Health Care Charities in The Netherlands

(1)

(2)

2 Health Care Charities

in The Netherlands

by ing. A.G. Sjoukema

University of Groningen

(3)

3 Abstract

Although a considerable amount of research has been done on charities and charitable giving in general, no research is done yet towards possible differences in giving behavior that occur when charities have different characteristics. In this research this gap is bridged for charities in the health care sector in The Netherlands.

An online experiment is done to obtain insights in these differences and possible moderating effects of demographics on the relationship between charities characteristics and the amount of money donated. The experimental findings show that although demographics do not have a significant moderating effect, differences in characteristics of charities do result in important significant differences in giving behavior. Well-known charities and charities aimed at all ages attract higher donations.

(4)

4 Preface

During the Field Course Marketing of the Master of Business Administration in Marketing Management at the Rijksuniversiteit Groningen, we had to make an assignment about charities. This assignment made me realize that non-profit organizations also need marketing to move forward, but very often don’t have the money to hire a marketer to determine how to gain as much money as possible to help them reach their goal. Realizing this, I decided to search for a charity in the Netherlands that could use a little help from a student graduating in marketing to write my thesis for.

Duchenne Parent Project is a Dutch charity, founded by parents of boys suffering the Duchenne disease (only boys will perceive the effects of the disease). The foundation always needs more money to be able of make it possible to doing more research on developing medicines to increase the quality of life of the boys and eventually – hopefully – to find a solution that will make it possible that the boys don’t have to die any more. Besides that Duchenne Parent Project is a relatively small organization, so they can really use some help in the field of marketing. Besides that, I also have a personal reason to write my thesis holding DPP in mind. Seven years ago, my best friend died because of the Duchenne muscular dystrophy. I hope that my research will contribute to make it possible to build friendships that can last longer than only 21 years for boys that will be born with Duchenne in the future.

I have had a lot of support writing my thesis. I want to thank dr. M.C. Non of the Rijksuniversiteit Groningen for coaching me trough the literature-struggle, her help and useful suggestions, the other supervisors, all participants of my experiment, my friends, family and colleagues for listening to the never ending stories about my research, Christiaan for helping after midnight, Jörgen for doing the typo-bingo, and last – but definitely not least – Gideon for the fact that I had the luck to know him.

(5)

5

6

3.3.1.1 Independent Variable: Type of charity ... 24

3.3.1.2 Dependent Variable ... 24

3.3.1.3 Moderating Variable ... 25

3.3.1.4 Controlling and manipulation check ... 25

3.4 Plan of analysis ... 25

4.Results ... 26

4.1 Descriptive results ... 26

4.1.1 Manipulation Check ... 27

4.2 Familiarity with the charities ... 27

4.3 Donations to charities in the research ... 28

4.4 Testing donations to charities ... 28

4.4.1 Relationship between gender and donation behavior ... 28

4.4.2 Relationship between religion and donation behavior ... 29

4.4.3 Relationship between education and donation behavior ... 30

4.4.4 Relationship between family situation and donation behavior ... 30

4.5 Predicting yearly donations ... 32

4.6 Testing hypotheses ... 33

4.6.1 Charity’s characteristics as predictors for donations ... 33

4.6.2 Including moderating variables... 34

Religion ... 34

Children ... 35

Grandchildren ... 35

Education ... 36

Gender ... 36

4.7 Hypotheses supported/not supported ... 37

5.Conclusions... 38

5.1 Experimental results compared to theory ... 38

5.2 Implications ... 39

5.3 Limitations and further research directions ... 40

References ... 41

(7)

7 1. Introduction

1.1 Background problem

Dutch people tend to give a lot to charities. They donate on average € 144,- per year per person, spread over an average of five different charities (GfK, 2011). This means that all charitable organizations in The Netherlands together have got more than 2,3 billion euro to spend annually.

Since the economic crisis, less money is given to charities in general, but a more important trend is that people are more critical when chosing which charity to support. They also want to know more than ever what happens with their donations (Centraal Bureau voor de Statistiek, 2012).

Charities in The Netherlands can have a CBF-mark. This mark tells that charities spend at least 75% of their income directly to the main goal of the charity (Centraal Bureau Fondsenwerving, 2012). This mark helps people to get confidence towards the charities involved which becomes more important in times of crisis. On the other hand, the 75%-rule makes that charities are not able to invest (a lot) on research. This implies that charities are also not able to invest in marketing research which can help them to make better choices in their marketing strategy.

(8)

8

According to the Stichting Nationale Goede Doelen (2011) and the Vereniging Fondsenwervende Instellingen (2012), the Dutch Charitable Market can be divided into four different main types of charity ‘Health Care’, ‘Socio-economic Development’, ‘Welfare, Sports and Culture’ and ‘Nature, Environment and Animal Care’. This is the most commonly used format. The focus in this research will be on Health Care Charities because of personal interest and preference.

1.2 Problem statement and research questions

A lot of literature can be found about charities. The existing literature describes research on factors influencing how much people donate to charities in general. However, no research can be found about the influence of the type of charity on the amount of money donated. For example, the different effects on the donation of a research-oriented or a relief-oriented health care charity or a health care charity for specific age groups like children. Also, no research can be found about the importance of moderating factors, like demographics such as income, family situation and church attendance, towards different types of health care charities. In this research this gap will be bridged by examining whether the type of health care charity has influence on the amount of money donated and if demographics have a moderating effect on this relationship. Therefore, the problem statement can be formulated in the following way:

Is there an effect of the type of health care charity on the amount of money donated by an individual and do demographics have a moderating effect on this relationship?

In this research there will be assumed that charities want to gain as much money as possible. This is their first goal which makes it possible to reach their main goal in health care.

The problem statement can be translated in the following – more specific - main question and sub questions.

Main Question

Which influencing charity characteristics can be defined for giving-behavior towards given health care charities and what is the moderating effect of the demographics on this relationship?

Sub Questions

(9)

9

1.3 Theoretical and managerial relevance

This study contributes to the existing literature about marketing of charities by determining how the specific characteristics of charities play a role in how much money people will donate to these certain charities.

Existing literature has researched which demographics influence the amount of money one will donate to charities in common, but no literature can be found whether these demographics influence the relationship between types of health care charities and the amount of money donated. And, if they do, in which way. Since there are multiple types of charities, the focus of this research will be on charities in the field of health care in The Netherlands.

The first contribution of this research to the existing literature is to investigate if there are differences in giving behavior when it comes to different types of health charities. The second contribution is the investigation of moderating effects of demographics towards these relationships.

According to the Stichting Nationale Goede Doelen (2011) health care charities in The Netherlands can be categorized using the following labels:

- International oriented/not international oriented - Research/Relief oriented

- Commonly known/not commonly known - Active for specific age groups or not

All health care charities can be categorized by using these labels.

In fact, existing literature only researches which demographics influence giving behavior in general. This information can be used very well when marketers want to market charities, but as explained in chapter 1.1, nowadays it becomes more important to waste as little money as possible in targeting people. In this research there will be investigated if there are differences in how much influence already distinguished demographics have on giving to health care charities with specific characteristics, using the labels provided by the Stichting Nationale Goede Doelen (2011).

The results will contribute to a conclusion in which it will be clear if marketers of charitable organizations have the possibility to target new donors using channels which match the group of people most likely to donate to the specific charity. Recommendations will be given, which can lead to optimization.

1.4 Structure of the thesis

(10)

10 2. Literature Review

Most charities are dependent on donations of people and organizations. The more money donated, the more a charity is able to reach its goal. It is interesting to investigate which people are most likely to give money to charities with certain characteristics.

In this chapter literature about marketing in a charitable environment, charity characteristics, demographics and charitable giving will be discussed in order to determine variables that influence peoples giving behavior. Based on this literature hypotheses will be presented in chapter 2.3.

2.1 Charities

2.1.1 Definition

A charity is a matter of common interest to which one can donate money, goods or time. For instance, people can donate time to a charity by being a volunteer at a charitable organization (Alle Goede Doelen, 2012).

(11)

11

2.1.2 Different types of charities in The Netherlands

In The Netherlands there are 117 charities that are member registered. These charities are active in different fields. There are four fields in which a charity can operate (Stichting Nationale Goede Doelen, 2011 and Vereniging Fondsenwervende Instellingen 2012), these fields are:

- Health care and medical research - Socio-economic development - Welfare, culture and sports - Nature and environment

In this research health care charities is the topic to be investigated. In the health care charity group, charities can have different focuses. According to the Vereninging Fondsenwervende Instellingen (2012) health care charities can be classified by using the following labels:

- Internationally oriented/not international oriented - Research- oriented/relief-oriented

- Commonly known/commonly unknown - Active for specific age groups or for all ages

A good indication for the size of a charity is the turnover (Sargeant and Kähler, 1999). In general, large charities will probably have more financial resources thanks to more donations and are, thanks to these resources, able to become more known trough media. Therefore in this study there will be assumed that a large charity is better known by the Dutch population than a small charity.

There are charities which are only active in The Netherlands but there are also Dutch charities that are part of an international organization. All charities included in this research will be specific Dutch parts of internationally active and internationally oriented organizations but they differ on all the other aspects. 2.2 Marketing for charitable organizations

2.2.1 Marketing charities

When you want people to donate to your charity, people have to know that your charitable organization exists. Marketing and promoting your product (in this case your organization’s goal) is a good way to reach your target group. Although the function of marketing in non-profit organizations differs from marketing in profit businesses, certain business concepts can be adopted to enhance their operations (Shapiro, 1972). The importance of knowing who your customers are and by which factors your customers are motivated to make a donation is as important for charities as it is for other organizations. Since charitable organizations are very often dependent on donations, it is interesting to know why and when a consumer makes a donation and in which way marketing instruments can contribute to more donations and to donations with a higher amount of money.

(12)

12

Figure 1: Giving behavior model (Sargeant and Woodlife, 2007)

Source

The model shows that charitable giving depends on many different factors. One of these factors is called the ‘source’ of the fundraising solicitation. This may be considered as the most important external input for peoples reaction to the ask-question of a charity. Branding, reputation, media, modes of ask, etcetera are all marketing instruments influencing the reaction of people to the ask-question. Kelman (1961) suggested that messages are more likely to be accepted by donors when the organization is already known to them and is perceived as reputable. This illustrates the importance of good marketing management of charitable organizations.

In 1959 Zielske already concluded that the manner in which one is asked to donate is crucial. Liu and Aaker (2008) examined how a focus on time instead of money can lead to two distinct mind-sets that affect consumers’ willingness to donate to charitable organizations. They conclude that when one is asked how much time someone would like to donate to a charity the monetary amount of the donation ultimately increases.

In addition to this, Lampe (2007) claims in his article that people are not so generous when it comes to charitable giving as we might expect. Only when people are confronted with the give-question very often, they consider giving by themselves. Otherwise they automatically think giving is important but it is something other people will do in their place. This implies that people have to be faced with the give-question frequently to make sure a donation will be eventually made.

(13)

13

H0,1 People donate more money to a well-known charity than to a charity that is commonly unknown.

Kelman (1961) suggested that people are more likely to donate when the charity is already known. This implies that when a charity is well-known, it will attract more money than a lesser-known charity.

Perceptual Reaction

The amount of money donated by people will be higher when there is a match between the donor and the charity. Research by Van Diepen, Donkers and Franses (2009) points out that you need to strike when the iron is hot; only when one is willing to donate in the first place, he or she will do so only when a matching charity asks for a donation. According to Frey and Meier (2004) identification with the organization and with specific groups is important as well, because when people can identify themselves with a charities goal, they are more likely to understand why a donation is needed. This will be discussed further in chapter 2.3.3.

Processing Determinants

Two key categories of variables appear to have impact on the way in which the giving decision is processed (Sargeant and Woodlife, 2007). These are the donor’s past experience with a certain charity and the judgmental criteria he/she might use to evaluate potential organizations for support. For example, higher levels of satisfaction in the quality of service provided by the fundraising department are associated with higher levels of donor retention and giving.

External Influences

According to Sargeant and Woodlife (2007) three types of external factors influence one’s giving behavior: models/experiences, communities of participation and public policy. Models and experiences from one’s youth will shape future adult giving behavior. Those growing up in a family with a strong tradition of charitable support will be significantly more likely to exhibit such behaviors themselves (Schervish and Havens, 1997). The provision of models like celebrities, can influence giving behavior as well.

Communities of participation are networks of formal and informal relationships entered by circumstances that brings an individual into contact with need. Also government contributions discourage, or crowd out private contributions (Sargeant and Woodlife, 2007).

Individual Characteristics

(14)

14

Motives

As shown in the model of Sargeant and Woodlife (2007, see page 12), there is a variety of intrinsic motives for charitable support, such as altruism and self-interest, empathy, sympathy, fear/pity/guilt, social justice, prestige and ‘making a difference’ and tax.

Individual giving behavior can be motivated by a need for emotional or material utility and/or pure altruism (Sargeant and Woodlife, 2007). Those individuals motivated by altruism are more likely to exhibit loyalty to the organizations they support.

Empathy may be defined as an individual’s emotional arousal elected by the expression of emotion in another (Shelton and Rogers, 1981). Sargeant and Woodlife (2007) state that individual giving behavior can be motivated by empathy for the beneficiary group. Higher levels of compliance and giving are associated with higher levels of empathy.

The motive ‘sympathy’ has also received a lot attention in literature. Sargeant and Woodlife (2007) say there is a relationship between the degree of sympathy engendered in fundraising appeals and both compliance and the value of support.

Individual giving behavior can also be motivated by feelings of fear, pity or guilt. Higher levels of compliance and giving are associated with higher levels of these factors (Sargeant and Woodlife, 2007). A need for social justice is a motivator for individual giving as well. The influence of this motive increases where the persistence of the need is likely to be short and where the beneficiary cannot be blamed for the origin of their need.

Two newer motives for philanthropy have been identified as prestige and the desire to make a difference. Prestige is about recognition and is therefore also relevant to the feedback aspect (see below) of the giving behavior model (see page 12). Impact philanthropy is based on a donor’s desire to personally make a difference. The defining assumption is that donors give because they enjoy increasing the output of a good themselves (Sargeant and Woodlife, 2007). For example, some donors prefer to give to a homeless person directly, rather than give to a charity for the homeless where their donation is distributed over a large number of recipients.

The last motive for giving behavior Sargeant and Woodlife (2007) identified has to do with taxes. There is a relationship between income taxes and charitable support. Individual giving is price elastic: a change in the level of taxation results in a proportionately higher change in the level of giving.

H0,2 People donate more to a charity for all people than to a charity with children as only beneficiaries

As mentioned, sympathy and empathy will play a role when it comes to giving-behavior. Many people do not have small children (in their direct environment). People without children might not feel empathy towards children but people with children probably still have empathic feelings towards ill people of all ages. So it will be very likely that people in general have more feelings of sympathy and empathy towards ill people of all ages than towards children only. So more will be donated to charities for people of all ages than to a charity with children as only beneficiaries.

H0,3 People donate more money to a charity with a relief-goal than to a charity with a research-goal

(15)

15

Inhibitors

A number of factors have been shown to inhibit individual giving. A lack of money, time, risk to an individual’s ego, and doubts over the worthiness of a cause can all serve to inhibit donor behavior (Sargeant and Woodlife, 2007).

Feedback

Having decided to offer a donation to a nonprofit, donors will typically be thanked by the respective organization in the hope this will be the first stage in building an ongoing relationship with the individual concerned (Sargeant and Woodlife, 2007). When thanking donors for their gift, organizations often append labels to the donor, such as kind, generous and helpful. Positive labeling is positively associated with donor loyalty and subsequent donations (Sargeant and Woodlife, 2007). Donors perceptions of the adequacy of the recognition they receive are positively related to loyalty and subsequent giving behavior.

Outputs

All the previous has got influence on the output of the model of Sargeant and Woodlife (2007): gifts of cash, size of gift and loyalty.

2.2.2 Competition between charities

Now that the model of Sargeant and Woodlife (see page 12) has been discussed, one important factor is not mentioned yet. Competition between charities is not part of the model but it is interesting to investigate whether an extra donation request makes the total revenues of charities increase.

It was assumed by the Direct Marketing Association (2007) that when people get an extra request for a donation, they might be irritated at some point and so this can have consequences for one’s reaction to mailings from other charities but also future requests from the charity itself. So consequently the real effect of a letter was not easy to observe, especially when one donates to more than one charity.

Additional research by Van Diepen et al (2009) on five large Dutch charitable organizations shows that there is a negative competitor effect, although that will fade out very fast. What remains, is a strong cannibalization effect of an extra request on the future revenues of the charity asking for money. The total effect however is positive and covers more than the costs. And the analysis of Van Diepen, Donkers and Franses (2009a) reveals that although direct mailings do result in irritation, this affects surprisingly neither stated nor actual donating behavior. The cannibalization effect is huge, with almost 80% of the revenues, which shows that the main competitor of the charity is the charity itself. So focusing only at the short term revenues can be very damaging for raising money for the charity.

In practice, charities most often use standards that relate to the direct costs of sending a mailing and ignore the effects of a mailing on future revenues. Improved evaluation methods should be taken into account by marketers and managers of charitable organizations.

(16)

16

2.3 Charitable giving

As shown in the model of Sargeant and Woodlife (2007) on page 11 and mentioned in their article as well, giving behavior cannot be described very easy since a lot of (complicated) factors play a role in the giving process. As remarked by Sargeant and Woodlife, individual characteristics of donors play an important role in donation behavior, but the model does not give an adequate consideration to the range of individual characteristics that would determine whether or not one would support a charity. One of the aims of this research is to fill this gap. In this chapter there will be investigated which characteristics play a role in giving behavior. When known which characteristics describe the target group of a charity best, it makes it possible for the charity to differentiate themselves from others and target consumers in a better way.

2.3.1 Definition

Charitable giving is described as a sense of genuine compassion, which is reacted by charitable practices (such as money, goods or time) towards others (Alle Goede Doelen, 2012).

2.3.2 Why people donate

Charitable giving has a positive impact on society and on the people living in it (Wiepking and Maas, 2009). Donors are likely to receive psychological and social benefits from their donation, such as feelings of joy and warm glow or enhanced self-esteem. Misje, Bosnes, Gasdal and Heier (2005) identified three factors that can explain why people donate; understanding (a positive experience associated with donating), value (altruistic or empathic reasons for donating) and esteem (volunteering in order to feel better about oneself). Personal moral norm or moral duty (a sense of moral obligation to society) was included as a fourth factor in some studies but this only had a weak influence on attitude (Misje et al, 2005), or a non-significant influence on intentions. Donating behavior can be influenced by the importance of the charity, the importance of the specific need and social values. Additionally, attitudes towards a charitable organization may also influence people’s behavior towards making donations. Furthermore, Wiepking and Bekkers (2011) distilled eight mechanisms which can be considered as the main stimuli of philanthropic donation behavior: (1) awareness of need, (2) solicitation, (3) costs and benefits, (4) altruism, (5) reputation, (6) psychological benefits, (7) values and (8) efficacy.

Moral emotions represent a key element of the link between moral standards and behavior (Tangney, Stuewig and Mashek, 2007). When one has to decide whether to donate or not, negative feelings such as shame, guilt and embarrassment come up in one’s mind. When the donation is done, the negative feelings will disappear (at least for a while) and will be replaced by the positive feelings elevation, gratitude and pride. Dahl, Honea and Machanda (2005) describe this; people get positive feelings after making a donation instead of keeping their negative feelings after being confronted by (the goal) of a charity. Basil, Ridgeway and Basil (2008) even claim that the impact of empathy on charitable donation is fully mediated by guilt and maladaptive responses.

In addition, Andreoni (1998) states that not only the result of making a donation is what matters for getting a positive feeling, but also the process of giving matters. In this process feelings of ‘warm glow’ may occur, which develops – according to Konow (2009) – a preference for giving per se.

(17)

17

endorse this and emphasize the impact of social information; when people know what their ‘neighbors’

are dealing with, they are more likely to donate money to a charity which matches this issue.

2.3.3 Which people donate

Known is, there are many factors influencing charitable giving. Giving to charitable causes is positively related to, among others, a higher household income, a higher education, church membership, gender of the donor and – as described in paragraph 2.3.2 - the ‘match’ between the donor and the charitable organization.

2.3.3.1 Gender difference

Mesch, Brown, Moore and Hayat (2011) investigated the difference between men and women towards charitable giving in general in the United States. They concluded women respond to emotional appeals more intensely and more frequent than men. Women are more likely to give, and give more than men. This may be so since women’s philanthropy has been shaped to a significant extent and their shifting economic position have given women the possibility to donate more.Another reason for the difference in charitable giving between men and women is given by Eckel and Grossman (1998); after research they concluded women are less selfish than men and more socially-oriented. It is interesting to investigate if this holds true for the Dutch population as well.

According to the fact that women would be more socially-oriented than men and the assumption that men believe more in facts than emotions, the following hypotheses about donation behavior differences between men and women will be tested:

H1,1 Women donate more money to charities of which the beneficiaries are children than men do

From a historical and (also) still an actual point of view, women are more likely to care for children than men. So there may be assumed that women care more about good health conditions for children and so are more likely to donate to charities of which the beneficiaries are children.

H1,2 Men donate a higher amount of money to charities with a research-goal than women

Women are more socially-oriented (Eckel and Grossman, 1998). This makes it more likely for women to donate to a charity with a relief-goal since the effect of the donation is more visible. There may be assumed men believe more in facts than emotions. So they are more likely to give more to charities with a research-goal than women would do.

H1,3 Women donate more money to charities which are commonly unknown than men

Since women are more socially-oriented they are more likely to know about commonly unknown charities than men and so more likely to donate to these charities.

2.3.3.2 Church membership

(18)

18

Bekkers and Wiepking (2007) found a rich literature in the sociology of religion on the relationship

between religious involvement and giving. Positive relations of church membership and frequency of both secular and religious philanthropy appear in almost any article in which this relation was studied. See for example: Bekkers and Wiepking (2007), Davidson and Pyle (1994), Forbes and Zampelli (2013), Van Slyke and Brooks (2005), Zaleski and Zech (1994).

According to Bekkers and Schuyt (2008), being more integrated in a religious network will increase the level of charitable giving. They found higher levels of volunteerism and generosity among church-members than among non-religious, which can be partly explained by levels of church attendance and social pressure to contribute. Part of the positive effect of having a more extended social network, like a religious network, on charitable giving can also be explained by the more frequent requests for donations that people with a more extended network receive (Bekkers and Schuyt, 2008). People with a more extended social network are exposed to solicitation by charitable organizations more often, very likely through people in their own network (e.g., a nephew asking to sponsor his laps in a school walk-a-thon or a neighbor standing at the doorstep around dinnertime asking for a small contribution to his favorite charity). But why do people with more social human capital - or resources - donate larger amounts to charitable organizations? A possible reason for the large donations of people in networks is that one finds positive norms for making regular and substantive charitable donations in those networks (Wiepking and Maas, 2009).

A religious network is a specific example of a general social network. Still, when person A is compared with person B, both with extensive social networks, and person A is religious and person B is not, person A will be more likely to donate more money to a charity. As mentioned before, research has shown the religious network of person A will lead to more intensive donation behavior than just the social network of persons A and B would do (Bekkers and Schuyt, 2008).

In the United States of America it is only a small step from religion to political preference. In the republican (red) states people are generally more religious than in the democratic (blue) states. In The Netherlands, the step between political preference and religion is a much bigger one since the political system in The Netherlands differs a lot from the system in the United States. Although religion plays another role in the Dutch society than in the American, some remarks can be made on the agreements between religion and political preferences in the US. Karlan and List (2007) have found a relation between generosity and political preference in the United States of America, provided that there is a match between the charity and the donor. The match between a donor and the charity has more effect on the gift results in ‘red states’ (republican states, more religious) than it has in ‘blue states’ (democratic states, less religious). This can be considered as additional support for the assumption that religious people donate more than non-religious people do.

(19)

19

2.3.3.3 Education

Besides gender and religion, education is also a factor influencing giving behavior. Having a higher level of formal education is positively related to charitable donations (Wiepking, 2008). As expected, this effect is partly due to the larger financial resources people with a higher education have access to. However the generosity of the higher educated is not only caused by their better financial situation. People with a higher level of formal education also have stronger verbal abilities, which facilitates a better understanding of the needs of (distant) other people, increasing charitable donations (Wiepking, 2008).

It has been suggested by Pentecost et al (2007) that people are less philanthropic than expected when holding their general willingness to donate in mind. People say they intend to donate to certain charities but the actual donation will not follow automatically. They found that different forms of donating behavior go along with different consumer perceptions. These perceptions influence the chance that one makes an actual donation. Pentecost et al (2007) suggest donating behavior to be influenced by the extent to which individuals perceive the relevance of the charitable organization, the importance of a specific donation activity, or their level of involvement behavior. People with a higher education are more able to see the relevance of a charity and a donation and so are likely to accept less in return than lower educated people. Causing they are more likely to make an actual donation when they have the intention to do so.

In addition, people with more formal education and stronger verbal abilities also have more trust that donations will be spent well, something which also increases their donations(Wiepking and Maas, 2009). We conclude that those with higher education are more charitable, because they have more financial resources, better verbal abilities and - to a smaller extent - a higher level of trust.

When people have got a higher education, one can imagine people are informed and up to date to a higher extent about charities and their goals. People with higher education read more newspapers, are able to interpret information better and can remember things more often. But a higher income and a better ability to understand needs in the society are not the only reasons why higher educated people are expected to donate more money to charities than people with lower education. People with a higher education also have more social contacts than people with a lower education (Wiepking and Maas, 2009). The effects of social and human resources on charitable giving are discussed in chapter 2.3.3.2 and are largely intertwined. Apart from this reasoning in chapter 2.3.3.2, the positive result of having a more extended social network on level of charitable giving can also be explained by the fact that individuals with a larger social network show more trust, more empathy concern and stronger verbal abilities and are more often aware of (health) issues in society (Wiepking and Maas, 2009).

(20)

20

H2,1 People with higher education are more likely to donate money to charities with a research-goal

than people with lower education

Higher educated people are expected to be more aware of the importance of investing in research.

H2,2 People with higher education are more likely to donate to charities which are commonly unknown

than people with lower education

Higher educated people are expected to be more aware of the society and existing charities than lower educated people. So higher educated people will know more charities, also the lesser-known ones. Besides that, higher educated people are also more able to understand the importance of supporting smaller charities and are expected to have more trust in organizations.

Not enough literature can be found to support any relation between higher education and supporting charities of which the beneficiaries are children. Therefore exploratory research on this relationship will be provided in chapter 4.

2.3.3.4 Family Situation

Chapter 2.3.2 already discussed that the match between the charity’s goal and the frame of reference of the donor plays a very important role when it comes to donating behavior. Since the biggest difference in beneficiaries in health care charities is between charities for children and charities for people of all ages, that difference will be distinguished in our present study. Meier (2007) concludes that a matching mechanism increases contributions. In our research there is a match between a participant and the charity for children’s goal when the particpant has children or grandchildren between 0 and 18 years old. Although Shapiro (1972) claims that marketing in non-profit organizations does not differ much from marketing in for-profit organizations, Rothschild (1979) investigated major differences between profit and non-profit organizations; the biggest difference is the presence of very high or very low involvement in non-profit organizations. Known is, that family is something very important (mostly most important) for people. This means that when it comes to matching charities based on family situation, the involment will be very high when there is a match between the consumer and the charity.

In this study health care charities from which the beneficiaries are children will be included. According to the literature there may be said that there is a match between the charity and the donor when the donor has children or grandchildren.

Considering this, the following hypotheses can be made:

H3,1 People with (grand)children donate more money to a charity with a research-goal

People with children are expected to be aware of the future since they want all the best for the next generation as well. Research takes more time to accomplish than relief so it might be possible that this awareness will also lead to higher donations to charities with a research-goal.

H3,2 People with (grand)children donate more money to a charity of which the beneficiaries are children

(21)

21

In the reviewed literature no real support for any statement about the influence of family situation on the

relationship between how well a charity is known and the donation can be found. Therefore this relationship will be researched in an exploratory way.

2.4 Conceptual Model

Summarizing the literature, four very important factors which influence the amount of money donated to a health care charity in The Netherlands are:

- Gender - Education - Religion

- Family situation

The different dimensions of the health care charities to be researched are: - Goal of the charity (research or relief)

- Beneficiaries of the charity (special age group or not)

- Acquaintance of the charity (known by a lot of people or not)

The hypotheses presented in the previous paragraphs are translated into the following conceptual model.

3.

4.

Figure 2: Conceptual model (many relationships in this model will be researched in an exploratory way, therefore no positive or negative expectations are shown in this model).

Demographics (MV) Gender Religion Education Family Situation Donation (DV) Charities’ characteristics (IV)

(22)

22 3. Methodology

3.1 Research Design

Research designs can broadly be classified as exploratory or conclusive. Exploratory research is conducted to develop initial insights and to provide direction and input for any further research needed, and is essential when a problem needs to be defined more precise, whereas conclusive research is conducted to draw conclusions about a problem and is more structured and quantitative. It is designed to assist the decision makers in determining, evaluating, and selecting the best course of action in a given situation. The output of exploratory research may serve as input for conclusive research, whereas the output of conclusive research may serve as input for managerial decision-making (Malhotra, 2009).

The objective of most of this research is to test certain hypotheses and examine presumed relations between different variables. Therefore, conclusive research is being practiced there. Based on the literature discussed in chapter 2.3.3.2, no hypothesis can be made about the relationship of religion and giving behavior. Analyzing this relationship thus will be an exploratory part in this research.

The objective of this research is to determine cause and effect relationships; three independent variables (characteristics of charities) are guaranteed, while the control over the moderating variables (demographics) are manipulated.

(23)

23

3.1.2 Experiment

The way in which data for testing the hypotheses will be collected is by using an online experiment. Experimentation is commonly used to infer causal relationships and it is a process of manipulation of one or more independent variables and measuring their effect on one or more dependent variables while controlling the extraneous variables (Malhotra, 2009). In this research, the experiment focuses on the causal relations between types of charity (independent variables) and the money donated (dependent variables) towards this charity, with different demographics as moderating variables. Therefore, the definition ‘experiment’ applies for this research.

3.1.3 Questionnaire and procedure

The online experiment exists of an online questionnaire, each addressed to the target group: Dutch people. Participants in this research will receive a hyperlink through e-mail, Twitter, LinkedIn or Facebook The questionnaire starts with a neutral introduction and an explanation of the procedure of the experiment, without explaining the purpose of the research. After the introduction a few exploratory questions will be asked about the knowledge of the participant towards the topic of research. After these questions, a small case is presented; the participants are asked to indicate their commitment towards the charities by dividing 100 euro’s over the charities included in the research. Each participant may divide the same amount of money since that makes it possible to research the influence of the moderating variables on the donations, unless the income of the participants. The amount of €100,- has been chosen since that will make it easy to translate the amount of money donated into percentages. At the end of the experiment, the demographics of the participants are asked in order to determine if these demographics give different moderating effects towards the different charities.

At the end of the questionnaire the participant can indicate if he likes to receive the results of the research when the research is done. After that, the results will be send to the researcher. When the researcher has (more than) 150 completed questionnaires, the online experiment is ended.

The experiment will be held in Dutch to make sure all participants are able to understand the questions as intended. The questions for the experiment can be founded in the appendix of this research paper. 3.2 Participants and sample

(24)

24

3.3 Variables

3.3.1 Description and manipulation of variables

In this research, the following variables will be considered:

- Independent variable: Characteristics of the charities included - Dependent variable: Amount of money donated

- Moderating variables: Demographics of participants

The variables will be described with regards to the how and why behind manipulating these variables in this experiment in order to test the different hypotheses as presented in chapter 2.

3.3.1.1 Independent Variable: Type of charity

The independent variable in this experiment is the type of charity. The focus of this experiment is on the characteristics a charity can be categorized by, according to the Stichting Nationale Goede Doelen (2011). As discussed in chapter 2, the health care charities in The Netherlands are categorized by three labels:

- Target group: For specific ages/ For all ages - Focus: Research/Relief

- Well-known by people: Yes/No

For this research three charities are used to determine which label has which impact on the amount of money donated. These charities and their labels are:

- Duchenne Parent Project: Children, research, not well-known - Make-A-Wish Nederland: Children, relief, well-known

- KWF Kankerbestrijding: All ages, research, well-known All these charities are part of an international oriented organization

According to facts and figures of the charities itself, MAW and KWF are well-known in The Netherlands and DPP is not (Make-A-Wish Nederland 2012, KWF Kankerbestrijding 2012, Duchenne Parent Project 2012). In the experiment a question is included to check if this is true for the participants of this research. 3.3.1.2 Dependent Variable

In the experiment, participants get a virtual € 100,- to divide between the three charities that are part of the research. The dependent variable is the amount of money donated to each charity. This is an amount between € 0,- and € 100,- per charity. The participants may also choose to donate (a certain part) to another health care charity. This makes it possible to donate the same amount to each charity included in the research or to indicate that you do not support the charities included at all.

(25)

25

3.3.1.3 Moderating Variable

The moderating variables used in this experiment are the demographics of the participants. Gender, religion, education and family situation are indicated as the most important factors which predict giving behavior of people. In this research it will be determined whether these factors influence the giving behavior of people differently when it comes to charities with different characteristics.

As a part of the experiment, participants fill out their demographics. This information will be used to determine if there is any moderating influence on giving behavior towards the different charities.

3.3.1.4 Controlling and manipulation check

The respondents of the sample group are asked to indicate their commitment to the charities included in the research by answering questions about if and to what extent they know the charity already before they participated in the research.

A manipulation check is included at the end of the online experiment. All participants are asked what the main target group of KWF Kankerbestrijding is to check if they understood the provided information correctly.

3.4 Plan of analysis

In order to test the relationship between the characteristics of the charities and the amount of money donated, a regression analysis is used. The characteristics of the charities are the independent variables and the amount of money donated is the dependent variable.

To test the moderating effect of demographics a regression analysis is used as well. A regression analysis is chosen since it can be used to predict the value of one variable on the basis of other variables (Keller, 2009), which is necessary for analyzing the data from the experiment.

(26)

26 4. Results

All tests are considered to be significant at a level of 5%.

4.1 Descriptive results

Subjects were drawn from a convenience sample of 211 respondents. 181 are retained for analysis, which means there was a usable response of 85,8 percent. The other information is removed from the dataset since almost no actual answers were given by these respondents causing this information to be not usable for analysis. Respondents who only did not give the answer on a single question are still included in the research. The remaining 181 respondents live spread all over The Netherlands, were born between 1937 and 1992 and have an average age of 41 years.

72 Men participated in the research and 107 women did, 2 respondents did not fill out their gender. 68 People are religious and 113 don’t have a religion. Most participants (137) are high educated (HBO/Bachelor or WO/Master) while 44 participants are low educated ((V)MBO/HAVO/VWO or primary school).

(27)

27

Grandchildren Yes No Total Children Yes 4 29 33 No 26 119 145 Total 30 148 178

Table 1: People with children and/or grandchildren

88,4 percent of the respondents do know someone in their immediate vicinity who has or had cancer. 22,1 percent do know a critically ill child and 28,7 percent of the respondents know someone with a muscular disease.

4.1.1 Manipulation Check

To check whether participants had read and understood all the provided information and asked questions which were part of the experiment correct, an extra question was included. 95,6 percent of the participants answered this question correctly, which indicates that the answers given during the experiment are reliable.

Frequency Percent Cumulative Percent Answer Incorrect 5 2,8 2,8

Correct 173 95,6 98,4

Missing 3 1,6 100

Total 181 100

Table 2: Answers on manipulation check

4.2 Familiarity with the charities

In the online experiment, three charities were included to represent a combination of the different characteristics a health care charity in The Netherlands can have. As mentioned in chapter 3, assumed is Duchenne Parent Project is the least known charity included in the research and KWF Kankerbestrijding and Make-A-Wish Nederland are well-known in The Netherlands.

The familiarity of the respondents towards these three charities is measured by using a 7-point Likert scale and the results are presented in the table below.

Mean Standard Deviation KWF Kankerbestrijding 5,36 1,363

Duchenne Parent Project 2,03 1,514

Make-A-Wish Nederland 4,48 1,590

Table 3: Mean scores of the familiarity of respondents towards charities included

These means all differ significantly from each other. After performing three t-tests, it shows that the significance level between the difference of the means of KWF and DPP is 0,000, between DPP and MAW 0,000 and between MAW and KWF is 0,001.

(28)

28

Kankerbestrijding. So KWF Kankerbestrijding may be considered as the most known charity included in

this research.

4.3 Donations to charities in the research

During the online experiment, participants got a virtual € 100,- to divide between KWF Kankerbestrijding, Duchenne Parent Project and Make-A-Wish Nederland. The fourth option was to donate to another health care charity in The Netherlands. 174 participants choose to donate to KWF, 153 participants to DPP, 163 participants to MAW and only 78 participants choose to donate to another health care charity. In the table below the descriptive results are represented. A quick scan of the results shows that KWF got on average the highest donation (€ 38,12). MAW got an average donation of € 28,99 and DPP got an average donation of € 22,92. The option of donating money to another charity is chosen as well with an average of € 9,40.

Minimum Maximum Mean Standard Deviation Donation KWF 0 100 38,12 21,790

Donation DPP 0 70 22,29 13,517

Donation MAW 0 100 28,99 19,156

Donation other 0 50 9,40 12,971

Table 4: Donations to charities included in the research

4.4 Testing donations to charities

As mentioned in chapter 1, Dutch people give, on average, € 144,- per year to charities. To investigate whether or not the participants of this research behave like the Dutch population, the answers given in the experiment are re-calculated onto real amounts of donation. During the online experiment participants indicated their yearly donation to charities using a range of amounts. A possible answer on this question could be ‘a yearly donation between € 120,- and € 200,-‘. To be able to use this answer in the analysis, the mean of the range of this answer is used for calculation, in this case € 160,-. This is also done for the other possible answers on this question.

As shown in the table below, the average donation of the participants of this research lies around the Dutch average amount, however the difference between the donated amounts is relatively large. One respondent did not gave an indication of his or her average donation per year.

Number Minimum Maximum Mean Standard Deviation Donation per year 180 60 700 198,56 200,095

Table 5: Average donations per year (all participants)

4.4.1 Relationship between gender and donation behavior

Gender Number Mean Standard Deviation Donation per year Men 72 209,31 218,879

Women 106 186,51 185,265

Table 6: Average donations per year (men/women)

(29)

29

behavior of men and women is not true. However, the significance level of a 2-sample t-test is 0,11 which

means that there may not be stated yet that men really donate more money than women.

When we have a look at the incomes of both men and women as represented in the table below, there seems to be a difference in incomes between men and women participating in this experiment. Besides that, the incomes also seem to be relatively high when compared to the average Dutch income (€ 33.000,- per year). This is the result of the fact that in the online experiment participants were asked to fill out the total family income. Therefore there may not be concluded that men earn more money than women.

Gender Number Mean Standard Deviation Income per year Men 72 62791,67 33875,991

Women 103 50941,75 35950,286

Table 7: Differences in income between men and women

In the next table the differences in amount of donation between men and women are shown when it comes to the charities included in this research. For both charities included in the research which have children as beneficiaries (DPP and MAW), women donate more money than men. Men donate more to charities for all people.

Gender Number Mean Standard Deviation Donation KWF Men 72 43,49 23,376

Women 107 34,37 20,129

Donation DPP Men 72 21,40 14,252

Women 107 23,95 13,094

Donation MAW Men 72 26,76 20,665

Women 107 30,80 17,952

Donation other Men 72 8,49 12,025

Women 107 9,81 13,359

Table 8: Differences in amount of donations between men and women on charities included in the research

4.4.2 Relationship between religion and donation behavior

Shown in the table below is the difference in the amount of money donated per year between religious and non-religious people. One respondent did not answer the question whether or not he or she is religious. As expected after reviewing literature, religious people do donate more money (€ 255,97) than non-religious people do (€ 164,51). This two-sample t-test has a significance level of 0,003 which means that we may state that religious people donate more money to charities than non-religious people do.

Religion Number Mean Standard Deviation Donation per year Yes 67 255,97 224,355

No 113 164,51 176,596

Table 9: Average donations per year (religion/no religion)

The next table shows the differences in the average amount of money donated between religious and non-religious people. The most remarkable result is that religious people donate more money to other health care charities than non-religious people do. A possible explanation might be that the charities in this research do not have a religious background.

(30)

30

choices than non-religious people when they know less about the charity since there is no significant

difference in how known religious people indicate DPP than the other participants of the experiment.

Religion Number Mean Standard Deviation Donation KWF Yes 68 36,40 17,571

No 113 39,15 23,990

Donation DPP Yes 68 25,32 12,481

No 113 21,48 13,959

Donation MAW Yes 68 24,72 12,036

No 113 31,57 22,030

Donation other Yes 68 12,09 13,525

No 113 7,78 12,407

Table 10: Differences in amount of money donated to different charities and religion

4.4.3 Relationship between education and donation behavior

Education Number Mean Standard Deviation Donation per year High 136 187,57 194,381

Low 44 232,50 215,602

Table 11: Average donations per year (low/high educated)

In the table above is shown how much people with higher education (HBO/Bachelor or WO/Master) and people with lower education (VWO/HAVO/MBO, VMBO and primary school) donate on average per year. Based on the reviewed literature in chapter 2, the opposite was expected; these results implies that lower educated respondents of this research donate more money per year than the higher educated respondents of this experiment do. The significance level of this two-sample t-test is 0,303 so the difference is not significant.

Education Number Mean Standard Deviation Donation KWF High 137 39,20 22,999

Low 44 34,73 17,293

Donation DPP High 137 22,12 14,330

Low 44 25,43 10,339

Donation MAW High 137 28,58 20,828

Low 44 30,27 12,719

Donation other High 137 9,42 13,431

Low 44 9,34 11,566

Table 12: Differences in amount of money donated between high and low educated people

The most remarkable result (see table 12) in differences in donating behavior between high and low educated people is that high educated people donate a whole lot more to charities for all people than low educated people. Besides that KWF is also the best known charity with a research-goal which might imply that higher educated people give more to charities with a research goal in general than low educated people.

4.4.4 Relationship between family situation and donation behavior

(31)

31

average € 313,50 per year and people without grandchildren € 175,57. The significance level of this

comparison is 0,005. A possible reason for the larger donations of people with grandchildren might be that grandparents are probably older people than the other participants of the research and older people are more likely to have enough money or do not have to save for the future any more.

Children Number Mean Std. Deviation Donations per year Yes 33 197,58 194,153

No 144 200,97 203,462

Table 13: Average donations per year (children yes/no)

Grandchildren Number Mean Std. Deviation Donations per year Yes 30 313,50 240,399

No 150 175,57 183,464

Table 14: Average donations per year (grandchildren yes/no)

In the tables below the differences in donations between people with or without (grand)children towards the different charities included in the research are shown.

Children Number Mean Standard Deviation Donation KWF Yes 33 39,61 24,290

No 145 37,81 21,356

Donation DPP Yes 33 24,42 12,676

No 145 22,57 13,854

Donation MAW Yes 33 29,27 15,513

No 145 29,05 20,110

No 145 9,86 13,570

Table 15: Differences in donations between people with children and people without

Children Number Mean Standard Deviation Donation KWF Yes 30 41,07 24,344

No 151 37,53 21,287

Donation DPP Yes 30 21,30 14,791

No 151 23,25 13,279

Donation MAW Yes 30 27,07 18,338

No 151 29,38 19,350

No 151 9,17 13,158

Table 16: Differences in donations between people with grandchildren and people without

When looking to these numbers remarkable is that the differences in the average amount of money donated between people with or without children is relatively small. The differences between people with or without grandchildren are a little larger but also not as big as expected based on the literature reviewed in chapter 2. Besides that the differences are also in another direction than expected based on the literature.

(32)

32

4.5 Predicting yearly donations

Regression analysis is used to predict the value of one variable on the basis of other variables (Keller, 2009). Gender, religion, education and family situation can be seen as predictors for the yearly donation of consumers (to all charities). A multiple regression analysis is done to determine how the relationship between these factors and the amount of money donated annually can be described or predicted. Age and income are asked in the online experiment as well, and may also have influence on the yearly donation. Therefore this information is included in the analysis, besides gender, religion, education level and family situation.

Unstandardized Coefficients t Significance Collinearity Statistics B Standard Error Tolerance VIF Constant -31,029 62,353 -,498 ,619 Age 3,937 1,313 2,997 ,003 ,449 2,229 Religion -54,364 30,072 -1,808 ,072 ,801 1,249 Children -16,466 34,923 -,471 ,638 ,928 1,078 Grandchildren 25,702 48,122 ,534 ,594 ,514 1,946 Education 2,653 32,667 ,081 ,935 ,855 1,170 Gender -10,628 27,266 -,390 ,697 ,939 1,065 Income ,002 ,000 4,556 ,000 ,793 1,261

Table 17: Coefficients of the regression analysis (dependent variable: donations/year)

The R2-value of this regression analysis is 0,293. This means that 29,3% of the yearly donation is predicted by this model. The Variance Inflation Factor (VIF) indicates whether a predictor has a strong linear relationship with the other predictors (Field, 2005). If the largest VIF of the analysis is larger than 10 there is cause for concern. As shown in table 17 the VIF’s of this regression give no cause for concern at all since the largest is VIF value is 2,233. The significance level of the f-test indicates the significance of the whole analysis. The significance level of this regression analysis is 0,000 which means that the analysis is significant. As shown in the table above, income and age are significant in predicting the yearly donation of the participants of the experiment.

The total yearly donation to charities of participants of this experiment can be predicted with the formula:

f(x) = b0 + b1*gender + b2*children + b3*grandchildren + b4*education + b5*religion+ b6*age+ b7*income According to the regression analysis, the formula for predicting the yearly donation is:

Yearly donation = -31,029 - 10,628 * gender + 0.002 * income + 3,937 * age – 54,364 * religion – 16,466 * children + 25,702 * grandchildren – 2,653 * education

The formula shows that there is a positive relationship between the yearly donation and income and age. Income seems to have only a very small influence, but 0,002 times an average yearly income (€ 33.000,-) is still € 66,- so that illustrates that income does have actual influence on the donation.

(33)

33

0 1

Gender Women Men

Children No Yes

Grandchildren No Yes

Religion Religious Non-religious

Education Low educated High educated

Table 18: Explanation dummy variables

When taking the dummy variables and the output of the regression analysis together, this gives the following information:

- Women donate € 10,63 more than men on a yearly basis.

- People with children donate € 16,47 less than people without children on a yearly basis.

- People with grandchildren donate € 25,70 more than people without grandchildren on a yearly basis.

- Religious people donate € 54,36 more than people without religion on a yearly basis. - Low educated people donate € 2,65 more than high educated people on a yearly basis. Since these parameters are not significant the results are not very reliable.

4.6 Testing hypotheses

4.6.1 Charity’s characteristics as predictors for donations

In the experiment of this research participants got a virtual € 100,- to divide between KWF Kankerbestrijding, Duchenne Parent Project and Make-A-Wish Nederland. It is interesting to analyze the influence of the charities’ characteristics on the amount of money the different charities got from the participants. To determine which role familiarity, goal and beneficiaries play in predicting the amount of donation a regression analysis is done.

Unstandardized Coefficients

t Significance

Collinearity Statistics B Standard Error Tolerance VIF Constant 18,603 2,763 6,734 ,000

Acquaintance 2,308 ,535 4,313 ,000 ,514 1,946

Goal -,385 2,322 -,166 ,868 ,508 1,968

Beneficiaries 7,454 2,622 2,843 ,005 ,399 2,508

Table 19: Coefficients of the regression analysis (dependent variable: yearly donation (euro))

The R2-value of this analysis is 0,133. This means that 13,3% of the amount donated towards the charities in this experiment can be explained by using this model. The f-test has a significance level of 0,000, which means that the analysis is significant. The VIF-values are all lower than 10 so there is no cause for concern. As shown in the table above, familiarity and beneficiaries are significant predictors for the amount of money donated to the charities included in the research and goal is not significant as predictor.

The total yearly donation to charities of participants of this experiment can be predicted with the formula:

Health Care Charities in The Netherlands

2

Health Care Charities

in The Netherlands

3

Abstract

4

Preface

5

Table of Contents

6

7

1. Introduction

8

9

10

2. Literature Review

11

12

13

14

15

16

17

18

19

20

21

22

3. Methodology

23

24

25

26

4. Results

27

28

29

30

31

32

33