Crisis communication between corporation, the news media and public : the case of Primark

Crisis communication between corporation, the news media and public:

the case of Primark.

Author: Anne van der Pijl Student number: 10705090 anne.vanderpijl@student.uva.nl

University of Amsterdam Graduate School of Communication

Master’s thesis Communication Science: Corporate Communication Supervisor: F. Schultz

Abstract

This present study aims to investigate crisis communication between Primark, the news media and the public. Previous research on crisis communication is mostly concerned with only analyzing the organizational perspective. This present study does not only focus on the

organizational perspective, but also focuses on the news media and public reactions. The use of framing, response strategies, tone and actors and issues was compared between the different groups. A cross-national comparative research compared UK and Dutch messages.

In this study a total number of 188 documents were analyzed. Both a qualitative and quantitative content analysis were performed in order to understand how Primark and the public interact in their crisis communication. The attribution of responsibility frame was found to be the most used frame in news articles and blogs. Furthermore, this study found in line with previous research that accommodative response strategies prove to be successful for the

protection of the organizational reputation.

Key words: crisis communication, response strategies, framing, agenda setting, tone, actors, issues, news media, public, Primark

Introduction

Public relations practices have changed a lot since the emergence of Web 2.0. It has both fostered challenges, as well as opportunities for corporate reputation management.

Corporations use the Web to present themselves as socially responsible citizens and to advance their legitimacy positions (Esrock & Leichty, 1998). The Internet has challenged the

communication dynamics between organizations and the public. The news media can easily distribute their news messages to the entire world. But also the public, especially protestors, increasingly got more power to affect the legitimacy of corporations (Castello, Morsing & Schultz, 2013). According to Jones, Temperley and Lima (2009) the World Wide Web has transformed stakeholders into active consumer watchdogs, investigation journalists and opinion influencers. This has significantly changed the power of the consumer. Once a negative story about a company is published on for example a blog or social media, it can reach billions of people all over the world within a few seconds. An organizational crisis can thus easily evolve. Research on crisis communication suggests that what organizations say and do during and after a crisis largely determines the aftermath of a crisis. Therefore it is

important for organizations to anticipate in the right way with proper crisis response strategies using the right type of medium if a crisis occurs (Schultz, Utz & Goritz, 2010; Coombs, Frandsen, Holladay & Johansen, 2010).

This study will look at the case of Primark, and how the corporation, the news media and public communicate about Primark’s crises. Primark is an Irish low budget-clothing retailer that sells fashionable clothes for fairly low prices. Primark claims to be a very ethical company. There are a lot of CSR related activities mentioned on their website

http://www.primark.com/en/our-ethics. For example, they claim to be involved in local

communities regarding women’s health and educational programs. Furthermore, Primark says they make sure the factories are safe and that they pay attention to human rights. They also

justify how they can offer the lowest prices on the high street (Primark, 2014). However, Primark recently was involved in two major crises. The first crisis was the collapse of the Rana Plaza, the garment factory in Dhaka in Bangladesh where more than 1000 people died. This happened on April 24th, 2013. The second crisis occurred a year later in June 2014. Several people reported that they found ‘cry for help’ labels in Primark clothing (for an example see appendix 1). Primark received a lot of negative media attention, and after both happenings there was a lot of debate about the work conditions in the Primark factories in Bangladesh and other developing countries. This study will focus on how Primark dealt with their crisis communication and response strategies in order to restore their reputation and maintain legitimacy.

Purpose

There seem to be a lot of ongoing crisis in the fashion and clothing retail industry. There is for example a lot of criticism on many brands about the working conditions and human rights of the people who work in their factories in developing countries. However, there is very little academic research devoted to crises within this sector.

There is a large number of studies on crisis situations, however, according to Coombs (2007) and Kim and Cameron (2011), this field is dominated by studies that show low ecologically validity and additionally, therefore these studies often present unidirectional effects. Moreover, most crisis research is approached from the organizational perspective, and does not take the perceptions of the media and stakeholders into consideration (Coombs, 2006; Coombs, 2007). Accordingly, the perspectives of the public and the media have often been ignored in academic studies. Hence, the objective of this study is to overcome these research gaps. Budget retail stores (e.g. Primark) are popular among the public, it is interesting to see how the public, news media and company report on crises in this sector. It is socially relevant to investigate whether

the public actually cares, or whether they just want to buy cheap clothes. The frequency of communication, the content of communication and the tone of the communication from Primark, the news media as well as the public will be investigated.

The key contribution of this paper is to deepen the understanding of crisis response strategies, how Primark interacts with the audience, how the news media responds to the crisis, how the audience responds, and what affect this has on corporate reputation and legitimacy. Both responses of the company, as well as responses from the public and news media will be investigated through a comparative study that compares the UK and the Netherlands, in order to find out how legitimacy is maintained or affected and whether Primark is successful in its crisis communication in both countries. The research questions will be investigated through a both qualitative and quantitative content analysis using Atlas.ti and SPSS.

The general research question of this study is:

How do Primark, the news media and the public in the UK and Netherlands report differently or similarly on both of Primark’s crises and how does this affect Primark’s reputation?

Model 1 was constructed to give a clear overview of the comparison. This model can be found in appendix 2.

Theoretical framework

Organizational reputation and legitimacy

Corporate reputation is essential in today’s corporate environment. Fombrun (1996) defines corporate reputation as ‘a perceptual representation of a company’s past actions and future prospects that describes the firm’s overall appeal to all of its key constituents when compared

with other leading rivals’ (p.70). A company’s reputation is constructed through the interaction between stakeholders and the company. Reputation is not a passive thing, but according to Christensen, Morsing and Cheney (2008) it must be ‘created, managed and sustained’ (p.90). Gray and Balmer (1998) argue that corporate reputation is an essential corporate asset, and link it directly to competitive success.

Corporate reputation management has emerged as an important field in public relations. However, reputation management is not an easy task because reputation is an intangible concept. Budd (1994) (as cited in Payne, 2006) argues that “because reputations are not a physical commodity, the challenge [of managing reputation] lies in managing all signals sent by a company – a story, an action, a report, a meeting, or an interview - for the best possible reputation”. This is especially the case in crisis communication. When a company is involved in a crisis, reputation management is crucial, because negative effects of a crisis can have long-term consequences on a company’s reputation. Reputational threats can emerge from bad publicity that appears in the media. The Internet has challenged reputation management since it is impossible to track everything that is being said on the Web. A damaged reputation can have negative impact on sales and profits, especially if the crisis is not managed properly

(Vanhamme & Grobben, 2009). A good corporate reputation is needed in order to maintain corporate legitimacy. Corporate legitimacy can be defined as the “license to operate”. The license to operate is based on social construction of social norms, values and expectations that organizations perform in their actions (Scherer & Palazzo, 2006).

Crisis communication

Even with good reputation management, it is impossible for organizations to avoid damage to a reputation when it is involved in a crisis (Coombs et al., 2010). A crisis can be defined as “a low-probability, high-impact event that threatens the viability of the organization and is

characterized by ambiguity of cause, effect, and means of resolution” (Pearson & Clair, 1998, p. 60). Because of the emergence of Web 2.0, there is no longer full control over media coverage. The web is a world full of critical voices. In order to restore the reputation that is affected by a crisis, crisis communication is an essential aspect in crisis times. The Internet has made crisis communication more complex compared to the old crisis communication that used traditional media. Macnamara (2010) calls this shift from controlled content (traditional media) to uncontrolled content (the Internet) the collapse of the control paradigm. Macnamara means that the Internet has made it much more complicated to steer messages in the right direction to a specific target audience. Procedures and practices to manage online reputation have therefore become a vital activity of public relations (PR). However, organizations can also use the Web to build and maintain their presence and legitimacy because they can go into a direct dialogue with the public and other stakeholders. For organizations it is therefore important to quickly anticipate to a crisis with an appropriate response strategy via the right medium type (Schultz, Utz & Goritz, 2011; Coombs, Frandsen, Holladay & Johansen, 2010).

Coombs (2007) developed the Situational Crisis Communication Theory (SCCT). This theory is a framework that is used to understand how organizations can maximize reputation management by post-crisis communication. SCCT is a framework that indicates how reputation is perceived by the public (i.e., favorable vs. unfavorable evaluations) (Kim & Cameron, 2011). SCCT also holds that information about past crisis events in a (news)

message can significantly affect perceptions of a recent crisis. How well companies perform in managing their reputation depends on the approach they take. A crucial aspect is how a

company engages with the audience. To achieve maximum success in managing the online reputation, companies have to engage in conversations with the public (Jones, Temperley & Lima, 2009).

Studies have shown that the type of medium that organizations use, as well as the crisis

response strategies influence corporate reputation (e.g. Schultz, Utz & Goritch, 2011; Coombs & Holladay, 2007). Responding is thus a key factor to solving a crisis and restoring one’s reputation. CSR communication is for example seen as an essential part of reputation management when restoring from a crisis (Jones, Temperley & Lima, 2009; Vanhamme & Grobben, 2009). It is thus very important how organizations frame and justify their response regarding a crisis.

RQ 1a: How much did Primark communicate in general in the crises? RQ 1b: What actors and issues did Primark use in its crisis communication?

Crisis Response strategies and framing

Research has shown that crisis response strategies serve as a toolkit in order to protect a reputation after a crisis (e.g., Coombs & Holladay, 1996; Coombs et. al., 2010). Crisis communication is essential because it can improve or worsen a crisis (Coombs et al., 2010). The attribution theory is a framework that is used to conceptualize crisis management. This theory focuses on three dimensions on which people judge a situation; these three dimensions are locus, stability and controllability (Coombs, 1995). Locus is the locus of control, whether internal or external factors caused a crisis situation. Stability refers to whether the cause of the crisis is always present, or just in times of the crisis. Lastly, controllability refers to the level of control the actor had on the crisis situation, whether it was beyond control of the actor or whether the actor influenced the cause. Those three attributions shape the perception of a crisis. Coombs (1995) argues that the stronger the attributions are, the more likely it is that an

organizational reputation is damaged. Proper crisis response strategies are therefore essential to restore this damage. Common used crisis response strategies identified by Coombs and

other parties or response to embarrassment (Coombs, 1995; Coombs, 2006; Coombs, Frandsen, Holladay & Johansen, 2010).

Hallahan defines public relations (PR) as ‘the construction of reality’ (1999, p. 206). PR is all about shaping organizational perspectives, framing is critical because frames can construct different perspectives. PR is regarded as successful when the content and framing of press releases are understood by the stakeholders as intended by the organization. Framing is thus an essential part of PR. Effective PR should (positively) affect the perceptions and reactions stakeholders hold towards the organization (Schultz, Kleinnijenhuis, Oegema, Utz, & van Atteveldt, 2012). According to Hallahan (1999) framing is a potential useful paradigm when it comes to creating public relations response strategies. Framing is connected to underlying psychological processes that humans use to examine and judge information. Framing provides contextual cues that guide human decision-making. This means that a frame defines a

message's meaning by shaping the assumptions that people make about the content of the message (Hallahan, 1999).

RQ1c: What response strategies and frames did Primark use in its crisis communication?

The construction of crises in the news media

How salient a topic is on the news agenda can determine how the rest of the crisis will evolve and how it is going to affect a company’s reputation. This indicates that agenda setting is a crucial factor in crisis communication. Caroll and McCombs (2003) define agenda setting as the ability of the media to influence the salience of a topic on the public agenda.

Which objects appear on the media agenda is the initial stage in agenda setting theory, attention is thus the starting point (Caroll & McCombs, 2003). Wartick (1992) found that media coverage does affect a firm’s performance. Agenda setting is an important factor for

determining a corporation’s reputation and social legitimacy. If the public forms an image about an organization, this is often based on messages that appear in the media. So corporate reputations are often based on how corporations are portrayed in the media (Caroll &

McCombs, 2003). Therefore positive media coverage and public attention are essential for organizations, and can help them to shape positive reputations. Negative publications about crisis situations can negatively influence the corporate reputation on the short –and long term.

There is a distinction between first level agenda setting and second level agenda setting. The first level of agenda setting refers to the salience of objects on the news agenda. These are the amount of times objects appear on the news – and public agenda. Weaver (2007) refers to this as the perceived importance of issues. The second level of agenda setting is concerned with the salience of the attributes of the object (Caroll & McCombs, 2003; Weaver, 2007). This can be seen as an evaluative dimension, where attributes are typically defined as positive, negative or neutral. Second level agenda setting is more than looking at facts, it also focuses on feeling and tone in a message (Deephouse, Caroll & McCombs, 2001; Caroll & McCombs, 2003).

Whereas first level agenda setting is focused on attention, second level agenda setting is more concerned with comprehension of the issue (Caroll & McCombs, 2003).

RQ 2a: How much did the news media communicate in general about the crisis compared to Primark?

RQ 2b: What actors and issues did the news media use in its communication compared to Primark?

The news media and framing

According to Schultz, Kleinnijenhuis, Oegema, Utz and van Atteveldt (2011) framing is especially important in crisis communication. Framing is a very important aspect of crisis communication because the media attention an organization receives during a crisis can

influence their legitimacy. The way a message is framed determines how people make sense of the event. Pattriotta, Gond and Schultz (2011) argue that legitimacy requires social actors to justify their positions towards the public. Thus in order to receive legitimacy, actions in which the organization is involved should be justified towards stakeholders. According to Pattriotta, et al. (2011) this has three important aspects. The first aspect is that, when developing

justifications, corporate actors have to come up with rationales that match the socially accepted definitions of the common good. Second, in order to do this, corporate actors have to take into consideration that there are different social groups have competing definitions of what is the common good. Corporate actors have to engage in these conversations about what is the common good to different social groups. And third, to develop effective justifications, specific competencies are needed in order to construct convincing accounts and arguments.

Previous literature has identified a wide range of different frames. In their study on crisis frames An and Gower (2009) used five frames that according to Neuman, Just and Crigler (1992) are the most dominantly used frames in U.S. news coverage. An and Gower use these frames to see which frames are most used in crisis communication. These five frames are human-interest, conflict, morality, economic consequences, and attribution of responsibility. The human-interest frame “brings a human face or an emotional angle to the presentation of an event, issue, or problem”. (Semetko & Valkenburg, 2000, p. 95). This frame is focused on how people are personally affected by an issue or event. This frame stimulates emotional response, and often leads to a more negative attitude toward the crisis. Within the conflict frame a conflict between certain groups, individuals or organizations is central. The morality frame is about the context of morals regarding the problem or event. The economic frame is concerned about economic consequences an issue or problem can have on individuals, groups or

organizations. The attribution of responsibility frame is how responsibility of an issue or event is attributed to a group, individual or organization.

According to Boin, ‘t Hart and McConnel (2009) crisis communication can be seen as a ‘contest’ between frames and counter-frames from various groups that are involved in the particular crisis. Those frames portray differences in the nature and severity of a crisis, its causes, the responsibility or escalation of the crisis, and the future implications for the organization and stakeholders. Within this ‘contest’ contestants are in a battle to have their frame accepted as the dominant frame. Framing is also about the tone of a message. This tone can be either negative, positive or neutral. Framing analysis is a useful tool for analyzing crisis communication because it can provide insights into which crisis response strategies are

appropriate to reduce the damage to the organizational reputation (An & Gower, 2009). RQ 2c: What frames did the news media use in it crisis communication compared to Primark? RQ 2d: What tone did the news media use in its communication?

The public and crisis communication

When a crisis occurs, organizations have to convey messages to their stakeholder groups to inform them about the crisis. Crisis communication is a way to reestablish organizational legitimacy (Lee, 2004). However, there are few practical insights on how the public actually responds to a crisis (Coombs & Holladay, 2008). This present study argues that blogs are a rich source to find out what the public opinion is on a crisis. Blogs are easy to create and often express personal opinions on contemporary issues that receive a lot of media attention (Thelwall & Stuart, 2007).

According to Lee (2004) the severity of a crisis is largely determining how consumers react to a crisis. High crisis severity may trigger negative emotional responses; this might also send danger signals to other observing consumers. It is for example an important factor if there are direct consequences for the consumers. Pennings, Wansink and Meulenberg (2002) argue that the attitude of the public towards a crisis largely depends on risk perception and risk attitude.

They suggest that when a company is involved in a crisis that poses direct risks to the public, effectively communicating correct information is a powerful tool that can change the attitudes and behavior of the public.

According to Lee (2004), the response strategy of the organization can influence the public’s evaluation of the organization. For example, denial as response strategy can lead to a negative evaluation where the public thinks the organization avoids the blame. In this case the public perceives the actions of the organization as blameworthy. Acceptance of responsibility appears to be more honest, which may reduce negative reactions of the public. Previous research (e.g. Lee, 2004; Coombs & Holladay, 2008) confirmed that acceptance of responsibility of a crisis can increase sympathy and forgiveness.

Highly accommodative strategies, to which Coombs and Holladay (2008) refer to as strategies that show apology, sympathy and concern for victims by for example paying compensation, are found to be the most effective response strategies in crisis communication. Previous research compared those accommodative strategies to less accommodative strategies (e.g. refusing to comment, denial, excuse, or justification of actions). These studies found that apology is the preferred strategy, because it has the strongest positive effect on perceptions of an

organization's reputation. This shows that honesty and responsibility are highly valued by the public.

RQ 3a: How much did the public communicate in general about the crisis compared to Primark and the news media?

RQ 3b: What actors and issues did the public communicate about compared to Primark and the news media?

RQ 3c: What frames did the public use in its crisis communication compared to Primark and the news media?

RQ 3d: What tone did the public use in its crisis communication compared to the news media?

Method

In order to understand how Primark, the news media and the public communicate in their crisis communication a content analysis was conducted. This study specifically looks at the messages (press releases, news publications and blogs) of Primark, the news media and public. The main focus of this study is on how they report differently or similarly on both the Rana Plaza and help label incidents and how this way of reporting possibly affects Primark’s reputation. In this study a total number of 188 documents were analyzed. This study used a combination of both qualitative and quantitative content analyses. This is also called triangulation; this involves the comparison of two or more research techniques to study the object of research interest. If the data from two or more methods point toward the same conclusion, then validation is enhanced (Lindlof & Taylor, 2011). Using these two different forms of content analysis enriches the results and implications of this study. According to Lindlof and Taylor (2011) organizational documents are a site of claims to power, legitimacy and reality. Crisis communication practices are encoded and preserved as documents. Therefore analyzing these documents is a rich source of data to gain insights in crisis communication practices and the implications it has.

This study focuses on UK and Dutch news media and messages from the public. It is interesting to see whether the news and public perception differ in these countries. First, a qualitative analyses was performed, Atlas.ti was used to analyze the unstructured qualitative data. A codebook was developed to label the relevant content that was found in the documents. A concept indicator model was created to summarize the findings. After the analysis in Atlas.ti the data about the frequency of the all codes were transformed to an Excel file. This Excel file was exported to SPSS in order to perform some basic statistical analyses.

Data collection and units of analysis

The messages that were analyzed for this study were specifically focused on two crises, the crisis of the collapse of the Rana Plaza in April 2013, and the cry for help labels that were found in Primark clothing in June 2014. This study was conducted from November 2014 – January 2015. The data that were used in this study were collected in December 2014. Therefore, the messages, publications and blogs that were analyzed for this study date from April 2013 – December 2014.

Documents were retrieved from various online sources such as newspaper websites and blogs. All Primark press releases were directly sourced from the official Primark website

(http://www.primark.com/en/our-ethics). Since there is an endless amount of blogs and

websites discussing issues related to Primark and its crises (some just one sentence, some entire documents), the websites with the highest search engine optimization (SEO) were selected for this study. All blogs and news releases had to specifically talk about Primark in relation to the Rana Plaza or ‘cry for help’ label incident. Just mentioning the name of Primark as supplier did not suffice.

Primark is active on Facebook and Twitter, but those social media account are solely used to promote their products, and are not used for corporate messages related to crisis

communication or CSR efforts. Therefore Primark’s social media pages are not taken into account in this study. News outlets that use social media often refer to the full article on their website, this makes analyzing social media posts ineffective. To make the comparison fair, social media messages from the public were also not taken into account.

For public messages the focus is on public blogs and websites/blogs/press releases from protest groups/NGO’s. As mentioned before, these were selected based on their SEO. For articles from the news media, the online versions of major UK and Dutch news outlets were used as source.

The following British outlets were selected: BBC, the Independent, Daily Mail, the Guardian, the Telegraph, the Financial Times, the Daily Mirror, the Huffington Post (UK). The Dutch news outlets that were studied were: de Telegraaf, Algemeen Dagblad, NU.nl, Volkskrant and Trouw. The English newspapers the Times and the Sun and Dutch newspaper NRC were left out because these newspapers required a paid subscription to enter the online database. In total 200 articles were selected. During the analysis 12 articles seemed to not meet the selection criteria, therefore n=188. Table 1 shows a specific overview of the units of analysis. This table can be found in appendix 3.

Data analysis

The Atlas.ti software was used to structure and code the data. All articles were coded according to a codebook. The first step in this coding process was to define the initial categories based on the theories that were relevant for this study. After the initial categories were defined, the further coding happened through an open coding process, this means that during the coding procedure new codes emerged. In order to analyze all messages a codebook was created as guideline to label all relevant data. The complete codebook is attached in appendix 4.

After this qualitative analysis in Atlas.ti, an additional quantitative analysis was conducted. All the data from Atlas.ti were exported to Excel, in order to create a list of all codes that occurred in the different groups (Primark press releases, UK news, Dutch news and UK blogs and Dutch blogs). This Excel file was exported to SPSS so that a second analysis could be performed. A one-way between subjects ANOVA was conducted to compare the mean differences of the variables between the groups. Fisher's least significant difference (LSD) was used as post-hoc analysis. The variable ‘groups’ was measured as independent variable, and the remainder variables were the codes that were used in the qualitative analysis (e.g. frames and response

strategies), these were measured as dependent variables. All the measures were tested on a 95% confidence interval.

Operationalization of concepts: codes and categories

The variables that were used in this study are source (press release, news release, blog), year, country of origin, actors and issues, frames, tone, response strategies and reaction of the public. These are the variables that were coded in the analysis. The actors and issues that were

identified are the key players in the crises. This study identifies several actors that were involved in both crises; these actors are Primark, the government, the industry, NGO’s and protest groups. The issues that are analyzed are the two crises; the collapse of the Rana Plaza factory and the help label incident. The frames that were used in this study were based on the frames that An and Gower (2009) used in their study on crisis communication. They used five frames that according to Neuman, Just and Crigler (1992) are the most dominantly used frames in U.S. news coverage. These five frames are human interest, conflict, morality, economic consequences, and attribution of responsibility. The response strategies that will be analyzed in all messages are taking action, taking responsibility, denial and apology. These are a few common strategies that are identified by e.g. Coombs, 1995; Coombs, 2006; Coombs,

Frandsen, Holladay & Johansen, 2010. In order to analyze reputational impact of the crises, the reaction of the public is also analyzed. This study looks at whether the public reacts favorable, unfavorable or wants to take action towards Primark. This analysis also looked at the tone of the message. Tone is a very subjective variable that is hard to measure. However, this study tried to look at whether there was a positive or negative tone towards Primark. Messages that were considered to be neutral were not coded.

Findings

This chapter discusses the results of the qualitative and quantitative content analysis. The differences between the groups, Primark, UK news media, Dutch News media, UK blogs and Dutch blogs are discussed. A concept indicator model was created to give an overview of the findings. A frequency table of the occurrence of all codes is provided in appendix 3.

Actors and issues Actors

In all articles (including news articles and blogs), Primark was 179 times found as an actor, this is in 95% of the articles. Actor Primark was found to have a small statistical significance between groups, as a one-way ANOVA showed an exact value of p=.05, F (4,74) = 2.42, p=.05). A post hoc test showed a small significant difference (p=.004) between Dutch news and UK blogs.Other retailers were also found to be actors, the word ‘retailer(s)’ occurred in total 162 times. Primark works together with other actors in order to improve the conditions in the garment factories. The garment industry, NGO’s and the Bangladeshi government were also mentioned as actors that were involved in “solving” the Rana Plaza crisis by providing aid to wounded victims and relatives of the victims that died.

The word ‘industry’ occurred 135 times, and was found to be an actor in 31 articles. A one-way ANOVA showed significant differences for actor industry between groups, (F (4,74) = 8.74, p=.000). A post-hoc LSD test showed that these significant differences occurred between various groups: between blogs and Primark, between UK news releases and Primark, and between Dutch news releases and UK blogs. The word ‘government’ occurred 72 times, but was only found as an actor in 7 articles. A one-way ANOVA did not indicate any significant differences between groups (F (4,74) = .34, p=.85). Another actor that frequently occurred in relation with Primark were NGO’s. The word ‘NGO’s’ appeared 37 times, and was found as an

actor in 29 articles. A one-way ANOVA showed for actor NGO’s (F (4,74) = 4.77, p=.001). Significant differences were found between UK news and blogs and between the Dutch news and Primark’s press releases. The last group that occurred as actors are protesters or protest groups, they were actors in 25 articles. A one-way ANOVA did not indicate any significant differences between the groups (F (4,74) = 1.35, p=.26).

Issues

The Rana Plaza issue had a higher position on the agenda than the help label issue. Primark released 16 press releases about the Rana Plaza incident and 3 press releases about the help label issue. In total, the Rana Plaza was found as an issue in 147 articles, whereas the help label issue was found to be an issue in 63 articles. The Rana Plaza issue was also frequently

mentioned in articles that discussed the help label.

For the Rana Plaza issue, a one-way ANOVA indicated significant differences between groups (F (4,74) = 5.55, p=.000). A LSD post-hoc test showed that the differences occurred between the following groups: UK news and Dutch news, UK news and Dutch blogs, Dutch news and UK Blogs, Primark and UK blogs and between Primark and UK news.

For the help label issue, a one-way ANOVA found significant differences between groups (F (4,74) = 4.45, p=.002). The post-hoc test indicated significant differences between Dutch blogs and UK news, Dutch blogs and Dutch news, UK blogs and Dutch blogs and between Primark and Dutch blogs.

Response strategies Primark

Primark’s response strategies were found to different in both crises. Their responses to the collapse of the Rana Plaza were apology, taking responsibility and taking action. Taking action is the most frequently used strategy in Primark’s press releases (used 22 times); this strategy is

also often mentioned in the news and in blogs. In each press release about the Rana Plaza incident Primark mentions the actions they will take. Words like ‘pay/ paid’ (221 times), ‘compensation’ (250 times), ‘improve’ (46 times), ‘inspect/ inspections’ (65 times),

‘investigations/ investigation/ investigate/ investigated’ (85 times) all relate to the actions that were taken by Primark. A one-way ANOVA showed significant results between groups for the use of the response strategy taking action (F (4,74) = 9.31, p=.000). Those differences occurred between UK blogs and all other groups.

Primark does not only provide help to victims, they also want to improve safety of the factories. An example of this kind of action mentioned in one of the press releases is signing the Accord on Fire and Building Safety in Bangladesh, initiated by the IndustriALL and UNI Global Unions. Primark states: “under the terms of this agreement, signatories commit to financing and implementing a joint fire and building safety programme. This will include safety inspections, remediation and fire safety training at supplier factories.”

Primark takes responsibility for the Rana Plaza crisis. Primark’s response strategy taking responsibility occurred in total 52 times (9 times in Primark’s press releases). The words ‘responsible/responsibility’ occurred 82 times in total. Primark immediately took responsibility for the Rana Plaza collapse. In one of the first press releases they state: “Primark confirms that one of its suppliers occupied the second floor of the eight storey building, which housed

several suppliers to the garment industry making clothing for a number of brands.” A one-way ANOVA indicated significant differences between groups (F (4,74) = 5.39, p=.000). A LSD post-hoc test showed that these differences occurred between UK blogs and Dutch news, UK blogs and Dutch news, UK blogs and Primark, UK news and Dutch blogs, Dutch news and Dutch blogs, Dutch blogs and Primark.

Primark also apologizes for what happened to the Rana Plaza. This strategy is only used a few times. They do not directly say ‘sorry’, but for example in one press release they state: "the company is shocked and deeply saddened by this appalling incident at Savar, near Dhaka, and expresses its condolences to all of those involved”. The following words are related to the apology strategy: ‘sorry’ (4 times), ‘apologizing’ (2 times), ‘condolences’ (10 times). These word frequencies are low. A one-way ANOVA indicated significant differences between groups (F (4,74) = 3.46, p=.009). These differences were found between UK news and UK blogs and between UK news and Dutch blogs.

The reaction towards the help label incident is the opposite, as Primark uses denial. Primark does take action by investigating the origins of the labels, but in the end they deny

responsibility by claiming the labels are a hoax. In a press release Primark states: “our investigation into the labels sewn onto two garments bought separately from our Swansea store in 2013, has led us to the conclusion that it is more likely than not, to have been a hoax carried out in the UK.” In all articles the words ‘deny/ denial/ denies/ denied’ occurred 21 times. A one-way ANOVA did not find significant results between groups (F (4,74) = .91, p=.46).

Primark and framing

The frame that occurred the most in Primark’s press releases is the morality frame (used 11 times). The second most used frame by Primark is the economic consequences frame (used 10 times), which is used to stress the economic actions that were taken by the company. The human-interest frame was used in 8 press releases. The attribution of responsibility frame occurred 5 times.

Primark uses the morality frame as a way to show that they are a morally right (and ethical) company.This frame is used in all their press releases. A one-way ANOVA showed a significant difference of the use of the morality frame between groups (F (4,74) = 11.59, p=.000). A LSD post- hoc test indicated that all p-values of Primark compared to the other groups are significant, since p=.000 for all groups. This indicates that Primark used the morality frame more than the news media and public.Primark expresses its morality by discussing the actions they take after the Rana Plaza incident. Primark also stresses that they are involved with local communities and the safety of the factories. They show their morality by for example talking about their ethical trading: “our Code of Conduct states that suppliers and factories must ensure products are made in good working conditions, and that the people making them are treated decently and paid a fair wage. We inspect each factory to ensure it is meeting the Code and support it by providing guidance and training when issues are identified and are a member of the Ethical Trading Initiative.”

The attribution of responsibility frame is also clearly visible in Primark’s press releases. In the first press release after the collapse of the factory they stated: “Primark confirms that one of its suppliers occupied the second floor of the eight storey building, which housed several

suppliers to the garment industry making clothing for a number of brands”. Primark shows its responsibility by calling other brands to take action, and by also paying compensation to victims that worked for other brands. Primark says: “Primark is calling on other brands involved in the Rana Plaza disaster to make a contribution by paying short-term aid to some 3,600 workers (or their dependents) who made clothes for their labels. To date these workers have been supported by Primark, even though the workers involved did not make clothing for the company but for other brands”. A one-way ANOVA showed a significant difference of the use of the attribution of responsibility frame between groups (F (4,74) = 10.85, p=.000). A

LSD post- hoc test shows that all p-values of Primark compared to the other groups are significantly different, since p=.000 for all groups.

Framing in the news media and in public blogs

Attribution of responsibility is by far the most dominant frame used by the news media and in blogs. It appeared 58 times in the news media, and 66 times in public blogs. Not only Primark, but the entire clothing industry is held responsible for the bad working conditions in the factories in order to produce cheap fashion. The news media hold various actors responsible. The following paragraph in the Huffington Post makes this very clear: "Factory owners show no responsibility for the very poor working conditions in Bangladesh, the government of Bangladesh is poor at enforcing their own weak laws about safety, and the international brands must also bear some responsibilities. Primarily, it is them putting extreme pressure on local factory owners to produce at a very low price. And because of that, factory owners have to slash many other areas to keep their profits. In the garment sector, the brands determine the price; if you cannot match it they go elsewhere. The price they want is so low it can barely cover the cost. So owners try to cut down on other costs, the wages, fixing problems in the factory. This all contributes to the situations like what we saw at Rana Plaza. It was not the first case and will not be the last one". Public blogs and protest groups also frequently use the attribution of responsibility frame in the same way as the news media. The public also holds various groups responsible. A one-way ANOVA shows a significant difference of the use of the attribution of responsibility frame between groups (F (4,74) = 10.85, p=.000). Post Hoc comparisons using LSD showed that Primark significantly differed from all groups.

The economic consequences frame is also frequently used in relation to the Rana Plaza incident. This frame is used when Primark’s economic compensation is stressed, and what economic consequences it has for the company and the victims. In the news media this frame

occurred 16 times, and it was identified in 10 blogs. A one-way ANOVA showed significant differences in the use of the economic consequences frame comparing Primark’s press releases to the public blogs (F (4,74) = 4.58, p=.002). There is no significant difference between other groups. Many news media articles discuss Primark’s economic actions to compensate the victims. The Telegraph for example reports: “British clothing firm Primark says it will pay a further $9 million (£5.4m) in long-term compensation to victims of the collapse of the Rana Plaza factory in Bangladesh. The compensation, announced on Tuesday, March 18, is being paid to the 580 employees of Primark supplier, New Wave Bottoms, which occupied the second floor of the eight-story structure. A further $1 million (£600,000) is being made to workers in the supply chain of competitors and will be paid to a trust fund run by the International Labour Organization”. The economic consequences frame is mostly used by Primark and the news media, and not so often by the public.

The public and news media also use the human-interest frame in order to provoke emotions. This frame is not used in Primark’s press releases. A one-way ANOVA indicated significant differences between groups (F (4,74) = 2.94, p=.022). A post-hoc LSD test showed that these differences occurred between Primark and the public blogs. There were no significant

differences between the other groups.

The conflict frame is the least used frame. In total it occurred 18 times. This frame was used when articles talked about a conflict between Primark and other groups, for example when protest groups were in conflict with Primark. A one-way ANOVA did not show significant results for differences between the groups (F (4,74) = 1.03, p=.393).

Tone in the news media and in blogs

The tone in the news media is most of the time quite neutral, whereas blogs expressed more negative tones in their messages. In the news media and blogs, the negative tone was visible in 73 messages, and the positive tone was apparent in 17 messages. A negative tone was for example coupled to words like ‘sweatshop’ (75 times), ‘shocking / shock/ shocked’ (61 times), ‘victims’ (265 times), ‘death(s) / dead’ (154 times). ‘slavery’ (9 times) in relation to Primark. A positive tone was for example used when the words ‘positive’ (20 times), ‘initiative’ (45 times), or ‘good’ (45 times) were mentioned in relation to Primark.

The tone is somewhat positive when Primark’s actions are compared with other companies. The Huffington Post for example quotes Oxfam's Bangladesh Country Director, Gareth Price Jones, he said: "Primark's decision to compensate survivors and families who lost loved ones in the Bangladesh Savar building collapse is welcome - other companies who had premises in the building should follow their example”. For a positive tone there were no significant differences found between groups (F (4,74) = .61, p=.66).

A negative tone was more often used in blogs than in the news media. A one-way ANOVA found significant differences between groups (F (4,74) = 10.47, p=.000). A LSD post hoc test indicated that these differences occurred between UK news and Dutch news, UK news and Primark, Dutch news and UK blogs, Dutch news and Dutch blogs, UK blogs and Primark, and between Dutch news and Primark.A negative tone is most of the time used to describe the bad conditions in the factories, which is also blamed on the clothing companies. One public

blogger from the UK writes for example “Primark, the UK clothing store often criticised for its use of ‘sweatshop labour’, has emerged as something of a white knight following the collapse of compensation talks surrounding April’s Rana Plaza factory disaster. The UK brand, owned by Associated British Foods, was the only one to agree to pay more short-term

aid at the talks in Geneva, to which only nine out of the 29 invited brands turned up”. Another blogger is maybe even more negative, by stating that: “Primark has worked over the years to improve on its sustainability image, but it isn’t about to shift its image as the bad boy of the high-street. This is an image build up over years, the knowledge that there has to be a sacrifice somewhere for the cheap prices that consumers desire. And when disasters like Rana Plaza happen, they highlight that no matter how shiny your sustainability plan, there are major problems“.

There are also examples of the news media using a negative tone towards Primark. The Huffington post writes for example: "if Primark had taken its responsibility to those workers seriously, no one need have died this week". The Daily Mirror is also quite negative by saying that “the company faced criticism for failing to carry out structural tests on the factories and for using low-cost suppliers”.

Reaction of the public

The reactions of the public towards Primark were diverse. There were many unfavorable reactions regarding the bad conditions in the Bangladeshi factories. But on the other hand part of the public reacts favorable to the fact that Primark takes action to financially compensate the victims. There are also reactions of protest groups to take action. A one-way ANOVA found significant differences between the groups for a unfavorable reaction (F (4,74) = 6.51, p=.000), and for the reaction take action (F (4,74) = 4.82, p=.001).

A LSD post-hoc test revealed that the unfavorable reactions significantly differed between UK News and Dutch news, Dutch news and UK blogs and between Dutch news and Dutch blogs. A quote that is often repeated is a quote from Rebecca Gallagher, who reacts unfavorable to the help label incident: “I was amazed when I checked for the washing instructions and spotted

this label. To be honest I’ve never really thought much about how the clothes are made. But this really made me think about how we get our cheap fashion. I dread to think that my summer top may be made by some exhausted person toiling away for hours in some sweatshop

abroad.”

Another reaction is a call for action; this is done by various protest groups. A LSD post-hoc test revealed that the reactions about taking action significantly differed between UK news and Dutch news, Dutch news and UK blogs, and between UK news and UK blogs.These actions vary from boycotting Primark or actually participating in public protests in front of Primark stores. The word ‘protest’ is used 48 times, and the word boycott(ting) is used 33 times in all articles. One common reaction of the public and news media is that people should carefully think about the origin of their clothing and in which circumstances they are produced. There are also a few public blogs that stress that people should not boycott Primark or other clothing stores. The following blog author argues that this can have negative consequences for the people that work in the Bangladeshi garment industry: “We zouden allemaal kunnen stoppen met het kopen van goedkope kleding: maar een morele algemene boycott zal niet veel goeds betekenen voor mensen in ontwikkelingslanden die hun banen als gevolg hiervan zullen verliezen en voor diegenen in ontwikkelde landen waar het kunnen kopen van een jeans van 5 euro het verschil maakt tussen armoede en het hebben van een klein beetje materieel geluk”.

There are also reactions in which it becomes clear that not everybody cares. Poor people and young girls like fashionable clothes, and do not care about the origin and bad circumstances their clothes are produced. One public blog reports about a survey that was conducted amongst the public: “Fashion forecaster Mintel conducted a survey with over 2,000 shoppers aged 16 and above, 3 months after 1,129 people were killed in the Rana plaza building collapse, to assess their opinions on ethics in fashion. The results were dismal – only 44% of women and

38% of men considered the treatment of workers as top priority. ‘When asked what was important when making a purchase, the treatment of workers was valued as “low” and environmental friendliness was “ranked even lower as a concern for clothes shoppers.”’

UK and Dutch articles

The English news media and also bloggers from the UK report more on both crises than the Dutch news media and bloggers. 58 UK news articles, and only 24 Dutch news articles were found for this study. There were also big differences in the amount of public blogs, 68 public blogs were found, and only 19 Dutch blogs. The ANOVA’s often indicated significant differences between the UK and Dutch messages. As reported above, for most variables significant differences were found between the two countries.

A concept indicator model was created to summarize the findings. This model can be found in appendix 5.

Conclusion

The aim of this study was to investigate how the public, news media and Primark interact in their communication about Primark’s crises by conducting a quantitative and qualitative content analysis. The frequency of communication, the content of communication and the tone of the communication from both Primark as well as the public were investigated. This study found similarities, as well as many differences between Primark, the news media and the public in their reporting on both crises.

General frequencies of communication

Primark only communicated through the 19 press releases that were published on their corporate website. It is remarkable that they only used one communication channel. Primark did not use social media in its crisis communication, and did not participate in a direct dialogue

with its consumers. This study only focused on online news publications, it is therefore impossible to say how often the news media in total communicated about the Rana Plaza and help label crises since paper newspapers were not taken into account. This study analyzed 82 news messages, but only from selected news outlets, since not all media outlets granted free access to their articles. In total, 87 blogs were analyzed. This does not mean that only 87 blogs regarding this issue were published. The blogs were selected through the highest SEO on Google, this means that not all public blogs on the Internet could be analyzed. However, it can be concluded that the UK news media, and also the UK blogs paid more attention to Primark and the crises.

Primark’s content of communication

The findings of this content analysis show that overall Primark did a good job with its responses by showing responsibility and taking action during the Rana Plaza crisis. Primark confirmed to be an actor that was involved in both crises. The company immediately decided to take action and pay compensation to victims. Primark clearly showed its corporate social responsibility (CSR) efforts in their communication. They even launched a website

http://www.primark-bangladesh.com to show what they did after the Rana Plaza incident. The

attribution of responsibility frame is also often used in Primark’s press releases.Previous research (e.g. Lee, 2004) confirmed that acceptance of responsibility of a crisis can increase sympathy and forgiveness among the public. According to Coombs and Holladay (2008), taking action by compensating victims is as effective as an apology, because it shapes the public perception of the organization taking responsibility for the crisis because they show that they focus on the wellbeing of the victims. These accommodative strategies also proved to be successful for Primark. An experimental study by Pace, Fediuk and Botero (2010) showed a positive relationship between accepting responsibility, showing regret or apology and reputation. Accepting responsibility is not just apologizing, but companies need to explicitly

state that they accept responsibility and show their regret. This can reduce the amount of anger of stakeholders and to leads to reputation protection. The findings of this study by Pace et. al (2010) show the importance of the content of crisis response messages.

Because Primark immediately took responsibility and promised to take action, the news media and the public were in general not so negative towards Primark as they were towards other actors (e.g. Carrefour, JC Penny and Matalan) that were also involved in the Rana Plaza accident and did not take action to help the victims. Primark not only said they would help the victims; they were also the first to actually provide money and other forms of emergency-aid to the victims. This positively contributed to the protection of Primark’s reputation.

However, Primark denied involvement in the help label issue. Primark claimed that it was a hoax. This reaction evoked more negative reactions among the public. Especially protest blogs reacted negative, by highlighting the bad working conditions in factories in developing

countries. The different response strategies indeed show that accommodative strategies work better than low accommodative strategies such as denial and shifting the blame.

The news media

In general, it was remarkable that there were more news reports on Rana plaza than on the help label issue. This could indicate that direct human suffering was considered to be more

important on the news agenda. The news media mainly focused on the attribution of

responsibility of both crises. An and Gower (2009) argue that it is logical that when the news media reports about a crisis, they tend to assign responsibility to organizations for a crisis, especially when it concerns coverage of preventable crises where organizational misdeed is the case.

Overall, Primark was not portrayed in a very negative way. Just like Primark’s press releases and the public blogs, the news media also stressed that Primark took action to help the victims. Since Primark was one of the first companies that admitted they were involved and promised to take action, the company was not as bad portrayed as other companies in the news media.

Reaction of the public

Just like the news media, the attribution of responsibility frame also occurred the most often in blogs. This indicates that the aspect of responsibility was the most important aspect in

reporting on both crises.

The human-interest frame is mostly used by protest groups who want to stress human suffering by focusing on the bad conditions in the Bangladeshi garment factories. This frame stresses death and suffering of victims and portrays Primark and the other actors that were involved with the Rana Plaza crisis in a negative way. The human-interest frame is most of the time coupled with a negative tone. The fact that people work in bad circumstances and thousands of people died because of this provoked many unfavorable reactions. The unfavorable reactions are towards the entire Bangladeshi garment industry but also towards the cheap fashion chains. Primark is also an actor that is being held responsible for this.

Unfavorable reactions often occurred in combination with a negative tone. A negative tone was often used when articles used the quotes from the women who found cry for help labels in their Primark clothes. In general people reacted favorable to Primark’s effort to compensate the victims of the Rana Plaza collapse, the actions taken by Primark were discussed in the majority of blogs. Kim and Cameron (2011) found that the use of emotional news in the media leads to more emotional and negative responses of the public. This means, that the use of

human-interest frame in the news could have affected the emotions of the public, which in turn led to a more negative evaluation of Primark.

Despite a negative tone that appeared in more than half of the public messages, on July 11th the Guardian reported a 20% increase in sales over the past 3 months. This indicates that despite the Rana Plaza accident, consumers decided not to boycott the retailer (Guardian, 2013). The company claims the sales have increased because of changes in the weather. But it can be concluded that both crises did not influence Primark’s sales. However, it shows that the unfavorable reactions of the public in blogs do not represent the reactions of the general consumers. On the other hand, it could also be the case that British people do not care about the origin of their cheap clothes.

Previous studies (e.g. Pennings, Wansink & Meulenberg, 2002; Lee, 2004) found that crises that pose direct threats to stakeholders tend to lead to more negative evaluations of an organization than when a crisis does not directly harm stakeholders. Both Primark crises did not directly harm consumers, which leads to less negative evaluations from the consumer perspective. Primark’s response strategies also show that they did a good job in their crisis communication, and that Primark’s legitimacy and reputational damage does not significantly affect their sales.

Discussion

The key contribution of this paper was to deepen the understanding of crisis response strategies, how Primark interacts with the audience, how the audience responds, and what affect this has on corporate reputation and legitimacy.

Implications

The practical implications of this study are confirmed by previous research, accommodative response strategies were found to be effective in crisis communication. The recent crises in which Primark was involved also show how hard it is for organizations to claim that their products are “ethically sourced”. Primark used its corporate social responsibility efforts as a way to show their morality. The company showed that expressing efforts of taking action are important in crisis communication as tool to protect organizational reputation.

Limitations and future research

This research found insights in the communication between company, news media and the public, but significant research gaps remain. A first limitation is that this study analyzed blogs, but blogs do not represent the general public opinion, therefore it is impossible to generalize what the public actually thinks about Primark’s crisis communication. Further experimental research is necessary, so that perceptions of the public regarding different response strategies can be directly measured. This type of research can help to draw conclusions on reputational effects of different crisis response strategies. Future research could also focus on how crisis communication and reputation after a crisis affect consumer behavior by conducting

experimental or survey research that measures consumer behavior. A second limitation is the sampling in this study. Not all messages from news outlets could be analyzed due to a paid subscription. Blogs were found through their SEO, which is not the most reliable way to select a sample. Another limitation is that social media were not analyzed in this study. Especially the public uses social media as an outlet for their personal opinion. Future research should focus on comparing social media messages from the company, news media and the public in order to find out what effect communicating through social media has on crisis communication and corporate reputation.

References

An, S.K., & Gower, K.K. (2009). How do the news media frame crises? A content analysis of crisis news coverage. Public Relations Review 35, 107–112.

Boin, A., 't Hart, P., & McConnell, A. (2009). Crisis exploitation: political and policy impacts of framing contests. Journal of European Public Policy, 16(1), 81-106.

Carroll, C.E., & McCombs, M. (2003). Agenda-setting Effects of Business News on the Public’s Images and Opinions about Major Corporations. Corporate Reputation Review, 6, 36–46.

Castello, I., Morsing, M., & Schultz, F. (2013). The communicative construction of corporate social responsibility in network societies: A mediation perspective. Journal of Business Ethics, 118, 683–694.

Christensen, L. T., Morsing, M., & Cheney, G. (2008). Corporate communications: Convention, complexity, and critique. London: SAGE Publications Ltd.

Coombs, W.T. (1995). Choosing the Right Words: The Development of Guidelines for the Selection of the ''Appropriate'' Crisis-Response Strategies. Management

Communication Quarterly, 8(4), 447-476.

Coombs, W.T. (2004). Impact of Past Crises on Current Crisis Communication: Insights From Situational Crisis Communication Theory. Journal of Business Communication, 41(3), 265-289.

Coombs, W. T. (2006). The protective powers of crisis response strategies: Managing reputational assets during a crisis. Journal of Promotion Management, 12, 241-260. Coombs, W.T. (2007). Protecting Organization Reputations During a Crisis: The Development

and Application of Situational Crisis Communication Theory. Corporate Reputation Review,10(3),163–176.

Coombs, W.T., & Holladay, S.J. (1996). Communication and attributions in a crisis: An experimental study of crisis communication. Journal of Public Relations Research, 8, 279–295.

Coombs, W. T., & Holladay, S. J. (2007). The negative communication dynamic. Exploring the impact of stakeholder affect on behavioral intentions. Journal of Communication Management, 11(4), 300–312.

Coombs, W.T., & Holladay, S.J. (2008). Comparing apology to equivalent crisis response strategies: Clarifying apology's role and value in crisis communication. Public Relations Review, 34(3), 252–257.

Coombs, W.T., Frandsen, F., Holladay, S.J., & Johansen, W. (2010). Why a concern for apologia and crisis communication? Corporate Communications: An International Journal, 15(4), 337-349.

Deephouse, D.L., Carroll, C.E. & McCombs, M.E. (2001). The role of newsroom bias and corporate ownership on the coverage of commercial banks in the daily print media’. Paper presented at the 5th International Conference on Corporate Reputation, Identity, and Competiveness, Paris.

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

Gray, E.R., & Balmer, J.M.T. (1998). Managing Corporate Image and Corporate Reputation. Long Range Planning, 31(5), 695–702.

Hallahan, K. (1999) Seven Models of Framing: Implications for Public Relations. Journal of Public Relations Research, 11(3), 205-242.

Jones, B., Temperley,J., & Lima, A. (2009). Corporate reputation in the era of Web 2.0: the case of Primark. Journal of Marketing Management, 25(9), 927-939.

Kim, H. J., & Cameron, G. T. (2011). Emotions matter in crisis: The role of anger and sadness in the publics' response to crisis news framing and corporate crisis response.

Communication Research, 38, 826–855.

Lee, B.K. (2004). Audience-Oriented Approach to Crisis Communication: A Study of Hong Kong’s Consumers’ Evaluation of an Organizational Crisis. Communication Research, 31(5), 600-618.

Lindlof, T., & Taylor, B. (2011). Qualitative Communication Research Methods. London: Sage.

Macnamara, J. (2010). ‘Emergent’ media and public communication: understanding the changing mediascape. Public Communication Review, 1(2), 3-17.

Neuman, W. R., Just, M. R., & Crigler, A. N. (1992). Common knowledge. Chicago: University of Chicago Press.

Neville, S. (2013, July 11). Primark sales rise 20% despite Bangladesh factory disaster backlash. The Guardian. Retrieved from:

http://www.theguardian.com/business/2013/jul/11/primark-sales-rise-despite-bangladesh-factory

Pace, K.M., Fediuk, T.A,, & Botero, I.C. (2010). The acceptance of responsibility and expressions of regret in organizational apologies after a transgression. Corporate Communications: An International Journal, 15(4), 410 – 427.

Pattriotta, G., Gond, J.P., & Schultz, F. (2011). Maintaining Legitimacy: Controversies, Orders of Worth and Public Justifications. Journal of Management Studies,48(8), 1804-1836. Payne, L. L. (2006). Synthesizing crisis communication and reputation management: an

experimental examination of memory. Journal of Promotion Management, 12(3-4), 161-187.

Pearson, C., & Clair, J. (1998). Reframing crisis management. Academy of Management Review, 23(1), 59-76.

Pennings, J.M.E, Wansink, B., & Meulenberg. M.T.G. (2002). A note on modeling consumer reactions to a crisis: The case of the mad cow disease. Internal Journal of Research in Marketing 19, 91–100.

Primark. (2014). Our Ethics. Retrieved from: http://www.primark.com/en/our-ethics

Scherer, A., & Palazzo, G. (2006). Corporate Legitimacy as Deliberation: A Communicative Framework. Journal of Business Ethics, 66, 71-88.

Semetko, H.A. & Valkenburg, P.M. (2000). Framing European politics: A content analysis of press and television news. Journal of Communication,50(2), 93-109.

Schultz, F., Utz, S., & Goritz, A. (2011). Is the medium the message? Perceptions of and reactions to crisis communication via Twitter, blogs and traditional media. Public Relations Review, 37, 20-27.

Schultz, F., Kleinnijenhuis, J., Oegema, D., Utz, S., & van Atteveldt, W. (2012). Strategic Framing in the BP crisis: A semantic network analysis of associative frames. Public Relations Review, 38(1), 97-107.

Thelwall, M., & Stuart, D. (2007). RUOK? Blogging communication technologies during crises. Journal of Computer-Mediated Communication, 12, 523-548.

Vanhamme, J., & Grobben, B. (2009). ‘‘Too Good to be True!’’. The Effectiveness of CSR History in Countering Negative Publicity. Journal of Business Ethics, 85, 273–283. Wartick, S.L. (1992). The relationship between intense media exposure and change in

corporate reputation. Business Society, 31, 33–49.

Weaver, D.H. (2007). Thoughts on Agenda Setting, Framing, and Priming. Journal of Communication, 57, 142–147.

Appendices

This is an example of a cry for help label that was found in Primark clothing.

Appendix 2 Model 1

An overview of the comparison in this study.

Appendix 3 Table 1

Units of analysis per group.

Source Number of articles Press releases 19

News papers UK 58 News Papers NL 24 Protest blogs 31 Public blogs 56 Total (N) 188 Appendix 4 Codebook Source:

- Press release Primark - Newspaper

- Other news media - Public blog - Protest blog Country of origin: - UK - NL Year: - 2013 - 2014 Issue: - Rana Plaza

- Help label Actor: - Government - Manufacturing industry - Primark - Protest groups - NGO’s Frames: - Human Interest - Conflict - Morality - Economic consequences - Attribution of responsibility Tone: - Negative - Positive

Response from Primark: - Apology

- Denial

- Taking responsibility - Taking action

Reaction public: - Favorable - Unfavorable - Take action - Don’t care - No Boycott

How is Primark portrayed in the message? - Positive

- Negative

Appendix 3 Table 2

Frequency table of all codes

Code Amount

Actor Government 7

Actor: Industry 31

Actor: NGO’s 29

Actor: Primark 179

Actor: Protest group 25 Frame: Attribution of responsibility 129

Frame: Conflict 18

Frame: Economic consequences 36 Frame: Human interest 37

Frame: Morality 23

Issue: Rana Plaza 147

Origin: NL 45

Origin: UK 120

Primark: Apology 10

Primark: Denial 36

Primark: Taking action 139 Primark: Taking responsibility 52 Reaction public: Don’t care 9 Reaction public: Favorable 5 Reaction public: No boycott 6 Reaction public: Take action 33 Reaction public: Unfavorable 57

Source: Newspaper 69

Source: Other news media 9

Source: Primark PR 19

Source: Protest blog 31

Source: Public blog 56

Tone: Negative 73

Tone: Positive 17

Year: 2013 88

Appendix 5 Model 2