MSc Marketing Thesis

Increasing charitable donations through disrupting

price frames in the context of fundraising messages

MSc Marketing Thesis

by

Paula Veldman

2

Increasing charitable donations through disrupting

price frames in the context of fundraising messages

MSc Marketing Thesis

by

Paula Veldman

University of Groningen

Faculty of Economics and Business

MSc Marketing

January 2017

Professor Enno Dirk Wiersmastraat 3/11

9713 GH Groningen

0634612401

p.a.veldman@student.rug.nl

Student number 2812614

3

TABLE OF CONTENT

ABSTRACT . 4 ACKNOWLEDGEMENTS . 5

1 INTRODUCTION . 6 2 LITERATURE REVIEW AND HYPOTHESES .

2.1 Conceptual model .

2.2 Literature review and hypotheses .

9 9 11 3 METHODOLOGY .

3.1 Data collection and research design .

3.2 Variable operationalization .

3.3 Sample and manipulation check .

3.4 Method of analysis . 13 13 15 18 19 4 RESULTS . 20

5 CONCLUSION, DISCUSSION, LIMITATIONS AND FURTHER .

4 ABSTRACT

Many articles have examined how to frame a charity message to increase charitable giving, but little is known about what influence the way a price is framed has on peoples’ donating behavior. The present research aims to address this gap by investigating how perceptions of donations framed as a small daily amount (pennies-a-day theory) or as a one-time relatively large amount (exceptional expense theory) influence the likelihood to donate. Based on the principles of mental accounting and budgeting processes it is assumed that these price frames have a disrupting effect, which increases the likelihood to donate. Furthermore, the effect of the disrupting price frames on the likelihood to donate might be moderated by the trust in the charity organization. Previous research showed that a high degree of trust leads to higher price tolerance and lower price sensitivity. This means that people are willing to pay more money to an organization they trust, while they also pay less attention to the price. Therefore, people may be less sensitive to the disruption of the price frame.

A total of 399 participants were randomly assigned to three scenarios, at which they saw an advertisement of a charity organization that differed in the way the price was framed. After presenting them the advertisement, the participants were asked to fill in a questionnaire to measure if the different scenarios influenced their likelihood to donate.

This research found evidence that suggests that when the price in charity messages is framed as a disrupting price frame (pennies-a-day or exceptional expense), this directly increases the likelihood to donate. This implies that the likelihood to donate increased when the price was framed as a small daily amount or a high amount framed in an exceptional time frame. Lastly, the proposed moderating role of the trust in a charity was not supported by this research.

Key words: Disruptiveness of price frame, pennies-a-day theory, exceptional expense

5

ACKNOWLEDGEMENTS

This master thesis marks the end of my Master Marketing Management at the University of Groningen. Writing this thesis has been very instructive and challenging as well. I want to thank several people that have helped me with writing this master thesis.

First I would like to thank my supervisor Dr. J.C. Hoekstra. She always gave me helpful feedback and kept me motivated to work on my thesis with her positive attitude. Furthermore, I would like to thank my fellow student and dear friend Rosanne Heijligers, not only for the past period but for all her help and support during the entire master period. I also want to thank Jasper Hidding who checked my thesis on my English writing.

Finally, a special word of thank to my parents Richard Veldman and Gea Veldman, I want to thank them for supporting me unconditionally and giving me the opportunity to do a Master. Without them I would never have been able to achieve this.

6

1. INTRODUCTION

Charity organizations are everywhere. Whether one is watching television, listening to the radio, being online or when traveling. People are confronted with messages from charity organizations everywhere. All these messages have the same goal: to try and get more donations. In the Netherlands there are a lot of charity organizations, each of which has its own specific cause. This means that there are a lot of organizations that have to compete with each other for a limited pool of money. Raising money is key to the organizations’ ability to serve their specific causes. According to Consumer Union (2013), each year nearly five billion euro is donated to charity organizations. But who is getting this money? Due to governmental regulations that have led to cuts in financial support for charity organizations, competitiveness among charities has increased. Due to this development, charity organizations have become more dependent on private donations (Das, Kerkhof and Kuyper, 2008). This in turn has caused charity organizations to spend a higher share of the donations to raise money. As a result of this, charity organizations need to spend donated money that could have otherwise been spent on the cause, to raise money instead. Andreoni and Payne (2011) indicate that most charities spend 5 to 25 percent of its donations on further fund-raising activities.

Marketing strategies have become a fundamental tool in the nonprofit sector to raise donations (Venable, Rose, Bush and Gilbert, 2005). One of the marketing tools that can be used is the way the content of a fundraising message is framed. Several message factors are critical in order to increase the perceived value of a charity’s cause and raise donations intentions (Bendapudi, Singh and Bendapudi, 1996). Smith and Berger (1996) argue that potential donors make judgments about the request the charity makes in the message. Is there a need? How great is the need? When donors see a charity message, they determine whether to give to the charity (a choice judgment) and how much to give (an estimation judgment). To raise more donations, it is important to know which marketing tactics influence this decision process and which marketing tactics, like framing, may enhance the marketing effectiveness of charity organizations.

7

(1 euro versus 30 euros) and the timeframe (per day versus per month) differ. This can influence the way in which people process the message. This has to do with mental accounting, which is related to how people process their budgeting (Gourville, 1998).

A price frame can be used to influence a person mental budgeting process (Sussman, Sharma and Alter, 2015). A budgeting process is the procedure by which an individual creates and manages a financial plan (Business dictionary 2016). When individuals start the budgeting process, they automatically use mental accounting. The budgeting process and mental accounting are interrelated. Thaler (1999) defined mental accounting as the set of cognitive operations used by individuals to structure, evaluate, and keep track of financial activities. The process of mental accounting consists of two stages: booking and posting. Booking the expense involves people noticing and recording the expense in their accounting system, while posting means they assign the costs to a specific account in their mind. To affect the budgeting process in the proportion to its magnitude, the expense should be booked and posted to a meaningful account. When people fail to execute one of the steps of the process, they can insufficiently account for the appropriate size or impact that the expense has on their budget. If people are not able to execute both steps, the budgeting process might be disrupted (Thaler, 1999). A disrupted budgeting process means that people are not able to put an expense into a budget category, which leads to affects such as overconsumption or underconsumption (Sussman et al., 2015).

8

small expenses could affect their budget as well. As a consequence of this, consumers fail to book small expenses like a daily cup of coffee (Thaler, 1999). The PAD model shows that consumers are more likely to donate when the budgeting process is disrupted (Gourville, 1998).

The effect of the price frame can be influenced by the trust the individual has in the charity organization. When a consumer is confronted with a fundraising message they have to deal with a social dilemma. A social dilemma can be explained as the difficult choice between self-interest and the interest of the collective. A factor that influences the social dilemma is the trust an individual has in the charity, as well as the believe that the charity can help solve the problem (Klandermans, 1992). Saxton (1995) and Sargeant (1999) argue that a higher degree of trust in a charity organization may be associated with a greater willingness to become a donor, as well as the willingness to donate a larger amount of money. Delgado-Ballester and Munuera-Aleman (2001) showed that trust leads to higher price tolerance. A result of the higher price tolerance is that people are less price sensitive (Anderson, 1996). When there is low price sensitivity, people assign less weight to the price when they evaluate the product (Erdem Swait and Louviere, 2002). People may be more willing to donate regardless of the price frame because they pay less attention to the price. As a result of this people fail to book the expense in their accounting system because they do not notice or record the price frame. Consequently, when people trust an organization, it is likely that they will donate anyway. They do not need the disruptiveness of the price frames to increase the likelihood that they will donate. According to these assumptions, it can be argued that the effect of the disruptive price framing on peoples intention to donate can be moderated by the trust an individual has in the specific charity organization.

9

because of its practical implications. Explorative research will be conducted to test which price frame is most effective. The third contribution is the moderation analysis which tests if the disrupting price frame effect still holds when there is a high degree of trust in the charity organization. This moderating effect has never been tested before in combination with the disrupting price frame theory and may therefore have academic and managerial relevance.

In chapter 2 the conceptual model and the underlying theories are described. This will result in the hypotheses as formulated at the end of chapter 2. These hypotheses will be tested by creating an online choice experiment. Chapter 3 describes the research methodology and is followed by chapter 4 which describes the results of the research. In the 5th and last chapter, the results will be discussed.

2. LITERATURE REVIEW AND HYPOTHESES

The following section consists of the conceptual model, literature review and hypotheses. First, in section 2.1 the conceptual model is shown. Section 2.2 describes the theories that are related to the variables used in this research. These theories are then explained in more detail which culminates into two hypotheses.

2.1 Conceptual Model

The conceptual model (Figure 1) assumes that there is a positive relationship between the disruptiveness of the price frame and the likelihood to donate. According to Sussman et al. (2015) and Gourville (1998) disrupting price frames have a disrupting effect on the individual budgeting process, which leads to a higher likelihood to donate. This relationship may be moderated by the trust in the charity organization.

10

Figure 1. Conceptual model

Income and age are included in this study as control variables. Older people generally donate a higher percentage of their income to charity organizations than do younger people (CBS, 2015). Other control variables are gender and education level. Research (Ogawa and Ida, 2015) found that individual characteristics, such as gender and education level significantly influence donating behavior. That is why it is relevant to control for these variables when assessing the outcomes. Also, the current donating behavior of the participants and the ties they have to a charity can affect the willingness to donate. Because of that, the research covers how often they donate, how much they donate to charity organizations per year, their donation proneness, their brand attitude relative to the charity used in the scenario and whether they already donate to this charity.

11 2.2 Literature Review and Hypotheses

The two main theories used in this research are the pennies-a-day (PAD) and the exceptional expense theory. These price frames both have a disrupting influence on the budgeting process and therefore are disrupted price frames. Both disrupting price frames are time related and are dependent on the way the price is framed in the fundraising message. While the requested donation is framed as a small daily donation in the PAD theory, it is framed as a relatively larger and uncommon, infrequent donation in the exceptional expense theory. In the PAD strategy, the requested donation is an aggregate expense that is reframed into a series of smaller ongoing expenses, which are subsequently framed per day (Gourville, 1998). For example, a consumer may be more likely to donate when asked for €1 per day than when asked for an annual donation of €365 to the same cause. Gourville (1998) indicates that PAD framing decreases the monetary magnitude of a consumer’s transaction relative to a more aggregate frame, like a monthly or annual price frame. It makes the product seem more affordable. This effect influences the attractiveness of the proposed transaction and therefore might increases the likelihood to donate. However, Thaler’s (1985) theory of hedonic editing predicts that a series of small costs may have a negative effect on the likelihood to donate. Hedonic editing means that consumers prefer to psychologically integrate a series of small costs, into one large bundled cost instead of experiencing many small costs.

12

for an exceptional expense than for an ordinary one, since exceptional expenses are harder to classify into a single category. Sussman et al. (2015) argued that framing donations as more exceptional (instead of ordinary/annual framing) may prevent consumers from processing the full magnitude of the expense, which leads to budgeting errors and ultimately results in overconsumption. This budgeting error will in turn influence the likelihood to donate.

According to the disrupting price frame theory, it is expected that the exceptional expense and the PAD price frame both have a positive effect on the likelihood to donate. This means that a disrupting price frame might have a significantly greater influence on the likelihood to donate than does a non-disrupting price frame. To test if this theory holds, the following hypothesis is formulated:

H1: The disruptiveness of price frames positively influences the likelihood to donate

13

less into consideration, and are therefore more willing to donate. If people trust the organization, they may pay less attention to the price frame in the fundraising message, because they put less weight on the price when they make the decision to donate. This might weaken the effect of the disrupting price frames on the likelihood to donate, because people do not notice or record the price frame due to their trust in the charity organization. This is also an essential stage in the effect of the disrupting price frames. Trust in a charity organization is already a disruption in the budgeting process in a different way. Trust therefore probably weakens the effect of the disrupting price frames. This moderating effect is formulated in H2:

H2: Trust weakens the effect of disruptiveness of the price frame on the likelihood to donate

3. METHODOLOGY

This chapter describes the methodology used for this research. First, section 3.1 explains in which way the data are collected and what the research design looks like. After that, section 3.2 describes how the variables used in this research are operationalized. Section 3.3 describes the sample used in this research and the manipulation check, and finally in section 3.4 the method of analysis is described.

3.1 Data Collection and Research Design

To test the hypotheses, a scenario-based experiment with a between-subject design was used. In this experiment, the participant read a specific scenario in which the price was framed (Figure 2). The disruptiveness of the price frame was divided into three levels being either a daily, monthly or a one-time donation. The two disrupting price frames are the daily (PAD frame) and one-time donation (exceptional expense frame). A third variable was needed to control for the disrupting price frames. It was also important for this third variable to not be disrupting. Therefore a monthly non-disrupting price frame was chosen. This non-disrupting price frame was then tested in combination with the two disrupting price frames to measure the effects of the PAD and exceptional expense theories independently.

14

budget processes like the disrupting price frames do. This is because it is easier for people to mentally account for a monthly expense in their budgeting process.

In the research of Sussman et al. (2015), the annual price frame is used as an ordinary expense frame. An ordinary expense, which consumers perceive to be common and frequently, is used to test the effect of the exceptional expense. However, in this message frame there is only a small difference in the formulation of the exceptional expense and the ordinary expense. In Sussman et al. (2015), the exceptional and ordinary expense are both framed in an annual time frame. The ordinary expense is formulated as ‘held annually’, while the exceptional expense is formulated as ‘only once a year’. In fact, both expenses are annual expenses. The only way in which the mental accounting process differs is in the way the message is perceived. To make a clear distinction between an annual price frame and an exceptional expense, the exceptional expense was not described as ‘only once a year’, but as a one-time expense in this research. Furthermore, in this research the non-disrupting price frame was not an annual, but rather a monthly frame.

The three different price frames were framed in a fundraising message to test the effect of the disrupting price frames. The charity organization used in the message was Unicef. Unicef is a well-known charity organization who fights for the rights of all children in the world. Before the participants saw the message they have to read the introduction with instructions for the questionnaire. In this introduction, participants were asked to look at the fundraising message for about 20 seconds, before they started the questionnaire. The fundraising message was focused on helping kids and to give them a good future. On the left picture (Figure 2) there was a girl who looked sad and needed help, on the right picture there were kids who went to school with the help of Unicef. The text in the message explained the problem, which is the situation a lot of children live in (left picture). The end of the message was about the goal to give these children a future (right picture).

15

TABLE 1

Scenario Price Time frame Theory

1 €0,15 Per day Pennies-a-day

For €0,15 a day, you can let them be a child again

2 €4,50 Per month Neutral condition

For €4,50 a month, you can let them be a child again

3 €55,- One time donation Exceptional expense

With a one-time donation of €55,-, you can let them be a child again

To help the children, respondents were asked by Unicef to donate to this cause. The requested donation was framed with one of the three different price frame levels: daily, monthly or the one-time donation frame. The total donation for each scenario was actually equal when calculated over a year. The participants were randomly allocated to one of the three scenarios. After the participants read and processed the scenario, they had to fill out a questionnaire to give an indication of their feelings about likelihood to donate and the degree of trust. In order to check the questionnaire, a test was performed with eight people. The results of the pre-test showed that people noticed the price frame which is a crucial factor to pre-test the hypotheses. They did not found any confounding things in the questionnaire. To control if the participants had seen the manipulation of the price frames there was a manipulation check. In the manipulation check, respondents were asked which price frame they had seen. To get a varied sample the participants were selected through different networks like fellow students, family, friends and colleagues.

3.2 Variable Operationalization

16

The scale of Sharma and Chan (2011) was used for donation proneness. Both scales were adjusted to fit into the right context. Finally, all items were measured on a 7-point scale, where 1 is totally disagree and 7 is totally agree. A 7-point scale was chosen because such a scale gives a good balance between scale reliability and discriminative demand of the respondent. The multi-item scales can be seen in Table 2 and the measurements of constructs of the control variables in Table 3. To measure the likelihood to donate, trust in the company,

TABLE 2

Factor analysis 1

Item

Factor loading

1 2 3

Trust in the charity organization Start question: I see Unicef as..

- .. Undependable – dependable - .. Dishonest – honest - .. Unreliable – reliable - .. Insincere – sincere - .. Untrustworthy – trustworthy Cronbach’s Alpha .812 .904 .864 .911 .903 (.967) .356 .321 Donation proneness

- Donating makes me feel good

- I feel excited when donating to charity

- When I donate, I feel that I am helping someone in need - I enjoy donating, regardless of the amount I give

Cronbach’s Alpha .890 .874 .779 .808 (.888)

Attitude towards the brand

- The decision to donate to Unicef is foolish - Donating to Unicef is a good decision

- I think that Unicef is a satisfactory charity organization - I think Unicef has a lot of beneficial characteristics - I have a favorable opinion on Unicef

Cronbach’s Alpha .330 .338 .339 .366 .810 .857 .870 .849 (.951) Factor analysis 2 Item Factor loading 1 Likelihood to donate

- The probability that I would consider to donate is - My willingness to donate is

- The likelihood that I would donate is

Cronbach’s Alpha

.771 .805 .801

(.866)

The bold was used to hi ghli ght t he fact that the factors hung

17

donation proneness and brand attitude, factor analysis was used to reduce the different item scale items into four different factors. Before doing the factor analysis, a KMO test and Bartlett’s test of sphericity were used to determine whether factor analysis was appropriate. The results of the first factor analysis with the independent variables showed a KMO of 0.916 and a Bartlett’s test of sphericity of 0.000 (P < .01). These scores indicated that the use of a factor analysis was appropriate for this data (Malhotra, 2010) and that it would factor well based on the observed correlation. The last check of appropriateness was testing the communalities. These communalities should be higher than 0.4 to be appropriate (Voerman, 2015). Except for one item (‘The decision to donate to Unicef is foolish’, 0.366) all extractions were higher than 0.4 (0.779 to 0.911). The rotated component matrix illustrated that this item loaded on two components with a factor below 0.4 which means that this item is

TABLE 3

Measurements of constructs control variables Manipulation check: Which donation

proposal was made in the advertisement?

What is your age?

(Open question)

- For €0,15 a day - For €4,50 a month

- With a one-time donation of €55,-,

….

What is your gender? Did you ever made a donation to Unicef?

- Male - Female

- Yes - No

What is your highest education level? What is your netto monthly income?

- Elementary education - LBO, MAVO, VMBO - HAVO or VWO - MBO

- HBO - WO

- Less than € 1.000 per month - € 1.000 tot € 2.000 per month - € 2.000 tot € 3.000 per month - € 3.000 tot € 4.000 per month - More than € 4.000 per month

How often do you donate to charity organizations?

How much do you donate to charity organizations per year?

- Never - Once a year - 2-3 times a year - 4-6 times a year

- More than 6 times a year

18

not appropriate for further analyses. In addition, the rotated component matrix indicated that the items can be classified into three factors. The factor analysis for the dependent variable, the likelihood to donate, showed a KMO of 0.738, a Bartlett’s test of sphericity of 0.000 (P < .01) and no factor loadings lower than 0.4 (0.771 to 0.805) which means that all three items were appropriate for further analyses.

A scale reliability test measuring the Cronbach Alpha of the construct was used following the factor analysis. The Cronbach’s Alpha for all factors was between 0.866 and 0.967 which is considered to be good to excellent reliability. Deleting one of the items would not increase the Cronbach’s Alpha, thus all items were kept and used to compute a multi-item scale for the likelihood to donate, the trust in the company, donation proneness and the brand attitude.

3.3 Sample and Manipulation Check

Out of the 371 questionnaires collected, 339 were found usable after the exclusion of seven straight-liners and 25 respondents who failed the manipulation check. The question asked in order to check for manipulation errors was: Which donation proposal was made in the advertisement? Respondents who did not choose the price frame as they saw in the advertisement failed the manipulation check. In this research it was decided to delete respondents who failed the manipulation check, because this meant the effect of the disruptive price frame could not be tested since the respondent did not notice or see the price in the advertisement. After conducting a linear regression analysis, the results showed that the sample with the manipulation errors weakened the significant effect, leading to a decrease of the p-value. This confirmed the suspicion that due to the manipulation check errors the direct effect of the disruptive price frames could not be measured correctly when the manipulation errors were included in the sample. Therefore, all manipulation check errors were deleted from the sample before further analyses were conducted. This resulted in a sample of 339 respondents which were used to run the regression analyses.

19 3.4 Method of Analysis

To test the hypotheses with the obtained data, several linear regression analyses were used. The control variables, (i.e. gender, age, income, education and donating behavior) were first included in a regression analysis in order to validate the results and account for possible confounds (model 1). Besides the direct effect, with and without the control variables (model 2 and 3), two additional regression analyses were conducted. In this way, the effects of both disrupted price frames were tested independently in addition to the direct effect (model 2A and 2B). Finally the moderating effect was regressed following model 4. Each disrupting price frame was tested in a regression analysis with the non-disrupting price frame, the neutral variable. The analyses were meant to estimate the following models:

Model 1 Dependent variable with control variables

Ltd =

α

ε

Model 2 Direct effect

Ltd =

α

ε

Model 2 A Pennies-a-day theory

Ltd =

α

ε

Model 2 B Exceptional expense theory

Ltd =

α

ε

Model 3 Direct effect + donation proneness + brand attitude

Ltd =

α

ε

Model 4 Moderating effect

Ltd =

α

ε

Ltd = Likelihood to donate Ba= Brand attitude

D = Disruptiveness of the price frame Dp= Donation proneness

Tr = Trust in the charity organization

ε

20

To test the effect of the disruptive price frames on the likelihood to donate, the disruptive price frames were coded as dummy variables. The disruptive price frames in scenarios one and three were used as a dummy with a value ‘1’, and the non-disruptive price frame at scenario two is coded with a value ‘0’. The value of the disruptiveness depended on the scenario the respondents were assigned to in the questionnaire.

4. RESULTS

An analysis is conducted using the responses from the 339 respondents. First, it is important to check for correlation between the control variables and the disruptiveness of the price frame. Therefore a correlation analysis is conducted. The correlation analysis only shows a significant correlation for the disruptiveness of the price frame with education level. No further relevant significant correlations are found. In order to determine that there is no correlating effect between any variables, a regression analysis is used to analyze multicollinearity. To see whether there is multicollinearity, the Variance Inflation Factor statistic is checked. None of the VIF scores exceeds a value of 4, it can therefore be concluded that there is no multicollinearity (Belsley, 1991).

Before testing the hypothesis all control variables are regressed on the dependent variable, likelihood to donate (model 1). As can be seen in Table 4, the control variable ‘donation proneness’ yielded a significant effect with a p-value of 0.000 and ‘brand attitude’ was significant with a p-value of 0.004. Since the control variables have no significant correlation with the independent variable, disruptiveness of the price, and none of the VIF scores is higher than 4, donation proneness and brand attitude are included as a control variable to test the hypothesis.

21

The model shows that the disruptiveness of the price frames has a positive significant effect on the likelihood to donate with a standardized beta of 0.087. This is in line with hypothesis 1, which stated that respondents who saw the disrupting price frame are more likely to donate than those who saw a non-disrupting price frame.

To see whether both actually have the disrupting effect that they are assumed to have, the PAD and exceptional expense frames are also tested independently from each other (model 2A and model 2B). To test these theories independently from each other, two separate regressions are run. One uses scenarios 1 and 2 as input (model 2A), and another one uses scenarios 2 and 3 as input (model 2B). These distinctions are made because scenario 2 is used as a neutral condition whereas scenario 1 and 3 represent the different disrupting price frames. The regression analysis shows that both theories have a significant positive effect on the

TABLE 4

Disruptiveness of the price frame H1 (+) .125** .133** .138** .087*** .081*** Control variables Donation proneness .344* .368* .362* Brand attitude .161* .184* .117*** Age -.074 Gender -.003 Income -.002 Education level -.013 Donations made to Unicef -.091 Amount donated to

charities per year

.103 How often they donate to

charities per year

-.048

Moderator

Trust in the charity .110***

Interaction effects

Trust in the charity* disruptiveness of the price frame H2 (-) .015 R2 (Adjusted R2) .251 (.231) .016 (.013) .018 (.014) .019 (.015) .247 (.240) .254 (.243) R2 – change F-value 12.270* 5.390** 4.176** 4.316** 36.534* .007* 22.642*

22

likelihood to donate. The PAD frame is found to be significant at a p-value of 0.042 with an adjusted R square of 0.014. The exceptional expense frame also has a significant effect at a p-value of 0.039 with an adjusted R square of 0.015. When looking at the likelihood to donate of the respondents in the case where they saw the disruptive price frames, the PAD frame shows a mean of 3.368 and the exceptional expense frame shows a mean of 3.355.

Now that the analysis shows that the direct effect is significant, another regression analysis is used to test the second hypothesis (model 4). This linear regression analysis assesses whether trust in the charity organization has a moderating effect on the effect of disruptiveness of the price frame on the likelihood to donate. Its adjusted R square is 0.254. However, with a p-value of 0.758, the moderating effect does not reach significance in the regression analysis. The regression did show a significant effect of trust in the charity organization (p < 0.01). The control variable ‘donation proneness’ remains significant at a p-level of <0.01 and so does the ‘brand attitude’ with a p-p-level of <0.1. The direct effect of the disruptiveness of the price frames also remains significant with a p-value of 0.091. When testing for multicollinearity with the VIF statistics, these again do not indicate that multicollinearity is an issue. Hence, due to the moderating variable not reaching significance, hypothesis 2 cannot be accepted. There is no statistical evidence to conclude that trust has a moderating effect on the effect of the disruptiveness of price frames on the likelihood to donate.

5. CONCLUSION, DISCUSSION, LIMITATIONS AND FURTHER

RESEARCH

23

Therefore, it can be concluded that both disruptive price frames have an approximately equal effect on the likelihood to donate.

Donation proneness and brand attitude are found to be highly significant, which indicates that the effect on the dependent variable, likelihood to donate, can also be explained by the respondents’ donation proneness and brand attitude. Hence, the willingness to donate generally has an influence on the likelihood to donate and this will not change much after the respondents see the advertisement. The significant effect of brand attitude towards the brand used in the scenarios also has a positive effect on the likelihood to donate. This implies that people who have a negative brand attitude are less likely to donate than people who have a more favorable attitude towards the brand. However, a correlation analysis showed that there is not a significant correlation between the significant control variables (donation proneness and brand attitude) and the disruptiveness of the price frames. Furthermore, the different disruptive price frames in the scenarios have a significant effect on the likelihood to donate. This implies that the disruptiveness of the price frame has a significant effect on its own. In conclusion, when the donation proneness and brand attitude are included in the regression analysis the effect of the disruptive price frames weakens because of the strong effect of the donation proneness and brand attitude on the likelihood to donate. Not because they influence the direct effect of the disruptiveness of the price frame on the likelihood to donate.

While trust in the charity organization does not significantly moderate the effect of the disruptiveness of the price frame on the likelihood to donate, there is a significant direct effect of the trust in the charity organization on the likelihood to donate. This is in line with Saxton (1995) and Sargeant (1999) who argue that a higher degree of trust in a charity organization may be associated with a greater willingness to donate. In conclusion, these results imply that trust in the charity organization has a direct effect on the likelihood to donate, but does not weaken the effect of the disruptive price frames on the likelihood to donate. It might be that trust in the organization is not strong enough to fade the effect of the disruptiveness of the price frame on the likelihood to donate. One explanation could be that a high degree of trust does indeed have a positive effect on the likelihood to donate, but that this does not imply that people are also really motivated to make a donation to Unicef whereby they pay less attention to the monetary value. Therefore trust might not interact with the effect of the disruptiveness of the price frame on the likelihood to donate as was expected. Hence, motivation might be a stronger predictor which can be tested in further research.

24

mean that the price in their communication should be framed as either a small daily amount or a high amount in an exceptional time frame. This result can also be used in a different context. It might be interesting for advertising companies. A disruptive price frame can be used in advertisements to increase sales. Besides, this research contributes to the existing research with a broader understanding of the application of both theories.

There are a number of limitations to the research. An important limitation is the high level of education of the respondents. Of the 339 respondents, 72% completed an HBO or WO education, which are relative high educational levels. People with a higher educational level might perceive messages in advertisements different than people with a lower educational level. However, there is no indication of how, and if, the outcomes might be different. In addition, the ANOVA analysis showed that the educational level is not equally distributed across the scenarios. When this research would be performed again, a more equal distribution of educational level across the scenarios could lead to different results. Another limitation is the influence of the brand used in the advertisement. In this research, a well-known charity organization, Unicef, is used for the scenarios. Approximately half of the respondents had at some point in their lives made a donation to Unicef. If a less known charitable organization would have been used, trust in the charity organization could have been lower, which in turn might have influenced the likelihood to donate in general. As a result of this, the disruptiveness of the price frame might have had a different effect on the likelihood to donate. This in turn could mean that the disruptiveness of the price frames has a different effect for a less known brand. Another limitation for this research is that the donation intention is measured instead of real donations.

25

References

Andreoni, J., & Payne, A. A. (2011), Is crowding out due entirely to fundraising? Evidence from a panel of charities. Journal of Public Economics, 95(5), 334-343.

Anderson, E., & Weitz, B. (1989). Determinants of continuity in conventional industrial channel dyads. Marketing science, 8(4), 310-323.

Anderson, E. W. (1996), Customer satisfaction and price tolerance. Marketing letters, 7(3), 265-274.

Belsley, D. A. (1991), Conditioning diagnostics. John Wiley & Sons, Inc. 1991 Bendapudi, N., Singh, S., & Bendapudi, V. (1996), Enhancing helping behavior: An integrative framework for promotion planning. Journal of Marketing, 60, 33-49. Business dictionary (2016), Budget process. Cited at: 17 oktober 2016. From: http://www.businessdictionary.com/definition/budget-process.html

CBS (2015), Nederlands steeds ruimhartiger voor geode doel. Cited at: 16 September 2016. From: https://www.cbs.nl/nl-nl/nieuws/2015/51/nederland-steeds-ruimhartiger-voor-goede-doel

Cheema, A., & Soman, D. (2006), Malleable mental accounting: The effect of flexibility on the justification of attractive spending and consumption decisions. Journal of Consumer

Psychology, 16(1), 33-44

Consumer Union (Consumentenbond) (2013), Slecht zicht op goede doelen,

Consumentengids, February, 38-41.

26

Delgado-Ballester, E., & Luis Munuera-Alemán, J. (2001). Brand trust in the context of consumer loyalty. European Journal of marketing, 35(11/12), 1238-1258.

Erdem, T., Swait, J., & Louviere, J. (2002), The impact of brand credibility on consumer price sensitivity. International journal of Research in Marketing,19(1), 1-19.

Gourville, J. T. (1998), Pennies-a-day: The effect of temporal reframing on transaction evaluation. Journal of Consumer Research, 24(4), 395-408.

Klandermans, B. (1992), Persuasive communication: Measures to overcome real-life social dilemmas. Social dilemmas: Theoretical issues and research findings, 307, 318.

Malhotra, N. K. (2008), Marketing research: An applied orientation, 5/e. Pearson Education India.

Ogawa, K., & Ida, T. (2015), Investigating donating behavior using hypothetical dictator game experiments. Review of Social Economy, 73(2), 176-195.

Ohanian, R. (1990), Construction and validation of a scale to measure celebrity endorsers' perceived expertise, trustworthiness, and attractiveness. Journal of advertising, 19(3), 39-52. Pennington, R., Wilcox, H.D., & Grover, V. (2003), The Role of System Trust in Business-to-Consumer Transactions, Journal of management Information Systems, 20(3), 197-226.

Putrevu, S., & Lord, K. R. (1994), Comparative and noncomparative advertising: Attitudinal effects under cognitive and affective involvement conditions. Journal of Advertising, 23(2), 77-91.

Sargeant, A. (1999), Charitable giving: Towards a model of donor behaviour. Journal of

Marketing Management, 15(4), 215-238.

27

Saxton, J. (1995), A strong charity brand comes from strong beliefs and values. Journal of

Brand Management, 2(4), 211-220.

Sharma, P., & Chan, R. Y. (2011), Counterfeit proneness: Conceptualization and scale development. Journal of Marketing management, 27(5-6), 602-626.

Smith, G. E., & Berger, P. D. (1996), The impact of direct marketing appeals on charitable marketing effectiveness. Journal of the Academy of Marketing Science, 24(3), 219-231. Sussman, A. B., & Alter, A. L. (2012), The exception is the rule: Underestimating and overspending on exceptional expenses. Journal of Consumer Research, 39(4), 800-814. Sussman, A.B., E. Sharma & A.L. Alter (2015), Framing Charitable Donations as Exceptional Expenses Increases Giving, Journal of Experimental Psychology: Applied, 21 (2) 130-139. Thaler, R. H. (1999), Mental accounting matters. Journal of Behavioral decision

making, 12(3), 183.

Venable, B. T., Rose, G. M., Bush, V. D., & Gilbert, F. W. (2005), The role of brand personality in charitable giving: An assessment and validation. Journal of the academy of

marketing science, 33(3), 295-312.

Voerman, L. (2015), Factor Analysis: Reducing data and finding dimensions. University of

28

Appendix 1. SCENARIOS

Scenario 1 – Pennies-a-day price frame

Scenario 2 – Neutral price frame