The effects of framing on Greenpeace : will the advertisements of Greenpeace be more effective when a positive frame is used instead of a negative frame?

(1)

The Effects of Framing on Greenpeace

Will the advertisements of Greenpeace be more effective when a

positive frame is used instead of a negative frame?

Master’s thesis Romana van Kuijk

Graduate School of Communication University of Amsterdam

(2)

The Effects of Framing on Greenpeace

Will the advertisements of Greenpeace be more effective when a

positive frame is used instead of a negative frame ?

Romana van Kuijk

Student number: 10316736

Master’s Thesis

University of Amsterdam

Graduate School of Communication

Master’s Programme of Communication Science

Supervisor: dr. Aart S. Velthuijsen Date of Completion: June 26, 2014

(3)

Table of Contents Title Page ... 1 Table of Contents ... 3 Abstract... 4 Introduction ... 4 Theoretical Background ... 6 Method ... 13

Participants and Design ... 13

Procedure ... 13

Stimulus Material ... 13

Measures ... 15

Manipulation Check ... 16

Results ... 16

Preliminary Operations and Analyses ... 16

Hypothesis 1 ... 17

Hypothesis 2 ... 19

Hypothesis 3 ... 22

Explorative Analyses ... 25

Conclusion and Discussion ... 25

(4)

Abstract

This study examines whether the marketing communication strategy of Greenpeace can be made more effective by using positively framed messages instead of the often used negatively framed messages. Former research on framing effects has indicated that

individuals who are exposed to positively framed messages have a higher purchase intention and a more positive attitude towards the advertised product than individuals who are

exposed to negatively framed messages (Smith, 1996). To test whether Greenpeace should also make use of positively framed messages an experiment was performed. For the

experiment, one Greenpeace advertisement was designed with a positive text and photo and one advertisement with a negative text and photo. The participants in the experiment (N = 51) were shown either the positively framed or the negatively framed advertisement and were asked questions concerning their intention to donate money, intention to sign a Greenpeace petition, and attitude toward Greenpeace. When controlling for mood effects, the individuals in the positive manipulation condition showed a slightly higher intention to donate money, a slightly lower intention to sign a petition, and a slightly more positive attitude toward Greenpeace than the individuals in the negative manipulation condition. Unfortunately, the results were not significant. Future research should further investigate this issue.

Introduction

Currently, there is a campaign for the protection of the Indonesian rainforests. Environmental organizations such as Greenpeace and WWF announced that for the production of palm oil, the Indonesian rainforest gets destroyed. This destruction has an enormous impact on the animals and people who live there, but also on climate change. To attract attention to this issue, Greenpeace has launched a campaign where they made the Sumatran tiger the symbol of the rainforest. This strategy can be effective, as many people

(5)

like tigers and are able to identify with them (Leidsch Dagblad, 2014). However, Greenpeace does not often use an animal as a symbol for its campaigns. Greenpeace is more concerned with the impact on the environment instead of its impact on the endangered animals that live there. Because without a healthy environment, the animals will become extinct anyway. But Greenpeace did decide to use the Sumatran tiger as the symbol for the protection of the Indonesian rainforest, as this can trigger people to sign their petition or even donate money. This strategy of using an animal as a symbol for a specific campaign is often used by another environmental organization, the World Wildlife Fund. They have often used cute animals to trigger positive emotions or feelings towards the campaign or towards WWF in general. WWF is one of the largest environmental organizations so one may conclude that this strategy has been effective for them. On the other hand, Greenpeace is also a successful environmental organization that uses shocking images rather than images with beautiful animals that evoke positive emotions. Greenpeace likes to use so called bearing witness photos of the destruction of the rainforests or the Arctic. However, these photos evoke negative feelings that may even cause people to stop donating money to Greenpeace or generate negative feelings toward the organization. Pascalle (personal communication, March 15, 2014) told me that she used to donate money to Greenpeace. She likes the organization and the goals it attempts to accomplish, but she really did not like that the emails she received were full of negative information and images. The communication in the emails focused on unethical events and that we should encourage companies with

unsustainable practices to stop what they are doing and become sustainable. Pascalle did not like the negativity within the emails and , therefore, decided to stop donating money to Greenpeace. If the emails would contain more positive messages or messages that evoked more positive emotions, she probably would not have stopped donating money. One may expect that Pascalle is not the only person that stopped her donations due to the negativity in the messages sent by Greenpeace. Greenpeace employee Christopher Hay (2011) also acknowledged the negative effects that the negative messages may have and said: “Today we are bearing witness in a nice way – showing the world what can be saved, rather than

(6)

what is being killed”. However, most advertisements still contain images and messages that are framed in a rather negative way. Not so much focusing on what can be saved but on the destruction itself. Framing the advertisements in a positive way may make the

advertisements more effective. People may generate more positive feelings toward Greenpeace, donate money or sign its petitions. The big question is whether the use of a positive frame instead of a negative frame will lead to more effective advertising campaigns.

Research question: To what extent does the use of a positive or a negative frame in an advertisement of Greenpeace influence the intention to donate money, the intention to sign petitions and the attitude toward Greenpeace?

Theoretical background

Message framing has been described in many ways. Gamson and Modigliani (1987) define a frame in communication to be the words, phrases, presentation styles, and images that a speaker uses when communicating information about a certain issue to an audience. Krishnamurthy, Carter, and Blair (2001) describe framing as the presentation of logically similar options in semantically dissimilar ways. For example, a product can be described with an 85% satisfactory rate (positive frame) or a 15% dissatisfactory rate (negative frame). Both descriptions present the exact same information, but in a very different way. Another clear example would be: “By donating money to our charity, you can save thousands of lives” (positive frame) versus “By not donating to our charity, thousands of lives can be lost” (negative frame). The persuasiveness of either a positively or a negatively framed message can be explained by the Elaboration Likelihood Model of persuasion by Petty and Cacioppo (1986).

The Elaboration Likelihood Model (Petty & Cacioppo, 1986) suggests that a variable can affect attitudes and persuasion by functioning as a simple cue, by serving as an argument, or by influencing the extent to which the information is being processed. Which role the variable serves is influenced by the motivation and ability of the recipient to process

(7)

the presented information (Petty & Cacioppo, 1986). When the recipient enjoys thinking carefully or is interested in the subject, the motivation to process information concerning the subject is high. However, when the recipient does not enjoy thinking carefully or has little interest in the subject, the motivation to process information concerning the subject is low. The ability of the recipient to process information depends on intelligence, prior knowledge on the subject and whether there are any distractions that hinder the recipient from carefully processing the provided information (Petty & Cacioppo, 1986). When the motivation and/or ability to process information are low, peripheral processing will take place via the peripheral route. The provided arguments receive less attention and more attention will be paid to peripheral cues, such as heuristics, affect, simple associations and identifications (Petty, Gleicher, & Baker, 1991). Positively framed messages often evoke positive affect and, therefore, receive more attention and are more persuasive than negatively framed message when the motivation and/or ability to process information are low. When the motivation and ability to process information are high, central processing will take place via the central route where prior knowledge and experience is used to carefully scrutinize the arguments.

Negatively framed messages are more persuasive than positively framed messages when central processing takes place. This can be explained by the Negativity Bias.

The Negativity Bias claims that negative messages receive more weight and are processed more carefully than positive messages (Meyerowitz & Chaiken 1987; Smith & Petty, 1996). When faced with a negatively framed message, individuals want to avoid the possible losses communicated in the message and are more likely to be persuaded to start behaving in a loss avoidant manner than when individuals are faced with a positively framed message (Kahneman & Tversky, 1979; Davis, 1995). For example, a campaign persuading smokers to quit smoking will communicate the positive aspects or benefits of not smoking in a positively framed message. A negatively framed message concerning smoking will

communicate the negative aspects or losses of smoking, for example lung cancer. Because the losses are possible and relevant to the target audience, their motivation to process the provided information concerning the subject will be high. With a high motivation and ability to

(8)

process the information, central processing occurs where the messages and the provided arguments are processed more carefully. Messages communicating the possibility of lung cancer will receive more weight and will be more persuasive than messages communicating the possibility of a long and healthy life. Therefore, when the motivation and ability to process information are high, negatively framed messages will be more persuasive than positively framed messages. The Positivity Bias shows a different explanation for the persuasiveness of a negatively framed message.

The Positivity Bias describes that individuals tend to expect information to be presented in a positive frame (Meyerowitz & Chaiken, 1987). Negatively framed messages are,

therefore, not expected, more attention grabbing, and receive greater scrutiny (Meyerowitz & Chaiken, 1987; Ditto & Lopez, 1992; Homer & Yoon, 1992; Pratto & John, 1991) than

positively framed messages. Negatively framed messages receive greater information processing than positively framed messages and are, therefore, processed via the central route (Smith & Petty, 1996).

The heuristic-systematic model (Chaiken, 1987) suggests that when there is a high processing motivation and the information is processed via the central route, negatively framed information is more persuasive than positively framed information (Maheswaran & Meyers-Levy, 1990; Rothman, Salovey, Antone, Keough, & Martin, 1993; Maheswaran & Meyers-Levy, 2004). However, when the personal relevance is low and the information is processed via the peripheral route, positively framed information is more persuasive than negatively framed information (Rothman, Salovey, Antone, Keough, & Martin, 1993; Maheswaran & Meyers-Levy, 2004). When the personal relevance is low or the individuals are generally uninvolved with the message, a simple cue effect occurs where the positive words simply generate positive associations (Maheswaran & Meyers-Levy, 1990).

The interaction between the Elaboration Likelihood Model and framing has also been shown in a study by Das, Kerkhof and Kuiper (2008). The researchers asked the participants to read a fundraising message for the Dutch Leprosy Foundation and to indicate their

(9)

attitude toward the message, and their intention to donate money to the foundation. The results indicated that an effective fundraising message should either combine a negative frame with abstract, statistical information or combine a positive frame with anecdotal evidence. This is in line with the previously mentioned theories and results. In order to process statistical evidence you need to use prior knowledge and experience. Therefore, statistical evidence is being processed via the central route where the negative frame is more persuasive. Anecdotal evidence, on the other hand, makes use of more peripheral cues. When the information is being processed via the peripheral route, the positive frame is more persuasive (Chaiken, Liberman, & Eagly, 1989).

Even though the previously mentioned results are highly useful and clearly show the differences in persuasiveness for positive and negative frames, they entail gain- and loss-framing. Gain- and loss-framing involves the presence of a certain risk. In this research, however, the perceived risk is quite low. To many people, environmental issues occur in countries and oceans far away and the devastating effects on the environment are not tangible to them. Therefore, some individuals fail to see how these environmental issues have an impact on their lives and perceive them to be of a low risk (Rathzel & Uzzell, 2009). Gain- and loss-framing does not seem to apply to this particular case. Therefore, the more general framing will be used where the focus is on positive versus negative framing.

Framing effects include the effects of all sorts of message framing and shows that the frame of a message has an effect on its persuasiveness (Smith & Petty, 1996). In a study by Levin (1987), participants were asked to evaluate ground beef that was either described as “25% fat” or “75% lean”. The study showed that the participants who read the positively phrased attribute labels (“75% lean”) had more favorable associations with the beef than the participants in the negative manipulation condition (“25% fat”). It has been suggested that positive frames may be more effective than negative frames because they generate positive associations that influence the persuasion and evaluation (Levin & Gaeth, 1988; Levin, Schneider, & Gaeth, 1998). Another study by Levin, Johnson, Deldin, Carstens, Cressey, and Davis (1986) showed that participants were more likely to accept a gamble that was

(10)

described with positive words (probability of winning) than when it was described with negative words (probability of losing). Buda and Zhang (2000) also showed these results. In their experiment, participants read a description of a stereo receiver that either

communicated an 85% satisfactory rate (positive frame) or a 15% dissatisfactory rate (negative frame). The participants in the positive manipulation condition rated the product significantly higher on its attractiveness, perceived performance, and their willingness to buy the product than the participants in the negative manipulation condition. Smith (1996)

provided the participants in his experiment with an ad of a video camera that had either negatively or positively framed sentences. He found that in the positive manipulation condition the attitude toward the ad and the purchase likelihood were higher than in the negative manipulation condition.

Even though the theories and researches mentioned above differ in many ways, there seems to be a general trend that indicates that the positive frame rather than the negative frame may lead to more effective advertisements for Greenpeace. In general, individuals do not consider environmental issues to be personally relevant and are, therefore, likely to process Greenpeace advertisements via the peripheral route. Research on positive versus negative framing effects has also shown that individuals in a positive manipulation condition have a higher purchase intention and a more positive attitude toward the ad and the

advertised product than the individuals in the negative manipulation condition. Unfortunately, not much research has been performed on the positive versus negative framing effects on the intention to donate money or to sign a petition of a charitable organization. However, as purchase intention is a behavioral intention, it is expected that a positive manipulation condition will also lead to behavioral intentions in relation to charitable organizations. The study by Das, Kerkhof and Kuiper (2008) also shows that individuals in a positive

manipulation condition, in combination with anecdotal evidence, will have a more positive attitude toward a charitable organization and a higher intention to donate money to the organization than individuals in a negative manipulation condition, in combination with anecdotal evidence. The aforementioned results and the observation that Greenpeace uses

(11)

small quotes rather than statistical evidence lead to the hypothesis that a positive frame in a Greenpeace advertisement would be more effective than a negative frame.

H1: Individuals who are exposed to a positive frame in a Greenpeace advertisement have a higher intention to donate money, a higher intention to sign their petitions, and a more positive attitude toward Greenpeace than individuals who are exposed to a negative frame.

However, the previously mentioned Elaboration Likelihood Model and research by Smith (1996) showed that involvement may influence the hypothesized relationship. Even though it was not mentioned explicitly, all the participants in the research by Smith (1996) were highly involved with the issue. They had either owned a video camera or were considering the purchase. The results showed that the (highly involved) participants in the positive manipulation condition had a higher purchase likelihood and a more positive attitude toward the ad than the participants in the negative manipulation condition.

An experiment by Donovan and Jalleh (1999) showed different results. Their results showed that for the negatively framed product label, there was a difference between the participants with a high involvement and a low involvement. Only in the negative

manipulation condition, the participants with a low involvement had a more favorable attitude toward the product and a higher purchase intention than the participants with a high

involvement.

In contrast with the other researches, Maheswaran and Meyers-Levy (1990) showed that negative frames are more effective than positive frames in changing behavior when the individuals are highly involved with the issue. However, positive frames are more effective than negative frames in changing behavior when the individuals have a low involvement with the issue. This is in line with the research on the Elaboration Likelihood Model and the heuristic-systematic model. Advertisements are generally more effective when attention is paid toward them and when individuals process the information more thoroughly (Gordon, McKeage, & Fox, 1998). When the involvement of individuals with an issue is high, a relevant

(12)

message will be processed in more detail and the negative frame is more persuasive than the positive frame (Petty and Cacioppo, 1983; Maheswaran and Meyers-Levy, 1990). However, when the involvement of the individuals is low, the positive frame is more persuasive than the negative frame (Maheswaran and Meyers-Levy, 1990). When the involvement is low, individuals pay more attention to peripheral cues, such as colors, symbols or other affective elements (Buda & Zhang, 2000).

Based on these findings it is hypothesized that the involvement with Greenpeace will influence the hypothesized relationship between the positive and negative frames and the persuasiveness of a Greenpeace advertisement. Individuals who have a high involvement with Greenpeace will generally have a high intention to donate money, sign petitions, and have a positive attitude toward Greenpeace. However, the positive versus negative framing effect will be stronger for individuals who have a low involvement with Greenpeace.

H2: The effects of a positive frame are moderated by the involvement with

Greenpeace. Individuals with a low involvement with Greenpeace will show stronger effects of the positive frame.

The involvement with the environment is also hypothesized to have an influence on the hypothesized relationship between the positive and negative frames and the persuasiveness of a Greenpeace advertisement. Individuals who have a high involvement with the

environment will generally have a high intention to donate money, sign petitions, and have a positive attitude toward Greenpeace. However, the positive versus negative framing effect will be stronger for individuals who have a low involvement with the environment.

H3: The effects of a positive frame are moderated by the involvement with the environment. Individuals with a low involvement with the environment will show stronger effects of the positive frame.

(13)

Method

Participants and Design

For this study, a secondary school was approached to gather participants. Participants were exposed to an advertisement (fabricated by the author) of Greenpeace and had to answer questions after the exposure. It is a one-factor between-subjects design with framing (positive and negative) as the independent factor.

Procedure

After getting informed consent from their parents, the participants were told they would see an advertisement of Greenpeace. Afterwards, they completed a questionnaire with the measures described below. They also answered standard questions concerning their age, sex, and demographics.

Stimulus Material

Participants viewed one of two advertisements showing either a positive frame or a negative frame; both advertisements can be found in Figure 1 and Figure 2. In both cases, the message concerned a Greenpeace advertisement that focused on the same campaign. The campaign concerned whaling, which is a part of the Oceans campaign.

The photos in the advertisements were chosen as a result of a pre-test. Respondents were asked to view five photos and to lay them down in the order of most positive to the least positive. They also were asked to view four other photos which they had to lay down in the order of most negative to the least negative. In total, 17 respondents participated and indicated that the negative photo of the whale was by far the most negative photo. The photos with the positive frame were all generally viewed as being positive and the rating was more divergent. One of the two whale photos with the positive frame was rated as the

second most positive frame. Therefore, the two photos chosen for this research are the photos with whales.

(14)

With the design, special care was taken to make the advertisements look like actual Greenpeace outdoor advertisements. In the Netherlands, the outdoor advertisements of Greenpeace usually have a shocking photograph with a short sentence. The message with the positive frame concerned a photo with a free, jumping whale and a text that read: ”Red de prachtige walvissen”, which means “Save the beautiful whales”. The message with the negative frame concerned a photo with a bleeding, captured whale and a text that read: “Stop de afslachting van de walvissen”, which means “Stop the slaughter of the whales”.

Figure 1: Positive frame

(15)

Measures

Behavioral intentions. Four items assessed participants' behavioral intentions toward Greenpeace (on four 7-point bipolar response scales ranging from 1 = “very unlikely” to 7 = “very likely”. Participants choosing the middle point, number 4, are not unlikely nor likely to perform the described behavior). The donating behavioral intention was measured with the items intention to donate money and intention to donate money in the future. The other behavioral intention consisted of the intention to participate in a petition of Greenpeace and the intention to participate in a petition in the future.

Attitude toward the brand. The participants’ attitude toward Greenpeace was measured with a scale developed by Spears and Singh (2004). This scale consists of five items that are rated on 7-point bipolar response scales ranging from 1 = “not at all” to 7 = “extremely”. The five items are unappealing/appealing, bad/good, unpleasant/pleasant, unfavorable/favorable, unlikable/likable. However, the item unfavorable/favorable did not have an appropriate translation in Dutch. Therefore, this item was deleted in the questionnaire.

Mood effect. The PANAS developed by Watson, Clark, and Tellegen (1988) was used to measure the mood of the participants. This is an important variable, as the results of the study should be assigned to the framing of the advertisements, not to the mood the

participants were in. The PANAS involves 20 items that describe emotions and feelings. The items are rated on 5-point bipolar response scales ranging from 1 = “very slightly or not at all” to 5 = “extremely”. The 10 positive items are added up and the 10 negative items are added up. The total mood can be measured by weighting both the positive and the negative items. For the purposes of the experiment, two items were deleted. The item proud was deleted, because it is not a relevant emotion in the context of the subject of the stimulus material and the questionnaire. The item distressed was deleted, because there is not a good translation for it in the context of the subject of the stimulus material and the questionnaire. This leaves the questionnaire with nine positive items (interested, excited, strong, enthusiastic, alert, inspired, determined, attentive, and active) and nine negative items (upset, guilty, scared, hostile, irritable, ashamed, nervous, jittery, and afraid).

(16)

Involvement with Greenpeace and involvement with the environment. Both types of involvement were measured with a single item asking the participants whether they feel involved with Greenpeace and with the environment on two 7-point bipolar response scales ranging from 1 = “strongly disagree” to 7 = “strongly agree”.

Manipulation check. The participants’ perceptions on the positivity or the negativity of the advertisement was measured with two items. Both the perceptions on the photo as well as the text were measured on two 7-point bipolar response scales ranging from 1 = “very negative” to 7 = “very positive”.

Results

The study had 51 participants (51% female, ages 15-17, M = 15.94 years, SD = 0.65) from two HAVO 4 classes from the Rijnlands Lyceum Sassenheim. The participants in one classroom were exposed to the positive frame (N = 25) and the participants in the other classroom were exposed to the negative frame (N = 26).

Preliminary Operations and Analyses

The 18 mood effect items were recoded into the variable mood according to the PANAS instructions. The nine positive items were added up and divided by nine to form the positive mood. The nine negative items were added up and divided by nine to form the negative mood. To measure the mood, the negative mood was subtracted from the positive mood.

Reliability analyses were performed to see whether multiple items could load into one factor. The items intention to donate money (M = 2.75; SD = 1.44) and intention to donate money in the future (M = 3.88; SD = 1.83) have a reasonably strong reliability (α = .78). Both items were recoded into the item intention to donate money. The items intention to

participate in a petition (M = 4.45; SD = 1.92) and the intention to participate in a petition in the future (M = 4.80; SD = 1.87) have a strong reliability (α = .95) and were recoded into the

(17)

item intention to sign a petition. The attitude items unappealing/appealing (M = 4.41; SD = 1.28), bad/good (M = 5.71; SD = 1.08), unpleasant/pleasant (M = 5.00; SD = 1.17), and unlikable/likable (M = 4.98; SD = 1.26) have a strong reliability (α = .81) and were recoded into the item attitude.

To measure and exclude the correlations between mood and the dependent variables, the median (0.78) was calculated to divide the mood in both a positive and a negative mood. The first group were all participants who had a mood of 0.78 or more (N = 25). These

participants had a positive mood. The second group were all other participants (N = 26). They had a negative mood. However, the continuous variable mood did not correlate significantly with the intention to donate money (r = 0.16; p = .127), the intention to sign a petition (r = -0.05; p = .366), or the attitude toward Greenpeace (r = 0.08; p = .284).

To make a division between low involvement and high involvement with Greenpeace in relation to the dependent variables, the median (4.00) was calculated. The participants with an involvement of 4 or less (N = 29) were considered to be less involved with Greenpeace. The participants with an involvement of more than 4 (N = 22) were considered to be more involved with Greenpeace.

To make a division between low involvement and high involvement with the

environment in relation to the dependent variables, the median (5.00) was calculated. The participants with an involvement of 5 or less (N = 30) were considered to be less involved with the environment. The participants with an involvement of more than 5 (N = 21) were considered to be more involved with the environment.

Hypothesis 1

The first hypothesis was analyzed with three t-tests. Hypothesis 1 claimed that individuals who are exposed to a positive frame in a Greenpeace advertisement have a higher intention to donate money, a higher intention to sign their petitions, and a more positive attitude toward Greenpeace than individuals who are exposed to a negative frame.

(18)

The first expectation of hypothesis 1 was that the participants in the positive

manipulation condition would have a higher intention to donate money than the participants in the negative manipulation condition. This expectation did come true. The participants in the positive manipulation condition (M = 3.42; SD = 1.40; N = 25) showed a higher intention to donate money than the participants in the negative manipulation condition (M = 3.21; SD = 1.59; N = 26). However, the difference between the positive and the negative manipulation condition was not significant: t(49) = .50; p = .622; 95% CI [-0.64, 1.05]. Therefore, it cannot be excluded that the results can be ascribed to chance.

The second expectation is that the participants in the positive manipulation condition would have a higher intention to sign a petition of Greenpeace than the participants in the negative manipulation condition. This expectation did not come true. The participants in the negative manipulation condition (M = 4.94; SD = 1.73; N = 26) showed a higher intention to sign a petition than the participants in the positive manipulation condition (M = 4.30; SD = 1.95; N = 25). However, the difference between the positive and the negative manipulation condition was not significant: t(49) = -1.25; p = .218; 95% CI[-1.68, 0.39]. Therefore, it cannot be excluded that the results can be ascribed to chance.

Another expectation of hypothesis 1 is that the participants in the positive manipulation condition would have a more positive attitude toward Greenpeace than the participants in the negative manipulation condition. This expectation did not come true. The participants in the positive manipulation condition (M = 5.04; SD = 0.89; N = 25) had approximately the same attitude toward Greenpeace as the participants in the negative manipulation condition (M = 5.01; SD = 1.04; N = 26). The difference between the positive and the negative manipulation condition was not significant: t(49) = .11; p = .911; 95% CI [-0.51, 0.58]. Therefore, it cannot be excluded that the results can be ascribed to chance.

Hypothesis 1 claimed that the participants in the positive manipulation condition would show a higher intention to donate money, to sign a petition, and a more positive attitude toward Greenpeace than the participants in the negative manipulation condition. Even though the participants in the positive manipulation condition did show a higher intention to

(19)

donate money, they had a lower intention to sign a petition and the same attitude toward Greenpeace as the participants in the negative manipulation condition. The results of the t-tests were also not significant. Therefore, hypothesis 1 can be rejected.

Hypothesis 2

The second hypothesis was analyzed with three ANCOVAs with the covariate mood. Even though the continuous variable mood did not show significant correlations with the dependent variables, I did choose to control for the mood effects in all ANCOVAs. In none of the ANCOVAs mood had a significant effect as a covariate. Therefore, the effects of the covariate will only be mentioned at the first ANCOVA.

Hypothesis 2 claimed that the effects of a positive frame are moderated by the

involvement with Greenpeace. Individuals with a low involvement with Greenpeace will show stronger effects of the positive frame.

The first expectation of the second hypothesis was that participants who were less involved with Greenpeace would show an increased difference between the positive and the negative manipulation condition in respect to the intention to donate money than the

participants who were more involved with Greenpeace. The expectation did not come true. As can be seen in Table 1, the differences between the positive and the negative

manipulation condition were approximately the same size in both involvement groups. The moderation of involvement with Greenpeace on the hypothesized relation between

manipulation and the intention to donate money was not significant (F (1, 46) = 0.09, p = .772). As previously mentioned, the effects of the covariate mood were also not significant (F (1, 46) = 1.31, p = .259).

(20)

Table 1:

Means of the Manipulation and the Involvement with Greenpeace on the Intention to Donate Money

Less involved More involved

Positive manipulation 3.13 4.33

Negative manipulation 2.50 3.66

A second expectation was that the participants who were less involved with Greenpeace would show an increased difference between the positive and the negative manipulation condition in respect to the intention to sign a petition than the participants who were more involved with Greenpeace. The expectation did not come true. As can be seen in Table 2, the differences between positive and the negative manipulation condition were approximately the same size in both involvement groups. The moderation of involvement with Greenpeace on the hypothesized relation between manipulation and the intention to sign a petition was not significant (F (1, 46) = 0.06, p = .804).

Table 2:

Means of the Manipulation and the Involvement with Greenpeace on the Intention to Sign a Petition

A third expectation was that the participants who were less involved with Greenpeace would show an increased difference between the positive and the negative manipulation condition in respect to the attitude toward Greenpeace than the participants who were more involved with Greenpeace. As can be seen in Table 3, this expectation did come true. Even though the difference between the positive manipulation condition and the negative

(21)

Greenpeace than for the participants who were more involved, the moderation of

involvement with Greenpeace on the hypothesized relation between manipulation and the attitude toward Greenpeace was not significant (F (1, 46) = 0.25, p = .617). Therefore, it cannot be excluded that the results can be ascribed to chance.

Table 3:

Means of the Manipulation and the Involvement with Greenpeace on the Attitude

According to the second hypothesis, the less involved participants would show an increased difference of the dependent variables between the positive and the negative manipulation condition than the participants who were more involved with Greenpeace. This expectation did only come true for the dependent variable attitude toward Greenpeace, but the moderation was not significant. Therefore, hypothesis 2 can also be rejected.

Even though there were no significant results, there were some interesting trends that ask for further investigation. For both the less involved and the more involved participants, the participants in the positive manipulation condition showed a non-significantly higher intention to donate money (F (1, 46) = 0.69, p = .409, M positive = 3.42, SD = 1.40; M negative =

3.21, SD = 1.59) and a non-significantly more positive attitude toward Greenpeace (F (1, 46) = 1.41, p = .242, M positive = 5.04, SD = 0.89; M negative = 5.01, SD = 1.04) than the

participants in the negative manipulation condition. In contradiction with the expectations, the participants in the negative manipulation condition showed a non-significantly higher

intention to sign a petition than the participants in the positive manipulation condition (F (1, 46) = 0.93, p = .339, M positive = 4.30, SD = 1.95; M negative = 4.94, SD = 1.73).

Another interesting trend is that the participants who were more involved had a significantly higher intention to donate money (F (1, 46) = 8.02, p = .007, M more involved =

(22)

3.84, SD = 1.43; M less involved = 2.91, SD = 1.43) and a significantly more positive attitude

toward Greenpeace (F (1, 46) = 15.53, p < .001, M more involved = 5.53, SD = 0.79; M less involved

= 4.64, SD = 0.90) than the participants who were less involved with Greenpeace.

Hypothesis 3

The third hypothesis was analyzed with three ANCOVAs with the covariate mood. Hypothesis 3 claimed that the effects of a positive frame are moderated by the involvement with the environment. Individuals with a low involvement with the environment will show stronger effects of the positive frame.

The first expectation of the third hypothesis was that the participants who were less involved with the environment would show an increased difference between the positive and the negative manipulation condition in respect to the intention to donate money than the participants who were more involved with the environment. The expectation did not come true. As can be seen in Table 4, the differences between positive and the negative

manipulation condition were larger for the participants who were more involved than for the participants who were less involved with the environment. The moderation of involvement with the environment on the hypothesized relation between manipulation and the intention to donate money was not significant (F (1, 46) = 1.85, p = .181). Therefore, it cannot be

excluded that the results can be ascribed to chance.

Table 4:

Means of the Manipulation and the Involvement with the Environment on the Intention to Donate Money

(23)

A second expectation was that the participants who were less involved with the environment would show an increased difference between the positive and the negative manipulation condition in respect to the intention to sign a petition than the participants who were more involved with the environment. As can be seen in Table 5, the expectation did not come true. Contradictory to the expectations, the difference between the positive and the negative manipulation condition was larger for the participants who were more involved than for the participants who were less involved with the environment. The moderation of

involvement with the environment on the hypothesized relation between manipulation and the intention to sign a petition was not significant (F (1, 46) = 1.56, p = .219). Therefore, it cannot be excluded that the results can be ascribed to chance.

Table 5:

Means of the Manipulation and the Involvement with the Environment on the Intention to Sign a Petition

A third expectation was that the participants who were less involved with the environment would show an increased difference between the positive and the negative manipulation condition in respect to the attitude toward Greenpeace than the participants who were more involved with the environment. As can be seen in Table 6, the expectation did not come true. Contradictory to the expectations, the difference between the positive and the negative manipulation condition was larger for the participants who were more involved than for the participants who were less involved with the environment. However, the

moderation of involvement with the environment on the hypothesized relation between manipulation and the attitude toward Greenpeace was not significant (F (1, 46) = 0.55, p = .461). Therefore, it cannot be excluded that the results can be ascribed to chance.

(24)

Table 6:

Means of the Manipulation and the Involvement with the Environment on the Attitude

According to the third hypothesis, the less involved participants would show an increased difference of the dependent variables between the positive and the negative manipulation condition than the participants who were more involved with the environment. This expectation did not come true. For the dependent variables intention to sign a petition and attitude toward Greenpeace, the more involved participants even had a non-significantly increased difference between the positive and the negative manipulation condition than the participants who were less involved in the environment. Therefore, hypothesis 3 can be rejected.

Even though there were no significant results, there were some interesting trends that ask for further investigation. Similar to the trends mentioned in Hypothesis 2, involvement with the environment showed interesting but non-significant trends. The more involved participants had a non-significantly higher intention to donate money (F (1, 46) = 2.70, p = .107, M more involved = 3.52, SD = 1.58; M less involved = 3.17, SD = 1.43), a non-significantly

higher intention to sign a petition (F (1, 46) = 0.11, p = .744, M more involved = 4.74, SD = 1.95;

M less involved = 4.55, SD = 1.81), and a non-significantly more positive attitude toward

Greenpeace (F (1, 46) = 0.45, p = .507, M more involved = 5.10, SD = 1.12; M less involved = 4.98,

SD = 0.85) than the participants who were less involved with the environment. Contrary to the expectations but similar to the trends mentioned in Hypothesis 2, the participants in the negative manipulation condition had a non-significantly higher intention to sign a petition than the participants in the positive manipulation condition (F (1, 46) = 2.28, p = .138, M positive =

(25)

Explorative Analyses

The findings of Smith (1996) were also replicated. Smith (1996) found that the highly involved participants in his experiment had a higher purchase intention and attitude toward the ad in the positive manipulation condition than in the negative manipulation condition. The results of this research show that the participants with a high involvement with Greenpeace have a higher intention to donate money (F (1, 46) = 8.02, p = .007, M more involved = 3.84, SD

= 1.43; M less involved = 2.91, SD = 1.43) and a more positive attitude toward Greenpeace (F

(1, 46) = 15.53, p < .001, M more involved = 5.53, SD = 0.79; M less involved = 4.64, SD = 0.90)

than the participants with a low involvement with Greenpeace. As can be seen from Table 1 and Table 3, the participants with a high involvement with Greenpeace also had a (non-significantly) higher intention to donate money and a (non-(non-significantly) more positive attitude toward Greenpeace in the positive manipulation condition than in the negative manipulation condition. This, however, is in contrast with the Elaboration Likelihood Model where the negative frame is more persuasive than the positive frame when an individual is highly involved with the issue.

Conclusion and Discussion

The aim of this study was to discover whether Greenpeace should design

advertisements with a positive frame rather than a negative frame. According to the research on positive versus negative framing effects, one may expect that a positively framed

advertisement would be more effective than a negatively framed advertisement in terms of behavioral intentions and attitude toward the brand.

The hypothesized relationships between framing and the persuasiveness of Greenpeace advertisements were not confirmed in this research. The participants in the positive manipulation condition did not have a significantly higher intention to donate money, to sign a petition, and did not have a more positive attitude toward Greenpeace than the

(26)

participants in the negative manipulation condition. Therefore, the conclusion to the research question To what extent does the use of a positive or a negative frame in an advertisement of Greenpeace influence the intention to donate money, the intention to sign petitions and the attitude toward Greenpeace? is that there is no significant difference between the influences of the positive and negative frame on the intention to donate money, the intention to sign petitions and the attitude toward Greenpeace. The results of this research did not show a preference for using either the positive or negative frame in Greenpeace advertisements. The involvement with both Greenpeace and the environment did also not significantly moderate the proposed relationship between framing and the dependent variables.

A limitation and an explanation for the results that were not significant is that the manipulation was not successful. The participants indicated that they believed the text in the negative manipulation condition to be not positive nor negative. A t-test showed that the text in the negative manipulation condition was rated a 4.23 (M = 4.23; SD = 1.58; N = 26) on a scale from 1 (negative) to 7 (positive). A manipulation with a more negative text may have led to stronger or even significant results.

A second limitation of the experiment is that the participants in the positive

manipulation condition were less focused than the participants in the negative manipulation condition. The students in the positive manipulation condition just had a lunch break and were very energetic, while the students in the negative manipulation condition were already focused. The students in the negative manipulation condition may, thus, have been more focused on and involved with the experiment. This brings us back to the heuristic-systematic model (Chaiken, 1987) where positively framed messages are more persuasive when the involvement is low and negatively framed messages are more persuasive when the

involvement is high. If the students in the negative manipulation condition indeed had a high involvement with the experiment and the less focused students in the positive manipulation condition had a low involvement with the experiment, the lack of significant results is not surprising.

(27)

A third limitation is that the sample in the experiment was really small. There were only 25 participants in the positive manipulation condition and 26 participants in the negative manipulation condition. Even though a sample with 51 participants is large enough for finding significant results, it is a too small sample to draw big conclusions. Even if there were

significant results, you cannot conclude that the results will be representative for the entire population. For this research it is not possible to draw any reliable and generalizable conclusions.

A fourth limitation is that the variables mood, involvement with Greenpeace, and involvement with the environment were recoded to make a division between a positive and a negative mood and lowly and highly involved participants. To make a division, the median was used. However, this may not be a very reliable division. The group of individuals who were actually highly involved may have been smaller than was calculated with the median.

A fifth limitation is that the participants were already familiar with the brand Greenpeace and had already formed their opinions about it before the experiment took place. Changing an already existing opinion or attitude is quite difficult to achieve with the exposure to only one advertisement. This may explain why there were only small differences in results between the two groups of participants.

Although the results were not significant, the findings do show that the issue should be further investigated. The results show interesting trends that indicate that the intention to donate money and the attitude towards Greenpeace may be higher when a positive frame is being used in a Greenpeace advertisement than when a negative frame is being used. When the focus is on gathering as much signatures for petitions as possible, it may be better to use a negative frame in an advertisement. Future research should investigate this and make the manipulation stronger and successful. Not only the photos should be pre-tested but also the accompanying texts. The pre-tests should lead to very strong positive and negative

manipulations, while also being perceived as credible (Greenpeace) advertisements. Future research should also strictly monitor that all the conditions in the experiment are exactly the same in both groups and that they are equally focused and involved with the experiment.

(28)

Even though the findings of this research were not significant, they do relate to the previous findings on positive versus negative framing effects. Previous studies showed that individuals who read a description or advertisement with a positive frame had more positive associations, higher purchase intentions, and a better attitude toward the advertisement than individuals who read a description or advertisement with a negative frame (Levin, 1987; Levin et al., 1986; Buda & Zhang, 2000; Smith, 1996).

Even though the results were not significant and further research may contradict the findings in this study, tentative recommendations are given. When further research does replicate the findings of this study with significant results, the advice for the Communication department of Greenpeace is to design an advertisement with a positive frame when the goal is to gather donations or to increase the general attitude toward Greenpeace. The photo should generate positive feelings. This may entail a photo that shows a beautiful natural scene or even a wild animal that lives serenely in its natural habitat. The text should use words that are perceived to be positive, such as “save”, “clean” or “victory”. When the goal is to gather signatures for the petitions, the advertisement should have a negative frame. The photo should generate negative feelings by, for example, showing destructions, oil spills, or extinct animals. The text should use words that are perceived to be negative, such as “stop”, “destruction” or “extinct”. When future research does find significant results for the

hypothesized relationship between positive versus negative framing and the persuasiveness of (Greenpeace) advertisements, these tentative recommendations may hopefully increase the persuasiveness of the advertisements and ultimately lead to a healthy planet.

References

Buda, R., & Zhang, Y. (2000). Consumer product evaluation: The interactive effect of message framing, presentation order, and source credibility. Journal of Product & Brand Management, 9(4), 229-242.

(29)

Chaiken, S. (1987) The heuristic model of persuasion. In M. P. Zanna, J. M. Olson & C. P. Herman (Eds.), Social influence: The Ontario symposium (Vol. 5, pp. 3-39), Hillsdale, NJ: Erlbaum.

Chaiken, S., Liberman, A., & Eagly, A. H. (1989). Heuristic and systematic information processing within and beyond the persuasion context. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 212_252). New York: Guilford Press.

Das, E., Kerkhof, P., & Kuiper, J. (2008). Improving the effectiveness of fundraising

messages: The impact of charity goal attainment, message framing, and evidence on persuasion. Journal of Applied Communication Research, 36(2), 161-175. DOI: 10.1080/00909880801922854.

Davis, J. J. (1995). The effects of message framing on response to environmental

communications. Journalism and Mass Communication Quarterly, 72(2), 285-299. Ditto, P. H., & Lopez, D. F. (1992). Motivated skepticism: Use of differential decision criteria

for preferred and nonpreferred conclusions. Journal of Personality and Social Psychology, 63(4), 568-584.

Donovan, R. J., & Jalleh, G. (1999). Positively versus negatively framed product attributes: The influence of involvement. Psychology & Marketing, 16(7), 613-630.

Gamson, W. A., & Modigliani, A. (1987). The changing culture of affirmative action. In R. D. Braungart (Ed.), Research in political sociology (Vol. 3, pp. 137–177). Greenwich, CT: JAI.

Gordon, M. E., McKeage, K., & Fox, M. A. (1998). Relationship marketing effectiveness: The role of involvement. Psychology and Marketing, 15(5), 443-459.

Hay, C. (2011, January 22). Last chance to see? Bearing witness. Retrieved April 4, 2014 from http://www.greenpeace.org/international/en/news/Blogs/makingwaves/last-chance-to-see-bearing-witness/blog/32517/.

Homer, P. M., & Yoon, S.-G. (1992). Message framing and interrelationships among ad-based feelings, affect, and cognition. Journal of Advertising, 21, 19-33

(30)

Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263-292.

Krishnamurthy, P., Carter, P., & Blair, E. (2001). Attribute framing and goal framing effects in health decisions. Organizational Behavior and Human Decision Processes, 85(2), 382-399. DOI: 10.1006/obhd.2001.2962.

Leidsch Dagblad (2014, February 26). Greenpeace met ‘tijgervrije’ shampoo op WNF-toer. p. 3.

Levin, I. P. (1987). Associative effects of information framing. Bulletin of the Psychonomics Society, 25, 85-86.

Levin, I. P, & Gaeth, G. J. (1988). How consumers are affected by the framing of attribute information before and after consuming the product. Journal of Consumer Research, 15, 374-378.

Levin, I. P., Johnson, R. D., Deldin, P. J., Carstens, L. M., Cressey, L. J., & Davis, C. R. (1986). Framing effects in decisions with completely and incompletely described alternatives. Organizational Behavior and Human Decision Processes, 38, 48-64. Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All frames are not created equal: A

typology and critical analysis of framing effects. Organizational Behavior and Human Decision Processes, 76(2), 149–188.

Maheswaran, D., & Meyers-Levy, J. (2004). Exploring message framing outcomes when systematic, heuristic, or both types of processing occur. Journal of Consumer Psychology, 14(1), 159-167.

Maheswaran, D., & Meyers-Levy, J. (1990). The influence of message framing and issue involvement. Journal of Marketing Research, 27(3),361-367.

Meyerowitz, B. E., & Chaiken, S. (1987). The effect of message framing on breast self-examination attitudes, intentions and behavior. Journal of Personality and Social Psychology, 52, 500-510.

(31)

Petty, R. E., & Cacioppo, J. T. (1983). Central and peripheral routes to persuasion: Application to advertising. In: L. Percy, & A. Woodside (Eds.), Advertising and consumer psychology (pp. 3-23). Lexington, MA: Lexington Books.

Petty, R. E., & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral routes to attitude change. New York: Academic Press.

Petty, R. E., Gleicher, F., & Baker, S. M. (1991). Multiple roles for affect in persuasion. In J. P. Forgas (Ed.), Emotion & social judgments (pp. 181-200). Oxford: Pergamon Press. Pratto, F., & John, O. P. (1991). Automatic vigilance: The attention-grabbing power of

negative social information. Journal of Personality & Social Psychology, 61(3), 380-391.

Rathzel, N., & Uzzell, D., (2009). Changing relations in global environmental change. Global Environmental Change, 19, 326–335.

Rothman, A. J., Salovey, P., Antone, C., Keough, K., & Martin, C. D. (1993). The influence of message framing on intentions to perform health behaviors. Journal of Experimental Social Psychology, 29, 408-433.

Smith, G. E. (1996). Framing in advertising and the moderating impact of consumer education. Journal of Advertising Research, 36(5), 49-64.

Smith, S. M., & Petty, R. E. (1996). Message framing and persuasion: A message processing analysis. Personality and Social Psychology Bulletin, 22(3), 257-268. DOI:

10.1177/0146167296223004.

Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase intentions. Journal of Current Issues & Research in Advertising, 26(2), 53-66, DOI: 10.1080/10641734.2004.10505164.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief

measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063-1070.