The gains and losses of Corporate Social Responsibility : how the gain/loss framing of CSR messages affect consumer evaluations

(1)

The gains and losses of Corporate

Social Responsibility

How the gain/loss framing of CSR messages affect consumer

evaluations

Master’s thesis

Master’s Track Persuasive Communication University of Amsterdam

Graduate School of Communication Thesis supervisor: Lotte Salome 27-06-2014

Studentnumber: 6178200 Raoul van Tol

Kudelstaartseweg 235 1433 GG Kudelstaart raoulvantol@gmail.com

(2)

1

ABSTRACT

For decades companies have been communicating their corporate social responsible (CSR) activities via various media to their consumers. Companies in this period mostly used gain-framed messages to promote their CSR activities. This study explores if the framing of CSR messages (gain/loss) has any effect on consumer behavior (C-C identification, reputation and purchase intention) and if there is a single effective method to communicate CSR activities to consumers. It contributes to existing literature by explaining the use of framing from corporation’s to present their environmental responsibility. The use of framing in messages plays an important role in shaping social issues and subsequent judgments in dealing with them. The research focuses on companies that were portrayed as positive or negative in the media and tried to find whether a gain-framed message or a loss-framed message will have more effect in these conditions. The expectations were that gain-framed messages would have a more positive effect on C-C identification and reputation in a negative news condition (risk averse) and that loss-framed messages would have a more positive effect on C-C identification and reputation in a negative news condition (risk seeking). This would indirectly influence the purchase intention from the consumer. The study (N=170) found no significant evidence that the way of framing has any effect on consumer behavior. The method of framing did have an indirect effect on purchase intention via C-C identification. The study did not find an indirect effect on purchase intention via reputation.

(3)

2

INTRODUCTION

The worldwide discussion about the climate change invites people to think about green consumerism. This discussion made an increasing number of companies become more aware of the impact of corporate social responsibility and the contributions to the environment (Sen & Bhattacharya, 2001). The main thought of corporate social responsibility (CSR) is that company activities serve a social good, and not the company’s needs. Therefore managers implement these activities into the core strategies of a company (Cone, 2013). These initiatives can have many advantages for companies. For instance, 96% of the people have a more positive image of a company that has CSR activities, 94% are more likely to trust companies that participates in CSR activities and 93% would be more loyal to a company when companies support social or environmental issues (Cone, 2013). These benefits make entrepreneurs aware of the necessity of CSR activities, and encourage them to take more initiatives to communicate these activities to the outside world (Cramer, Jonk & Heijen, 2004). Therefore, the amount of CSR advertisements increased substantially in the past 30 years (Bortree, Ahern, Smith & Dou, 2013). The way how these advertisements are framed is an important for the evaluation of the messages by the consumers. Bortree et al. (2013) found that organizations have a tendency to use gain-framed messages in CSR communications. These gain-framed messages are messages that emphasize the potential gains of a certain behavior. Opposite to these messages are loss-framed messages that emphasize the potential losses of a certain behavior. Bortree et al. (2013) suggest that it would be good to encourage the use of loss-framing to let people think about environmental issues. Interestingly little research has been done to the effect of gain/loss framing on CSR messages, while the consequences of CSR behavior are an important component of these messages. Therefore framing looks perfectly applicable for CSR messages, because framing plays an important role in shaping social issues and subsequent judgments in dealing with them (Gergen, 1992). This study will examine if the differences between gain/loss framing in the context of CSR messages, have effects on the reputation of a company, the consumer-company identification and which effect this has on the purchase intention of the consumer.

Prior scientific literature examined the effect of CSR messages on consumer-company identification, reputation and purchase intention. Despite this there is still a lot we do not know about which influence framing has on this relation. These previous studies explored the impact of message framing on consumer behavior (Detweiler, Bedell, Salovey, Pronin & Rothman, 1999; Rothman, Salovey, Antone, Keough & Martin, 1993). A leading theory about

(5)

4 message framing is the prospect theory from Kahneman and Tversky (1979). This theory explains how people’s decisions are sensitive to the way a message is framed. People will avoid risks when considering potential gains are afforded by a decision, but are willing to take risks when considering the potential losses afforded by their decision. This method of framing is mostly applied in health communication (Detweiler, Bedell, Salovey, Pronin & Rothman, 1999; Rothman, Salovey, Antone, Keough & Martin, 1993). Framing is also implemented in relation with CSR activities. Wang (2007) found that priming and framing may work together to achieve a higher level of persuasion, because priming raised the awareness of the people.

It is important to communicate a company’s CSR activities on an effective method because it influences consumer behavior. Underlying processes of this effect is that CSR initiatives generate higher company evaluations by the consumers. Important determinants of company evaluations are consumer-company identification and a company’s reputation. This is because individuals who identify themselves in the behavior of an organization incorporate these beneficial facets to their own identity (Marin, Ruiz & Rubio, 2009; Sen & Bhattecharya, 2001). A higher consumer-company identification has a positive effect on the reputation of a firm when it engages in CSR activities (Bhattecharya & Sen, 2004; Brammer & Pavelin, 2006; Hsu, 2012). Brown and Dacin (1997) suggest that CSR has an indirect effect on purchase intention via a consumer’s evaluation of a company.

With few studies about the relation between framing and CSR there is still a lot to explore in this topic of research. There is still little research available about the prospect theory and its effects on the framing of CSR messages. This research provides more information about how the framing of CSR messages can help companies communicate their business activities to their audience. This study is the start for the creation of a framework that gives companies suggestions how they can communicate their CSR messages in the right way. This will help managers to create effective communication strategies for their businesses, what would lead to a better company evaluation and higher profits. Therefore the research question of this study is:

RQ: To what extend does the gain/loss framing of a CSR message in a positive/negative news condition influence a consumers’ company evaluation and does this relation affects the purchase intention of a consumer?

(6)

5

THEORETICAL FRAMEWORK

If you want to convey your CSR message effectively to your consumers, it is necessary to know on what effect framing has on these messages. The study of Bortree, Ahern, Smith and Dou (2013) examined how corporations have framed their CSR messages the past 30 years. By coding articles from the National Geographic magazine they have found trends about the framing of CSR messages. They found an increase in the amount of CSR advertisements over the past 30 years and besides an increase of messages, the study also found that organizations have a tendency to use gain frames in their communication about CSR. The authors suggest that the use of gain-framing would create a positive perception of CSR and would minimize the awareness of environmental issues. Besides that they also suggest that it would be right to use more loss-framed messages so that people would be encouraged to think more about environmental issues. This approach would have its effect because people will hold companies responsible for their actions towards the environment. While companies recognize the importance to communicate their CSR initiatives effectively, scientific research has been slow to provide theory-based guidance on how firms should achieve this. In this study we will search empirical evidence for these statements from Bortree, Ahern, Smith and Dou (2013). Furthermore we will look for differences in consumer responses between gain-framed and loss-framed messages in CSR communication.

CORPORATE SOCIAL RESPONSIBILITY

Throughout the years CSR has been defined several times. The problem in this field of research is that the term CSR can have different meanings. In 1972 Votaw wrote: “corporate social responsibility means something, but not always the same thing to everybody” (Votaw, 1972, p. 25). In this definition he describes the different ways how people interpret CSR. He names the idea of legal responsibility, social responsibility behavior in ethical sense, a charitable contribution and that a company is socially conscious. The complexity of this definition gives CSR a wide range of meaning. Therefore the following definition is used to describe the term CSR in this study: ‘CSR associations reflect the organization’s status and activities with respect to its perceived societal obligations’ (Brown & Dacin, 1997, p.68). Because an increasing number of companies implement CSR initiatives, it is important for them to find a method to communicate these initiatives effectively to their customers. Without good communication about their CSR initiatives people would not be aware of the fact that a company is involved with their local community (Pomering & Dolcinar, 2009). A range of

(7)

6 studies found that consumer awareness of CSR initiatives is low, and that brings great opportunities for companies to increase the consumer’s awareness by using communication (Sen, Bhattacharya & Korschun, 2006; Pomering & Dolcinar, 2009). Therefore this study will explore which effect the framing of CSR messages has on consumer behavior.

FRAMING

An effective method to influence people is framing. Framing is used to influence the receivers’ attitude and behavior towards the topic of a message. A message could influence the decision of the receiver and is controlled by the formulation of the problem and the habits, norms and personal characteristics of the decision-maker (Tversky & Kahneman, 1981). The information in a message can be presented in various ways. Prior research has shown that not all individuals respond the same to an identical message. A message that is presented in a positive way (positive framing) could be interpreted in a different way as when the receiver sees a message that is presented in a negative form (negative framing). This choice in framing influences the attitude of the receiver and affects its behavior (Tversky & Kahneman, 1981). A regularly used expression as when a glass of water is labeled ‘half-full’ or ‘half-empty’ is an example of framing in everyday life (Martin & Marshall, 1999). In 1993 Entman researched the different functions of framing and used the following definition to explain framing: “To frame is to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote in particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (Entman, 1993, p.52). In his definition he distinguishes four functions of framing: the problem that is defined by the message, the cause of this problem, the moral evaluation of this problem and the solution of this problem. With this definition he states that when a person frames a message, he or she will make a piece of information more noticeable and meaningful to an audience. The various ways how people evaluate the problem declare the different interpretations to framed messages.

PROSPECT THEORY

A leading theory about framing is the ‘prospect theory’ from Kahneman and Tversky (1979). Throughout the years a range of studies used the prospect theory as a framework for their framing studies (Rothman, Bartels, Wlaschin & Salovey, 2006; Rothmann et al., 1993). The prospect theory explains how people’s preferences are sensitive to the way how

(8)

7 information is framed. For understanding this theory it is important to know how people make rational decisions. Kahneman and Tversky (1979) suggest that every person has its own reference point as it comes to health, prestige and wealth, per example; the level of wealth that a person has can imply poverty for the one person and great riches for another person. This depends on the reference point that has been taken. The current wealth of the person is the reference point, and this reference point differs for everybody. An important factor is the magnitude of change that occurs from that reference point. When the value decreases below the reference point it will be considered as a loss. A person will try to minimalize this loss by taking more risks in their behavior to make profits (risk seeking). When the value increases over the reference point it will be considered as a gain. A person always wants a potential gain for its reference point. According to the prospect theory people try to avoid risks when they take the potential gains into account that are produced by the behavior, these people are ‘risk averse’ (Kahneman & Tversky, 1979; Rothman & Salovey, 1997). Gain-framing accentuates the profits of the behavior and therefore people will give their preference to a safe decision. The people who consider potential losses that are produced by their behavior are ‘risk seeking’ in their preferences, because people have the feeling that they are in risk. It suggests that people are more up to take risks when they read a message where the potential losses are emphasized (Kahneman & Tversky, 1979). In their study they also mention the tendency of people to prefer avoiding losses instead of obtaining gains. This tendency is called ‘loss aversion’ and it leads to people choosing ‘risk aversion’ preferably above ‘risk seeking’. With the earlier example of wealth this would mean that a person that loses 10 euro from their reference point will lose more satisfaction as a person gains satisfaction from a 10 euro profit to their reference point. This suggests that people prefer avoiding losses to making gains (Kahneman & Tversky, 1979). Meyerowitz and Chaiken (1987) criticize the effects of the prospect theory and suggest that the prospect theory does not take the negativity bias into account when people process information at low risk. The negativity bias can be explained as: “When a positive attitude or positive affect does not have an effect on measured behavior oppositely equivalent to the effect of a negative attitude or negative affect” (Jordan, 1965, p. 315). This suggests that negative information has a stronger influence on decisions as positive information. Kahneman and Tversky (1979) already took the negativity bias into account with their ‘loss aversion’ statement. Meyerowitz and Chaiken (1987) also suppose that the prospect theory only focuses on behavior that contains risks and does not pay attention to the fact that some decisions may have no risk involved, what could lead to different results. This could be important for the use of gain/loss-framing in the field of CSR because the intended behavior

(9)

8 does not contain a high amount of risk. This study will explore if the prospect theory is applicable to CSR related behavior and if the gain/loss framing of CSR related messages will have any influence on consumer evaluations.

GOAL FRAMING

The studies that were conducted after the Kahneman and Tversky (1979) study differed in their outcomes (Bohm & Lind, 1992; Fagley & Miller, 1990; Highhouse & Yuce, 1996). Levin, Schneider and Gaeth (1998) thought that a closer examination of the different types of framing effects could reveal an explanation for these mixed results in framing research. Their study found three different types of framing manipulations: Risky choice framing, attribute framing and goal framing. Risky choice framing is the form of framing were the outcomes of a potential choice involving options with a different level of risk are described in several ways. Another form of framing is ‘attribute framing’ were characteristics of an object or event are centered in the message. The last type of framing is goal framing were the goal of an action and/or behavior is framed (Levin, Schneider & Gaeth, 1998). This study focuses on goal framing because the goal that is tried to communicate with CSR communication, are the advantages of CSR activities. Goal framing affects the impact of persuasion of a message and can be measured with a comparison of the rate of adoption of the intended behavior (Levin, Schneider & Gaeth, 1998). Hallahan (1999) explains the effects of goal framing or ‘the framing of actions’ how he calls this form of framing in his study. He describes that the framing of actions gives people the opportunity to accentuate the positive consequences to perform a certain behavior or to reinforce negative consequences of not taking a particular action. Both ways of framing promote the same behavior, but the formulation of the intended behavior in the message makes a difference for the persuasiveness of the message. Gain-framing focuses on reaching a particular profit by emphasizing the profits of a behavior, while with loss-framing the focus lays on reaching the same profit by emphasizing the disadvantages for when the desired behavior will not be adapted. Both ways of framing still have the same goals and try to influence people to perform a certain behavior.

HEALTH BEHAVIOR AND FRAMING

The prospect theory is used frequently in studies about health behavior (Cox & Cox, 2001; Meyerowitz & Chaiken, 1987; Schneider, Salovey, Apanovitch, Pizarro, Mccarthy, Zullo & Rothman, 2001). The theory fits the area of research because most sorts of health

(10)

9 behavior contain risky decision making. Rothman and Salovey (1997) presented that when people consider a risky behavior (detection behaviors) loss-framed messages would be more effective to promote a certain behavior. A range of studies support that loss-framed appeals are more effective than gain-framed appeals in promoting detection behaviors (Cox & Cox, 2001; Meyerowitz & Chaiken, 1987; Detweiler, Bedell, Salovey, Pronin & Rothman, 1999; Schneider et al., 2001). These studies found effects of loss-framed appeals on cancer screening behaviors from people. On the other side prevention behaviors help people to maintain their health and minimize the risk to get sick, for this reason gain-framed messages are predicted to have a higher effect on the persuasion of people. Prior research supported the assumption that gain-framed messages are more effective for people who want to perform prevention behaviors (Detweiler, Bedell, Salovey, Pronin & Rothman, 1999; Rothman et al., 1993). All these studies presented the applicability of the prospect theory to healthcare behavior, and show that the prospect theory could also be applicable in other areas of research.

NEWS AND FRAMING

Another area of research is the relation between framing and news coverage in the media. These studies made no distinct between risky choice framing, attribute framing and goal framing (Levin et al. 1998) like the health behavior studies mentioned earlier. The fact that news messages do not present any persuasive content and only cover facts could be an explanation for this. Nevertheless is there a high use of framing in these messages. ‘Valence framing’ is mostly used in news coverage’s, because the creators of the news (journalists, news agencies) add their own specific perception to an issue. Bizer and Petty (2005) found that negative framing in the news has more influence on the perceiver as when people are exposed to a positive framed news item. Iguarta and Cheng (2009) supported these results and measured the effects of positive and negative frames from news coverage about immigration. The outcomes of this study showed that negative framed news messages (the one that linked immigration with criminality) have a stronger effect on the attitude towards the news topic as the positive framed news messages did.

CSR AND FRAMING

While there has been research about the framing effects on health behavior and news coverage, there has been little research on the effects of framing on corporate social

(11)

10 responsibility (CSR). This absence of research in this area is remarkable, because framing plays an important role in shaping social issues and subsequent judgments in dealing with them (Gergen, 1992). Wang (2007) tested the effects of priming and framing on the judgment of a target corporation. In his study he exposed people to framed news messages and the results suggest that priming and framing may work together to achieve a higher level of persuasion. This is because priming raises the awareness of the participants. That is the reason why the effect of framing stayed out when the participants were not primed before the exposure to the news message. This suggests that people’s judgments may not be susceptible to framing when an issue is not activated by priming. As the Bortree et al. (2013) study found there is a high amount of gain-framed CSR messages in CSR advertising. More research could explain this preference for gain framing, and would explore more about the consumer reactions towards different ways of framing. The lack of research about gain/loss framing in the area of CSR is remarkable because the way of framing could be an important determinant for the intended behavior. This study explores whether message framing about CSR has any influence on the company evaluation, and if this has any influence on the intended purchase behavior or the consumer.

CSR AND C-C IDENTIFICATION

The consumer-company identification (C-C identification) is an important determinant for the purchase intention of a consumer. Identification satisfies the need for social identity and self-definition and positively influences a member’s organizational loyalty (Mael & Ashforth, 1992). Bhattacharya and Sen (2003) proposed that consumers identify themselves with the subset of company associations that constitutes the company’s identity. This identity is built up by two dimensions: The core values of a company and the company’s demographic characteristics. The company’s core values embody the operating principles, organizational mission and leadership of a company (Whetten & Godfrey, 1998), the demographic characteristics consist of the industry, the size, age, location and employees of the company (Pelled, Cummings & Kizilos, 2000). These company associations are often communicated by official documents as annual reports and press releases. A high level of congruence between the company and the individual evokes a high level of consumer-company identification. This leads to a more positive word-of-mouth about a company (Ahearne, Bhattecharya & Gruen, 2005), a higher brand reputation, more brand loyalty and a higher purchase intention (Marin, Ruiz & Rubio, 2009). Brown and Dacin (1997) support these results by suggesting that C-C

(12)

11 identification has a positive effect on the evaluation of a company. Consumers will be loyal to already existing products what leads to a higher willingness to try new products, and more resistance to negative information about the company in question (Bhattacharya & Sen, 2003). This means that if a consumer has a positive attitude towards CSR, it would see less negative information of a company that acts with CSR activities. CSR activities enhance the C-C identification, because of the positive image that is created by these activities (Marin et al., 2009; Sen & Bhattacharya, 2001). Therefore it is important that the CSR message that is communicated by a company is phrased effectively. This study explores if a certain way of framing (gain vs loss) would have a more positive influence on C-C identification, and if this effect would be different when the consumers were exposed to positive news or negative news about these CSR activities.

CSR AND REPUTATION

Companies recognize brand reputation as an important factor to achieve goals and to keep competitive with their rivals. Therefore companies use CSR in their corporate branding to raise the reputation of their company (Melo & Garrido, 2012). Fombrun (1996) defined reputation as: “a perceptual representation of a company’s past actions and future prospects that describe the firm’s overall appeal to all its key constituents when compared to other rivals” (Fombrun, 1996, p.72). Because reputation has to be build up during a longer time span, is it a valuable good for a company because it is hard to replicate or to create in a short amount of time (Drejer, 2000). The benefit of a positive reputation is the sustained competitive advantage that a company acquires. A method to raise this reputation for a company is to use CSR initiatives. Melo and Garrido (2012) found that CSR has a positive effect on reputation, and creates competitive advantages for a company. Outcomes of this competitive advantages are improved relationship with bankers and investors what has a positive effect on a company’s capital (Spicer, 1978), attracting better employees (Greening & Turban, 2000) and attracting new customers (Fombrun & Shanley, 1990). The study presented proof that CSR activities lead to a higher reputation of the company. People see their relationship with a company as a valuable good, and will try to preserve this relationship. Therefore this study will explore which kind of framing is the most effective for building a stronger reputation.

(13)

12 CSR AND PURCHASE INTENTION

A range of studies examined the direct effects of CSR activities on purchase intention (Kacperczky, 2009; Lee & Shin, 2009; Tian, Wang & Yang, 2011). These studies examined the relation between the exposure to a CSR message and the purchase intention of the consumer. The outcomes of these studies found that consumers have to be aware of a company’s CSR initiatives before this factor can influence their purchase decisions. This awareness led to positive financial performances for the companies. But awareness of CSR could also have a negative effect on the purchase intention of the consumer. Mohr, Webb and Harris (2001) present that irresponsible CSR behavior can also have a negative influence on the purchase intention of the consumer. Companies can use cause-related marketing to create this awareness at the consumers. Webb and Mohr (1998) found that when a concept was explained, 79 percent of the people could recognize a cause-related marketing campaign. The majority of the consumers does not make purchase decisions on CSR criteria but mostly use traditional criteria as price, quality, reputation, location and service. Besides that, CSR only influences consumer perceptions about companies, and not about products or services (Brown & Dacin, 1997). For this reason Brown and Dacin (1997) suggest that it only seems that CSR activities have an indirect effect on purchase intention. This effect has been proven by the study of Keh & Xie (2009) who found an indirect effect of CSR activities on the consumers purchase intention via the company evaluation. The reputation of a company is a part of a consumer’s company evaluation (Brammer & Pavelin, 2006). Past research found that reputation has a positive influence on the financial performance of a company (Fombrun, 1996; Roberts & Dowling, 2002). A reason for this positive effect is the amounts of quality were the company is associated with due to the positive reputation. Firms with a positive reputation are associated with high quality products, services and consumers are open to pay a price premium for these high quality features of a company (Shapiro, 1987). This is because a high reputation can strengthen people’s confidence and reduce risk perceptions of people to buy products from a company. Maden, Arikan, Telci and Kantur (2012) support these findings and found that the purchase intention of a consumer depends on the reputation of a company. Another determinant of a consumer’s company evaluation is the C-C identification. A high level of consumer-company identification leads to a more positive word-of-mouth about a company (Ahearne, Bhattecharya & Gruen, 2005), a better brand reputation, more brand loyalty and a higher purchase intention (Marin, Ruiz & Rubio, 2009). A company with a high C-C identification benefits through the consumer’s loyalty to their products, the

(14)

13 willingness to try new products and the resistance towards negative information about the company (Bhattacharya & Sen, 2003). These relations are supported by the study of Keh and Xie (2009) who found that C-C identification and the reputation of a company both have a positive effect on the purchase intentions.

HYPOTHESES

When people are confronted with positive news about the CSR activities of a company, they could consider it as a safe decision to buy a product from this company. Because the company handles in line with their CSR strategy, people will perceive the company as a trustworthy firm. Prior research supported the assumption that gain-framed messages are more effective for people who want to perform behaviors with low risk (prevention behaviors) (Detweiler et al., 1999; Rothman et al., 1993). Because the company is positively in the news, we can assume that there is a low amount of risk for the consumer to buy their products from that company. Therefore the use of gain-frames will lead to a more positive company evaluation, and a high purchase intention of the consumer as when a loss-frame is used. Besides that, research suggests that the use of gain-loss-framed CSR messages would lead to a more positive company evaluation as when a loss frame is used (Bortree, Ahern, Smith & Dou, 2013), and therefore would create a positive perception of CSR and would minimize the awareness of environmental issues. A more positive company evaluation would say that there will also be a positive effect for C-C identification and a company’s reputation. These two factors would have a more positive effect on the purchase intention of the consumer (Keh & Xie, 2009). These outcomes have lead to the following hypotheses:

H1: When people are exposed to positive news about CSR, a gain-framed message will have a

more positive effect on C-C identification (A) than when a loss-framed CSR message is used,

(B) which enhances the purchase intention of the consumers

H2: When people are exposed to positive news about CSR, a gain-framed message will have a

more positive effect on reputation (A) than when a loss-framed CSR message is used, (B) which enhances the purchase intention of the consumers.

When people are confronted with negative news about the CSR activities of a company, it could be considered as a ‘risky choice’ to buy a product from this company. Because the company does not act in line with the expectations of the consumer, people would think that the company is hard to trust and that choosing other companies to buy their

(15)

14 products from would be a saver choice. Rothman and Salovey (1997) presented that when people consider a risky behavior (detection behaviors) loss-framed messages would be more persuasive. So when people are exposed to negative news, they would consider a risky behavior to maintain the old attitude towards the company. When a loss-frame is used people would be more likely to link the company with a more positive company evaluation (C-C identification and reputation) as when gain-framing is used. These two factors would have a positive effect on the purchase intention of the consumer (Keh & Xie, 2009). The two following hypotheses are conducted to measure this relationship:

H3: When people are exposed to negative news about CSR, a loss-framed message will have

a more positive effect on C-C identification (A) than when a gain-framed CSR message is used, (B) which enhances the purchase intention of the consumers.

H4: When people are exposed to negative news about CSR, a loss-framed message will have

a more positive effect on reputation (A) than when a gain-framed CSR message is used, (B) which enhances the purchase intention of the consumers.

(16)

15

METHODOLOGY

RESEARCH DESIGN

The hypotheses of this study were tested by using a 2 (News: positive versus negative) x 2 (Framing: Gain versus Loss) between-subjects design. The respondents were randomly assigned to one off the four experimental conditions. The respondents were exposed to a positive or negative news message with information about the CSR activities of the fictional company Datex Electronics. After being exposed to the news messages the respondents were shown a statement from the company that contained ‘gain’ or ‘loss’ framing. The dependent variables: Consumer-company evaluation, reputation and purchase intention were used to measure the participants’ reaction on the different types of framing. The variable involvement was added as a control variable to the survey.

PARTICIPANTS

A total of 170 respondents (N = 170) participated in this online experiment. The survey was in Dutch, so only Dutch citizens completed the survey. The participants have been reached through snowball sampling in the network of the researcher by using social network site Facebook and by using Email. The participants were randomly assigned to one of the four conditions. The conditions with that contained positive news in combination with gain-framing consisted 42 participants, the condition that contained positive news with loss-framing 48 participants, the condition that contained negative news with gain-loss-framing 44 participants and the condition with negative news in combination with loss-framing 36 participants. The sample consisted of 42% male and 58% female respondents. The age of the participants ranged between 16 and 67 years with a mean of 26.65 years (SD = 9.69). The educational level of the respondents was distributed as followed: Master degree: 19%, Bachelor degree: 22%, higher education: 45%, secondary school: 14%.

PRE-TEST

The stimuli material that is used in this research has first been tested in a pre-test. The pre-test (N=13) measured if people perceived any differences between the negative and positive framed news messages. The paired samples T-test found a significant difference between the negative framed news message (M = 2.46, SD = 1.20) and the positive framed news message (M = 5.15, SD = 1.35) (t(12) = -4.72, p = .00).

(17)

16 The pre-test measured if people perceived any differences between the loss and gain framed material. The paired samples T-test found a significant difference between the loss-framed message (M = 3.93, SD = 1.66) and the gain-loss-framed message (M = 5.77, SD = 1.23) (t(12) = -2.98, p = .01). Therefore it can be concluded that the manipulation has worked and the manipulated material can be used in the survey.

STIMULUS MATERIAL

The stimulus material of this study contained two news messages and two statements from a company’s website. The news messages consisted positive or negative news about the CSR activities of the focal brand. Both messages contained a text about a ranking of an environmental group that rates the CSR activities of companies. The text in the messages was based on a news message about an environmental group ‘climate counts’ from the website: www.emerce.nl. In the negative news condition, the brand scored low on the sustainability ranking and in the positive news condition, the brand scored high on this ranking. The messages were displayed in a lay-out from the website of the Telegraaf, a big news agency from the Netherlands. The design was similar to the webpage of the newspaper Telegraaf www.telegraaf.nl. The amounts of words that are used in these news messages were around the same amount of words.

A neutral message was used as a startup point for the framing manipulation. The text was manipulated in a ‘gain’ and a ‘loss’ frame to test the effect of framing on CSR. The message explained the company’s view about their CSR activities and the consequences of these activities. The message that contained the ‘loss’ framing dimension emphasizes the negative outcomes when the company would not participate in these CSR activities. The ‘gain’ framed message accentuated the positive outcomes when the behavior would be adapted by the company. The messages were displayed in a lay-out from the website of the brand that is used in the questionnaire: Datex Electronics. The lay-out of this page is based on the website www.philips.nl from the electronics company Philips BV. The amounts of words that are used in the news messages were around the same amount of words.

PROCEDURE

All participants were asked to fill out a survey on the web. They received a personal request via e-mail or got a request on Facebook to participate in a study about CSR. The message contained a link to the questionnaire. The survey software of Qualtrics was used and

(18)

17 was accessed by www.qualtrics.com. The questionnaire started with a short instruction about the design of the survey. In this introduction, the participants were instructed about the fictional brand that was used, and that they had to interpret this brand as a company that exists in their real life. It also contained the promise that all answers would be used anonymously and would only be used for this research. After the introduction the participants were randomly exposed to one of the two news messages that they had to read carefully. The next screen contained a manipulation check that tested if the people had read the text. After reading the news message, the participants were randomly assigned to one of the two framed messages. After being exposed to the two stimuli, participants filled out a questionnaire that measured the 1) consumer-company identification, 2) reputation, 3) purchase intention, 4) involvement and 5) demographic variables. The survey ended with a small text that thanks the respondents for completing the survey and a link with the email address from the researcher for when the respondents have any questions about the research.

MEASURES

The dependent variables were measured with a 7-point Likert scale. The scores ranged from 1- disagree to the score of 7 that indicated a strong agreement with the statement.

Consumer-Company identification

A scale of Leach, Zomeren, Zebel, Vliek, Pennekamp, Doosje and Spears (2008) was used to measure the consumer-company identification. Three items from the social identification scale were used to measure C-C identification. One of the items was: ‘After reading the text, I feel involved with Datex Electronics’. The reliability of the scale was α = .93. The items were merged into a mean index, forming the variable ‘Consumer-company identification’ (M = 3.16, SD= 1.59).

Reputation

A scale of Hsu (2012) was used to measure the consumer-company identification. Five items were used to measure the construct reputation. An example of the questions was: ‘Datex has status.’ The reliability of the scale was α = 88. The items were merged into a mean index, forming the variable ‘Reputation’ (M = 4.36, SD= .57).

(19)

18

Purchase intention

To measure purchase intention, a three item scale was used based on Ajzens (1985) scale ‘Direct measures of perceived behavioral control, subjective norm, attitude and intention’. From this scale, the three items that measured intention were stripped and formed into theses that measure purchase intention. One of the items was:’After reading the text, I intend to buy a product from Datex Electronics within a few weeks’. A reliability analysis showed that the scale was reliable with a Chronbach’s alpha of α = .95. The three items form the variable ‘Purchase intention’ (M = 2.86, SD = 1.48).

Involvement

The variable involvement was measured using five items from the involvement scale constructed in the Zaichkowsky (1985) research. An example of an item was: ‘CSR is … for me’ with a 7-point Likert scale from important to unimportant. A reliability analysis showed that the reliability of the scale was α = .86. The three items form the control variable ‘involvement’ (M = 2.67, SD = 1.00).

Demographic variables

The demographic variable age was measured with the item: What is your age in years? The participant had the opportunity to answer the question in numeric values. The second demographic variable was gender, were the participants had the choice to answer with male or female. The last demographic variable that was measured was the level of education of the participants. This question asked the highest level of education from the participants were they had the choice in: master degree, bachelor degree, higher education, secondary school and no education.

Construct Number of items Alpha

C-C identification (Leach et al., 2008) 3 .93

Reputation (Hsu, 2012) 5 .88

Purchase intention (Ajzen, 1985) 3 .95

Involvement (Zaichkowsky, 1985) 3 .86

(20)

19

RESULTS

The relationship between the framing of CSR messages on company evaluations was measured using the dependent variables: reputation, C-C identification and purchase intention. The independent variables were four conditions that contained a positive or a negative news messages about CSR combined with a gain or a loss framed message about CSR. The control variable involvement with CSR was measured to check if involvement has any influence on the relationship between the independent and dependent variables.

MANIPULATION CHECK

A manipulation check was conducted to test the people’s perceptions about CSR messages that were used in the study. The news messages consisted out of a positive message and a negative message. The manipulation check was measured using a 7-point Likert scale were the participants were asked to rate the message as positive or as negative. An independent T-test revealed that people who were exposed to the positive news message scored significantly higher on the scale (M = 5.92, SD = 1.79) than the people who were exposed to the negative news message (M = 1.79, SD = 1.22) (t (168) = -23.63, p = .00). Therefore it can be concluded that the manipulation was effective and that people saw a difference between both framed messages.

The framed messages consisted out of a Gain-framed message and a Loss-framed message. The manipulation check was measured using a 7-point Likert scale were the participants were asked if the message emphasized the possible losses that could be reached by the CSR activities or if the message emphasized the possible gains that could be reached by the CSR activities. An independent sample T-test revealed that people who were exposed to the Gain-framed message scored significantly higher on the scale (M = 5.75, SD = 1.26) than the people who were exposed to the Loss-framed message (M = 5.04, SD = 1.74) (t (151) = 3.02, p = .00). Therefore it can be concluded that the manipulation was successful and that the respondents saw a difference between gain-framed and loss- framed messages.

Positive Negative t p News M = 5.92 M = 1.79 -23.63 .00 SD = 1.06 SD = 1.22 Gain Loss t p Framing M = 5.75 M = 5.04 3.02 .00 SD = 1.26 SD = 1.74

(21)

20 CONTROL VARIABLES

The control variables were used to check if other variables would have an effect on the outcomes of the dependent variables. In this study the demographic variables age, education and gender were used to check for influence on the dependent variables. The extra control variable involvement with CSR was added to see if the relation between involvement and framing would also be applicable in the context of framing about CSR with identification’, ‘reputation’, ‘purchase intention’ as dependent variables. A Chi-square test was conducted to test the relation between the control variables and the dependent variables. The test found no significant relations between the control variables and the dependent variables what means that these background variables have no influence on the results of the study.

HYPOTHESES TESTING

A two way analysis of variance was used for testing the hypotheses H1a, H2a, H3a and H4a. The dependent variables in these hypotheses are C-C identification and reputation and the independent variables are positive/negative news framing of CSR messages and the goal/loss framing of CSR messages. All the results from the analyses were tested with a 95% confidence interval.

The first two hypotheses measured the effect from gain and loss framing on C-C identification and reputation in a positive news setting (A). The second parts of these hypotheses suggest a mediated effect from C-C identification and reputation on purchase intention (B). The third and four hypotheses measured the effect from gain and loss framing on C-C identification and reputation in a negative news setting (A). The second part from these hypotheses suggests mediation from C-C identification and reputation on purchase intention (B).

The effect of gain and loss framing on reputation/C-C identification might differ across a positive/negative news message. A two-way analysis of variance found that the positive news messages had a significantly higher rating on the variables reputation (F (1, 166) = 5.08, p = .025) and C-C identification (F (1, 166) = 70.06, p = .00) than those who have seen the negative news messages. No significant difference was found between the gain framed messages and loss framed messages on reputation (F (1, 166) = .284, p = .595) and C-C identification (F (1, 166) = .337, p = .563). A two-way independent analysis of variance test found no interaction effect (see table 4 for means) between the framing of CSR and the positive/negative news setting of CSR on the variable C-C identification (F (1, 166) = 1.153,

(22)

21 p = .285). Therefore it can be concluded that the hypotheses H1a and H3a are rejected because there is no significant difference between gain framing and loss framing in the two news conditions on C-C identification. The two-way independent analysis of variance test found no interaction effect between the framing of CSR and the positive/negative news setting of CSR on the variable reputation (F (1, 166) = .158, p = .692). Therefore it can be concluded that the hypotheses H2a and H4a are rejected because there is no significant difference between gain framing and loss framing in the two news conditions on reputation.

Gain framed Loss framed

C-C identification Positive Mean 4.04 2.08

News SD 1.41 1.08

Negative Mean 3.94 2.43

SD 1.48 1.36

Reputation Positive Mean 3.60 3.77

News SD 0.56 0.65

Negative Mean 3.62 3.85

SD 0.47 0.56

- Table 3: Mean level of C-C identification and Reputation by News coverage and Gain/Loss Framing.

The hypotheses H1b, H2b, H3b and H4b were analyzed using regression analyses. Hypotheses H1b and H3b proposed that a higher C-C identification would lead to a higher purchase intention from the consumer. The linear regression analyses found a significant relation between the framed CSR messages and C-C identification (b = -.69, t(169) = -6.90, p = .00), a significant relation between the framed CSR messages and purchase intention (b = -.49, t(169) = -5.03, p = .00) and a significant relation between C-C identification and purchase intention (b = -.68, t(169) = 14.06, p = .00). In the second analyses, C-C identification was added to the independent variables. In this analyses the significant relation between the framing of CSR messages and purchase intention disappeared (b = -.28, t(169) = -.35, p = .73) and that there was still a significant relation between C-C identification and purchase intention (b = -.67, t(169) = 12.21, p = .00). The relation between the framing of CSR messages and purchase intention disappeared when the variable C-C identification was added to the model. Therefore it can be concluded that the relationship between the framing of CSR messages and purchase intention is a fully mediated relationship with C-C identification as a mediator. The hypotheses H2b and H4b proposed that a higher reputation would lead to a higher purchase intention at the consumer. The linear regression analyses found no significant relation between the framed CSR messages and reputation. The analyses found no significant relation between reputation and purchase intention. Therefore it can be concluded that

(23)

22 reputation is no mediator in the relationship between the framing of CSR messages and purchase intention.

(24)

23

CONCLUSION & DISCUSSION

This study contributes to the knowledge about the communication of CSR messages. The experiment explored the influence of the framing of CSR messages on consumer responses (reputation, C-C identification and purchase intention). This is a first step in creating a framework that marketers can use when their company experiences positive or negative publicity. By comparing the outcomes of message framing in a positive/negative news condition it can be concluded that gain-/loss-framing has no influence on a consumer’s company evaluation and on partially on the purchase intention of the consumer via C-C identification.

FINDINGS

The study expected that people would react differently to framed CSR messages when they were primed with positive or negative news messages. The hypotheses that predicted an effect of gain-framing and loss-framing on company evaluation and purchase intention were rejected. The first two hypotheses (H1a and H2a) predicted that when a consumer was exposed to positive news about a company, gain-framed messages would have a more positive effect on C-C identification and reputation as when a loss-framed CSR message was used. The analyses in this study found that gain-framing or loss-framing has no influence on C-C identification or reputation. The other two hypotheses (H3a and H4a) predicted that when a consumer is exposed to negative news about a company, loss-framed CSR message would have a more positive effect on C-C identification and reputation as when a gain-framed CSR was used. These hypotheses were also rejected. The outcomes of this study are contradictory to the predictions by the prospect theory from Kahneman and Tversky (1979). This theory suggested that all people have their own reference point and therefore will react differently on gain/loss framed messages. A person would minimalize their losses by taking more risks to make profits (risk seeking) and would try to avoid losses and choose for a safe solution when they are exposed to a gain framed message (risk averse). Rothman, Bartels, Wlaschin and Salovey (2006) applied the prospect theory to health behavior and found that gain-framed messages were more effective with prevention behavior and that loss-framed messages would be more effective with prevention behaviors. These findings were supported by a range of studies (Cox & Cox, 2001; Meyerowitz & Chaiken, 1987; Schneider et al., 2001).

As shown in this study; the prospect theory could be implemented in the framing of CSR messages. This means that gain-framed messages would be more effective to people

(25)

24 who are exposed to a positive news message, but interestingly the effect was not found in this study. An explanation for the absence of this effect could be explained by using the study of Wang (2007) about CSR and framing. Wang (2007) found that the effect of framing stayed out when people were not primed enough. The outcomes of this study suggested that priming and framing works together to achieve a higher level of persuasion because of the increased awareness of the participants. Therefore it is possible that people’s judgments about a framed CSR message would not be susceptible to the framing effects because the issue was not activated enough by priming. In further research it will be better to prime the respondents with more than one framed message. Therefore people would be more aware of the message and the issue that is being communicated. As a result, that message would be activated better. Another method to raise the awareness of the respondents is to reduce the text that is used in the message. When the text only contains a framed sentence, people would be more aware about the topic that is being communicated with the framed message. People would be less derived and therefore more concentrated on the content of the message.

Another argument that could be an explanation for the absence of the effects from framing on C-C identification and reputation was supposed by Meyerowitz and Chaiken (1987). They suggest that the prospect theory only focuses on behavior that contains risk and does not pay attention to the fact that some decisions may have no risk involved, what could lead to different results. This means that people who have low involvement with CSR would pay no attention to the risk that is involved with some decisions, what will lead to the absence of an effect between framing and company evaluation. This proposed relation between involvement and framing was tested in the control analyses and the outcome of this test was contrary to the results from the study of Meyerowitz and Chaiken (1987). The control analyses found no influence from CSR involvement on the relation between CSR framing and company evaluations. Another possibility is the lack of risk from CSR behavior in general. It is possible that people do not feel personally threatened by the consequences of CSR because most of the implications will only occur in the long term, what is contrary to health behavior because most cases do have immediate influence on the personal lives of people. Therefore it is possible that the prospect theory is not suited to influence CSR related behavior. Further research should explore this by expanding on this study by measuring the perceived risk that a consumer experiences when reading a CSR message. This addition to the research will help to find explanations for the outcomes of this study.

But when there are no differences between gain-framing and loss-framing in different news situations, why did marketers mostly use gain-frames in communicated CSR messages

(26)

25 over the past 30 years? Bortree, Ahern, Smith and Dou (2013) discussed the use of gain-frames in CSR communication in their study. The use of gain-gain-frames would create a positive perception of CSR and would minimize the awareness of environmental issues. They encouraged marketers to use loss-framing in CSR messages to let people contemplate about environmental issues. The outcomes of this study gives empirical evidence that there is no difference between gain-framing and loss-framing in either a positive or a negative news condition. This opens new possibilities for marketers to start using loss-framed advertisements in their CSR communications. Loss-framed messages would lead to the same company evaluations and purchase intention at the consumer as gain-framed messages do, but as suggested by Bortree et al. (2013) would let people think more about environmental issues and will hold companies more responsible for their actions towards the environment. This would raise the consumer awareness towards CSR and the companies that are involved with their environment (Pomering & Dolcinar, 2009).

This study also tried to find an indirect relation between the framing of CSR messages and the purchase intention from a consumer. The hypotheses H1b and H3b expected a mediated relation between the framed CSR messages and purchase intention, with the variable C-C identification as a mediator. First the analyses found a significant relation between the framing of CSR messages and purchase intention, but after the variable C-C identification was added to the model, this relationship disappeared and made place for a significant indirect relationship via C-C identification. Brown and Dacin (1997) earlier suggested that it only seems that CSR activities only have an indirect effect on purchase intention. This suggested relation is supported by Keh and Xie (2009) who found relation between a consumers purchase intention with the company evaluation of a consumer. They say that consumers have to be aware of a company’s CSR initiatives before this factor could influence their purchase decisions. But even when the consumers are aware of these initiatives, the majority still bases their purchase decisions on traditional criteria such as price, quality, reputation, location and service. It is self-evident that C-C identification can be seen as one of those traditional criteria on which consumers base their purchasing decisions on. Another important factor of C-C identification is the resistance that consumers will have to negative information about the company in question (Bhattacharya & Sen, 2003). Therefore consumers will be more loyal to existing products and have a higher willingness to try new products as people who have a low C-C identification. Marin, Ruiz and Rubio (2009) found that a high level of consumer identification would lead to a more positive company reputation, brand loyalty and a higher purchase intention. Therefore it was expected that the hypotheses H2b and H4b were a

(27)

26 mediated relation between the framed CSR messages and purchase intention, with the variable reputation as a mediator would also be significant. But the outcomes of the analyses were contrary to the expectations in the hypotheses. The outcomes showed no significant relation between the framed CSR messages and reputation and no significant relation between reputation and purchase intention. These outcomes are contrary to the results from earlier studies that found that reputation has a positive effect on a company’s capital (Spicer, 1978). A more recent study from Keh and Xie (2009) found a positive relation between the reputation of a firm and the purchase intention of the consumer. These outcomes are supported by Melo and Garrido (2012) who found that CSR initiatives raised the reputation of a company, what eventually would lead to a competitive advantage for a company. Why this study has not found a relation between reputation and purchase intention is hard to say. The high quality were the company is associated with should strengthens people’s confidence and reduce people’s risk perceptions to buy a company’s products (Maden et al., 2012). An explanation for this could be construction of reputation, because it is hard to duplicate and has to be built up over a longer time span (Drejer, 2000). Therefore it can be hard for people to form a clear judgment about a company’s reputation when it is the first time they have heard from the company and will not have a higher purchase intention when they rate the company as a firm with a positive reputation.

IMPLICATIONS

This is the first study that explored the situation of a company in the news (positive/negative news), the effects from gain-framing/loss-framing on these various news situations and their relation with C-C identification, reputation and purchase intention. The study showed some first steps in making a framework for marketers about the relation between the framing of CSR messages and the consumer evaluations of a company. This study made clear that there is not one correct method to frame a CSR message, what opens ways for marketers to experiment with loss-framing in their CSR messages. This would open people’s minds about the CSR activities from companies and would let the people think more about environmental issues (Bortree et al., 2013). Companies can choose themselves which message they want to send their audience with their CSR communications. The use of gain-framing promotes the CSR activities of a company and tries to create a positive attitude towards CSR. On the other side marketers can try to use loss-framing in their CSR communications and try to let people think more about environmental issues. It would raise

(28)

27 the awareness of these environmental issues and would have a better evaluation of a company’s CSR activities. But whether the marketer chooses for a gain frame or for a loss frame, both ways of framing will have the same effect on C-C identification, reputation and a consumer’s purchase intention.

GENERAL CONCLUSION

In the end it can be concluded that the answer to the research question of this study: ‘To what extend does the framing of a CSR message in a positive/negative news condition influence consumers’ company evaluation and which effect does this has on purchase intention?’ is that the framing of a CSR message does not have influence on the C-C identification, reputation or purchase intention from the consumer, neither in a positive news condition nor in a negative news condition. From the analyses it can be concluded that there was no difference between the effects of gain-framing and loss-framing on C-C identification and reputation in both the positive and the negative news condition. C-C identification was a mediator in the relationship between the framed CSR messages and purchase intention. This implies that the relation that was found between the framing of CSR messages and purchase intention would be an indirect one. Therefore it can be concluded that the prospect theory is not suited to influence CSR related behavior, what means that there is no right or wrong way to frame your messages in CSR communication. Both ways of framing (gain/loss) will lead to the same consumer evaluations.

(29)

28

BIBLIOGRAPHY

Ahearne, M., Bhattacharya, C. B., & Gruen, T. (2005). Antecedents and consequences of customer-company identification: expanding the role of relationship marketing. Journal of Applied

Psychology, 90(3), 574-585.

Ajzen, I. (1985). Intention, perceived control, and weight loss: An application of the theory of planned behavior. Journal of Personality and Social Psychology, 49(3), 843-851. Bhattacharya, C. B., & Sen, S. (2003). Consumer-company identification: A framework for

understanding consumers’ relationships with companies. Journal of marketing, 67(2), 76-88.

Bhattacharya, C., & Sen, S. (2004). Doing better at doing good: When, Why and How consumers respond to corporate social initiatives. California Management Review, 47(1), 9-24. Bizer, G. Y., & Petty, R. E. (2005). How we conceptualize our attitudes matters: the effects of

valence framing on the resistance of political attitudes. Political Psychology, 26(4), 553-568.

Bohm, P., & Lind, H. (1992). A note on the robustness of a classical framing result. Journal of

Economic Psychology, 13(2), 355-361.

Bortree, D. S., Ahern, L., Smith, A. N., & Dou, X. (2013). Framing environmental responsibility: 30 years of CSR messages in National Geographic Magazine. Public Relations Review, 39(5), 491-496.

Brammer, S. J., & Pavelin, S. (2006). Corporate reputation and social performance: The importance of fit. Journal of Management Studies, 43(3), 435-455.

Brown, T. J., & Dacin, P. A. (1997). The company and the product: corporate associations and consumer product responses. The Journal of Marketing, 61(1), 68-84.

Cone. (2013). The 2013 Cone Communications/Echo Global CSR study. Retrieved 6 18, 2014, from www.concecom.com: http://www.conecomm.com/2013-global-csr-study-release Cox, D., & Cox, A. D. (2001). Communicating the consequences of early detection: The role of

evidence and framing. Journal of Marketing, 65(3) , 91-103.

Cramer, J., Jonker, J., & van der Heijden, A. (2004). Making sense of corporate social responsibility. Journal of Business Ethics, 55(2), 215-222.

Detweiler, J. B., Bedell, B. T., Salovey, P., Pronin, E., & Rothman, A. J. (1999). Message framing and sunscreen use: gain-framed messages motivate beach-goers. Health Psychology, 18(2), 189.

Drejer, A. (2000). Organisational learning and competence development. Learning Organization,

7(4), 206-220.

Entman, R. M. (1993). Framing: towards clarification of a fractured paradigm. Journal of

Communication, 43(4), 51-58.

Fagley, N. S., & Miller, P. M. (1990). The effect of framing on choice: Interactions with risk-taking propensity, cognitive style, and sex. Personality and Social Psychology Bulletin, 16(3), 496–510.

(30)

29 Fombrun, C. J. (1996). Reputation: Realizing value from the corporate image. Harvard Business

Press.

Fombrun, C., & Shanley, M. (1990). What's in a name? Reputation building and corporate strategy. Academy of Management Journal, 33(2), 233-258.

Gergen, K. (1992). Social construction and moral action. In D. Robinson, Social discourse and

moral judgment (pp. 9-27). San Diego: CA: Academic.

Hallahan, K. (1999). Seven Models of Framing: Implications for Public Relations. Journal of Public

Relations Research, 11(3), 205-242.

Highhouse, S., & Yüce, P. P.-t. (1996). Perspectives, perceptions, and risk-taking behavior.

Organizational Behavior and Human Decision Processes, 65(2), 159-167.

Hsu, K. T. (2012). The Advertising Effects of Corporate Social Responsibility on Corporate Reputation and Brand Equity: Evidence from the Life Insurance Industry in Taiwan.

Journal of Business Ethics, 109(2), 189-201.

Inguarta, J.-J., & Cheng, L. (2009). Effect of Group Cue While Processing News on Immigration.

Journal of Communication, 59(4), 726–749.

Inoue, Y., & Lee, S. (2011). ‘Effects of different dimensions of corporate social responsibility on corporate financial performance in tourism-related industries. Tourism Management,

32(4), 790-804.

Jordan, N. (1965). The" asymmetry" of" liking" and" disliking": A phenomenon meriting further reflection and research. Public Opinion Quarterly, 29(2), 315-322.

Kacperzyk, A. (2009). With greater power comes greater responsibility? Takeover protection and corporate attention to stakeholders. Strategic Management Journal, 30(3), 261-285. Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk.

Econometrica: Journal of the Econometric Society, 47(2), 263-291.

Keh, H. T., & Xie, Y. (2009). Corporate reputation and customer behavioral intentions: The roles of trust, identification and commitment. Industrial Marketing Management, 38(7), 732-742.

Leach, C. W., van Zomeren, M., Zebel, S., Vliek, M. L., Pennekamp, S. F., Doosje, B., et al. (2008). Group-level self-definition and self-investment: A hierarchical (multicomponent) model of in-group identification. Journal of Personality and Social Psychology, 95(1), 2008. Lee, K., & Shin, D. (2009). Consumers' responses to CSR activities: The linkage between

increased awareness and purchase intention. Public Relations Review, 36(2), 193-195. 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.

Maden, C., Arikan, E., Telci, E., & Kantur, D. (2012). ‘Linking corporate social responsibility to corporate reputation: a study on understanding behavioral consequences,’. Social and

The gains and losses of Corporate Social Responsibility : how the gain/loss framing of CSR messages affect consumer evaluations