables but were motivated more than women by the taste. This fundamental difference in the different ways males and females approach shopping decisions is of paramount importance to marketers. For example, one strategy marketing specialists could adopt is advertising fruit and vegetables products to men through

highlighting the positive health benefits and taste of the products, and for women emphasising the positive effects of such products on their appearance and well-being.

6.3.Socio-Economic Factors and their Affect on Willingness to Purchase Fruit and Vegetables: Price

On the whole, women were more willing to pay a higher price for a green salad than men were, regardless of socio-economic status. This pattern of behaviour was also

observed when it came to their willingness to purchase a fruit salad. Nonetheless, whilst this seems to point to ultimately a higher women’s willingness to purchase in general, one must be slightly wary of sweeping statements when it comes to price indications.

According to Ning, women spend money gradually over time on longevity products

whereas men spend large amounts of money in a short period of time on more emotionally driven products (2017). This might be a reason why men are not as willing to purchase more mundane nutritional products such as fruit and vegetables. Thus, when it comes to price although women are willing to pay a higher price for fruit and vegetables, this might be due to their spending patterns as opposed to their actual dietary motivations and

(27)

The price women were willing to pay was more dispersed equally for a green salad regardless of socio-economic status compared to men. Men on the whole were willing to pay less in total, but there was a small sub-section of men willing to pay a higher price for a green salad. Thus, indicating that there is perhaps a niche market potential for male consumers who are willing to pay a higher price for vegetables. For a fruit salad, both genders had a similar proportion of those willing to pay a high price but a disproportionate price for a lower price. For a fruit salad, men had a more fluctuating willingness to

purchase with those willing to pay high and low prices, whereas women were largely willing to pay a medium to high price on the whole.

6.4.Educational Factors and their Affect on Willingness to Purchase Fruit and Vegetables Men were considerably better educated than women in the study, with a higher proportion obtaining higher university and university level education. Higher numbers of women obtained a high school education only, whilst a smaller margin of men obtained a high school education as a sole form of education. The results from the study seem to suggest that education, at least academically, does not translate to consumers having a higher willingness to purchase healthier food choices. This might be because academic education is not necessarily the same as nutritional education, and further still, many forms of education do not even teach health and nutritional subjects in their curriculum. Because the data was collected in higher socio-economic demographics, lower educated males did not make up a large part of the study compared to women. Nevertheless, it appears that gender perhaps plays a greater role than education when it comes to healthy food choices. Because of the state funded education system, the relationship between education and income in the Netherlands might be slightly different in other countries where higher levels of education are correlated with household income. Therefore when it comes to willingness to purchase fruit and vegetables, a higher education does not necessarily lead to higher income or greater spending power. Therefore, when it comes to marketing

products for a educational demographic in the Netherlands, there is a large, evenly spread proportion of educated people. According to LongReads, there are more women in the higher education system in the Netherlands than men (2017) and this trend is predicted to be accentuated over the coming decades. Therefore, this pattern of men being more educated than women is a trend that is perhaps going to change over the coming years as women are overtaking men. However, this changing trend will not necessarily change the willingness to purchase fruit and vegetables for the different genders.

(28)

6.5.Society’s Role and their Affect on Willingness to Purchase Fruit and Vegetables Having looked at gender, educational and socio-economic factors, one of the most influential factors affecting willingness to purchase fruit and vegetables is society’s role and its changing expectations on consumers, especially in Western society. Men are starting to become more image-conscious to the point where some researchers, such as Moss, who have found that men and women are equally conscious about their body image in this current era (2016). This changing status, or at least merging of values between men and women, means that more men are starting to investigate different ways for them to improve their appearance and well-being, which in the past was a stereotypically female desire. Men are being bombarded with images of the ideal physique through the internet, movies and applications, such as Instagram, that are projecting healthy dieting and physical fitness programs. As a result, this could increase the proportion of men who are willing to purchase fruit and vegetables, as men are becoming more health concerned regardless of their socio-economic status. This is perhaps one of the most important factors affecting willingness to purchase because as society’s values change, so do people’s values change. If one relates this to Maslow’s Hierarchy of Needs, this can

change men’s positioning on the pyramid as fruit and vegetable consumption moves closer to females’ needs which is more towards self-esteem and body image.

Further still, advancements in technology means that the nutritional values of fruit and vegetables are increasing to the point where the true value of eating healthily is identified. For example, studies are starting to show that certain vegetables have both positive and negative effects on body health and at preventing certain diseases (Zelman, 2018). On the flip side, while technology has identified the benefits of fruit, researchers have also found that fruit can have detrimental affects on your teeth due to acidity

(OralWellness, 2018). Thus, these findings all have an effect on consumers’ willingness to purchase fruit and vegetables and will continue to be a factor that will affect certain trends. Along with advancements in technology, people are also starting to explore different nutritional-based diets such as vegetarianism and vegan diets. This is also a product of society, as more alternative forms of vegetable-based diets are becoming mainstream. There has also been an influx of new jobs emerging in the nutrition industry such as nutritionists and dieticians as vegetarian foods are becoming linked to health benefits and body-health in general, whilst meat-based diets have been progressively linked to health

(29)

6.6.Limitations

There are several limitations of the study which will be addressed in this section. Although a number of techniques were used to minimise limitations, which will be discussed, naturally a series of limitations arose which ultimately affected the data and should be adjusted in future research.

Firstly, it has been discovered by researchers that people perceive their eating habits to be healthier than in actual reality (Tobin, 2015). When it comes to the self- analysis of those who completed the survey, one of the limitations of such a study is that people will overestimate their fruit and vegetable consumption, and therefore provide somewhat distorted figures in which to analyse. This is a very challenging limitation to address because it is linked to human psychology and one’s self-perception. In society, looking after your health is perceived as being responsible and is essentially a positive trait of character and therefore to admit perhaps a ‘less healthy’ dietary habit can be quite difficult for respondents. One way of attempting to eradicate this limitation is to emphasise the anonymity of the survey which might perhaps alleviate any concerns respondents have if their actual eating habits are not classed as ‘healthy.’ A second solution might be to phrase questions using fictional scenarios that depict characters who do not have healthy diets and then asking respondents to reply with their own diet choices, which has the effect of removing any expectations they might have to exhibit an healthy, ideal diet.

Secondly, the sample size was relatively small (n=99) which although gives a reasonable insight into the health trends of respondents, is not large enough to make generalisations and in future research should be enlarged to allow for more accurate conclusions. The sample demographic itself was also slightly distorted. One of the aims of this study was to collect data from ‘non-students’ because this this would avoid any

selection bias and also not have any adverse effects on the education independent

variable. As a result, the surveys were distributed through the Internet and also at various tennis clubs in Amsterdam to the members. However, whilst this avoided a predominantly student demographic, this consequently targeted an older generation. Whilst this is not a limitation in itself, it did mean that a large proportion did own their home but did not work as they were retired. This subsequently affected the socio-economic independent variable measurement as it, at times, provided data sets which this study did not necessarily wish to investigate.

This leads to the third limitation which was the criteria of the data collection especially when it came to socio-economic factors and also education measurements.

(30)

Education in this study was measured mainly through academic education and suggested that someone with a higher university’s degree would be more knowledgable to the

benefits of healthy eating compared to someone with a primary school education. However, academic education does not translate to greater nutritional knowledge as someone might have studied health-food education at high school compared to a PhD student who studied geography their entire career, for example. Therefore, future research could try and expand on the variable education and perhaps categorise education into different sub-sections to make it clearer as to what kind education affects willingness to purchase fruit and vegetables.

Further still, this paper used housing as its main indicator of socio-economic status however this indicator does not necessarily highlights one’s earning level and therefore is not an all encompassing criteria. As a result, those males who were from a higher-socio economic status, as measured in the survey distributed, may have been retired or married to a woman who is responsible for the majority of the shopping in the household.

According to the Private Label Manufacturers Association, women are the primary grocery shopper (2013) and this might affect men’s shopping habits and motivations behind their willingness to purchase fruit and vegetables. This should also be considered by marketers who may want to target households with the knowledge that women in this day and age are still the primary grocery shopper of fruit and vegetables. To combat the conflict of measurements, such as the housing and hours worked scenario mentioned earlier, one of the factors this study didn’t measure was respondents’ income on an annual basis. This was deemed too intrusive but perhaps if the survey emphasised privacy more strongly or perhaps was collected in a different demographic, this important measurement for socio- economic status could be collected and provide more accurate results.

Fourthly, perhaps different independent variables could be measured instead of using education and socio-economic status. On the surface, these two variables seem logical indicators of willingness to purchase fruit and vegetables but there are perhaps other indicators such as childhood diet or cooking ability which certainly would be useful to investigate for future research. According to research done at the University of Calgary, one of the most important factors affecting adult health is their childhood diet (2009) and this will naturally also affect their fruit and vegetable consumption and willingness to purchase. As mentioned earlier, a large part of willingness to purchase fruit and

(31)

because if someone is an assured cook, they will naturally use vegetables and fruit in their cooking and again, as a result, have a higher willingness to purchase such products. Lastly, relationship status and its relation to willingness to purchase fruit and vegetables is also a an important factor to mention which wasn’t included in this survey and could be investigated in future research. According to Stevens, one pivotal factor affecting men’s diets, in particular, is that they are waiting longer to settle down (2017) and this relationship between marriage and single status is something that could be

researched more thoroughly in the future because it would then therefore provide marketers with data to then target consumers depending on their relationship status.