Framing the BP Deepwater Horizon Oil Spill
A Comparative Analysis Between
Social Advocacy Organizations, Multinationals, and the Media
Erik Lauret (10876219) Master Thesis
Graduate School of Communication Master’s Programme Communication Science
dr. G.L.A. Toni van der Meer October 11, 2016
ABSTRACT
This paper examined whether there are significant framing differences between social
advocacy organizations, multinationals, and the US-media. A case study was conducted of the BP Deepwater Horizon oil spill, with the key actors BP, Greenpeace, and two newspaper outlets: the Washington Post and the New York times. The analysis was aimed on two measures of framing, namely generic frames and frame functions. Results showed that there are significant differences between the media and multinationals and social advocacy organizations, yet no significant differences were found between social advocacy organizations and multinationals. The significant differences found concerning generic framing specifically highlighted differences amongst the US-media sources for the human-interest frame. Further differences were found concerning the conflict frame and the economic frame; with The New York Times significantly differing in these framing efforts from Greenpeace, BP, and The Washington Post. Framing functions similarly highlighted significant differences with BP and The New York Times focusing on treatment
recommendation, Greenpeace on moral definition, and The Washington Post highlighting causal intepretation. It was further hypothesized that there would be significant frame differences between multinationals and social advocacy organizations, and frame alignment between social advocacy organizations and the media. Results however, highlighted that significant frame differences nor significant frame alignment was found.
INTRODUCTION
Similar to individuals, organizations are part of society and have obligations to support society (Seeger, Sellnow & Ulmer, 2003), often requiring businesses to look beyond their technical or financial interests (Carrol, 1991). In accordance, when an organizational crisis occurs, it doesn’t only directly and indirectly affect its members and those dependent on its inputs and outputs, but also society at large (Milburn, Schuler & Watman, 1983).
Characterized by rapidly succeeding emotional and stressful events (Weick, 1988),
organizational crises continuously pose significant threats to the reputation of the organization and in turn change how stakeholders interact with the organization (Coombs, 2007).
People seek information through the media to evaluate the cause of the crisis as well as the organizational responsibility (An & Gower, 2009). In realizing that the scope and impact of a crisis on organizational reputation affects how stakeholders interact with an organization (Coombs, 2007), the media holds significant power in having the potential to strategically spread crisis related information via framing and agenda setting (Patriotta, Gond & Schultz, 2011). Creating not only saliency, the media define and limit meaning to stories and in turn shape people’s interpretation (Hallahan, 1999).
Changes in media landscape have influenced how organizations discuss ideas and issues that are of interest to stakeholders and the organization (Luoma-aho & Vos, 2010). Traditional media, tangible, and new media, virtual, highlight that debates can be held in different media sources and simultaneously emphasize that different actors’ level of activity may similarly depend on media source (Luoma-aho & Vos, 2010). Characterized by
flexibility one medium may dominate initially, yet may be taken over by another medium later on in the discussion (Luoma-aho & Vos, 2010). It is thus of key importance that organizations participate in multiple discussions at the same time, addressing different stakeholders with different levels of activity (Luoma-aho & Vos, 2010). This is especially
important when communicating to change people’s attitudes (Chong & Druckman, 2011). As individuals receive competing frames over an extended period of time, rather than all at the same time (Chong & Druckman, 2011), this may specifically convince different stakeholders to share their opinions more widely and gain support easier.
This paper specifically investigates framing differences during an organizational crisis amongst three key actors, namely (1) the media, (2) social advocacy organizations and (3) multinationals. The crisis investigated is the BP Deep Horizon oil spill of April 2010, highlighting the communication efforts by BP, Greenpeace, and the US media, specifically ‘The New York Times’ and ‘The Washington Post’, whom were all highly involved actors during the crisis. his case serves of specific interest as it identifies stakeholders with vastly differing roles. Namely a responsible party, BP as a multinational, an opposition party, Greenpeace as a social advocacy organization, and an external stakeholder, the US media in a single organizational crisis. In acknowledging this, Chong & Druckman (2007) highlight that issues can have different interpretations through which frame building results in a competition of conflicting frame.
This research aims to contribute to future research by highlighting the interplay
between three independent actors in their framing efforts. Previous research in communication has often focused on framing, yet has not focused significantly on investigate the changing dynamics of a crisis in comparison to multiple actors involved from the beginning to the end. Instead it has highlighted differences mainly in news and political events (de Vreese, 2005;; Semetko & Valkenburg, 2000; Semetko, Valkenburg, & de Vreese, 2005) Thus in
contributing, this research aims to prove its relevance by specifically highlighting the
interplay between opposing stakeholders during a crisis situation and the effects each have on each other’s communication activities. In doing so, this research aims to clarify common
misconceptions about information communicated during crisis situations and identify where these come from.
Through conducting a quantitative content analysis of secondary sources of US newspaper articles from ‘The New York Times’ and ‘The Washington Post’, organizational press releases from BP, and press releases and Facebook posts from Greenpeace, the
following research question will be investigated:
How does the framing during an organizational crisis contrast between profit-driven organizations involved in the crisis, social advocacy organizations, and the media?
THEORETICAL FRAMEWORK
Schultz, et. al. (2011) highlight that previous research on crisis communication has focused mainly on image repair strategies (Benoit, 1995) and crisis response strategies (Coombs, 1998). Beyond this there is a lack of analysis on the interaction between
organizations and stakeholders over time in crisis situations (Schultz, et. al, 2011). This paper aims to contribute to this topic by investigating the interplay of organizational communication during crisis situations by multinationals and social advocacy organizations, whilst
simultaneously investigating the role of the media. More specifically, this research will focus on identifying framing differences in stakeholder communications during crisis situations.
Framing theory highlights that a certain issue can be viewed from various perspectives (Chong & Druckman, 2007). Framing is viable due to communication being a dynamic process (de Vreese, 2005). A frame organizes an idea or story line and provides meaning to events (Gamson & Modigliani, 1987), whilst simultaneously mentally storing clusters of ideas guiding an individuals’ processing of information (Entman, 1993). In specific, framing
communicating (Entman, 1993). Through placing salience on different aspects of a topic (de Vreese, 2005), arguments are constructed for issues, including their causes, evaluations, and solutions. (Entman, 1993). In this way framing acts a process, in which people conceptualize a certain issue and reorganize their thinking on an issue (Chong & Druckman, 1994).
The frames selected by the parties involved in an issue are partially controlled by the formulation of the issue at hand but are also influenced by the communicators’ norms, habits and personal characteristics (Kuhberger, 1998). Luoma-aho & Vos (2014) further
acknowledge that multiple stakeholders have different interests. In realizing this, framing may provide support to the parties involved in an issue by having their preferred interpretation compete and pre-dominate so that others will see the event from a perspective similar to their own (Hallahan, 1999). As crises are often sudden and unexpected events with the potential to pose as severe threats to organizations and disrupt operations (Coombs, 2007), framing may help understand what is going on in organizational crisis situations (van der Meer, 2014).
In the interest of this study, specific emphasis will be placed on generic framing and
frame functions. Generic frames have the ability to surpass time, issues, and space limits
allowing comparisons to be made between frames, topics, and potentially framing practices as well (de Vreese, Peter, & Semetko, 2001). Based on previous research, Semetko & Valkenburg (2000), identified five generic news frames, namely the ‘human interest’ frame, ‘conflict’ frame, ‘morality’ frame, ‘economic consequence’ frame, and ‘attribution of responsibility’ frame. Semetko, Valkenburg, & de Vreese (1999) first define the ‘conflict’ frame as a frame that emphasizes conflict between individuals, groups or institutions. The ‘human-interest’ frame is defined as bringing the story of an individual, or placing an emotional perspective, to an event, issue, or problem (Semetko, Valkenburg, & de Vreese, 1999). The ‘attribution of responsibility’ frame presents an individual, group, or government to be responsible for causing or solving an issue or problem (Semetko, Valkenburg, & de Vreese, 1999). The
‘economic consequence’ frame highlights the economic impact individuals, groups, region, institution, or country will face due to an event, issue, or problem (Semetko, Valkenburg, & de Vreese, 1999), and the ‘morality’ frame, places issues or events in the context of religious tenets and moral prescriptions (An & Gower, 2009)
In relation to frame functions, Entman (1993) highlights that frames have specific functions, namely to ‘promote a particular problem definition, causal interpretation, moral
evaluation, and/or treatment recommendation’ for the issue at hand. Entman (1993) defines problem definition as determining the costs and benefits of what the responsible entity is doing, causal interpretation as identifying the forces creating the issue, moral judgements as
evaluating the responsible entities and their effects, and treatment recommendation as offering and justifying treatments for the issues whilst predicting their likely effects.
Based on this literature, the following two hypotheses have been formulated:
H1: Generic framing will significantly differ depending on the origin of the communications.
H2: Frame functions will significantly differ depending on the origin of the communications.
Even though framing is often used to gain support and diffuse complex problems, certain occasions activate disputants to secure and activate participants (Snow, 1986). By placing focus on activating and securing participants, the disputants integrate interests and points of view that are in accordance with the primary objectives of the issue at hand (Snow, 1986). Perceptions are more important than reality, which highlights that the thought is more important than the act itself (Benoit, 1995). Organizations believed to be responsible for crises start to form unfavorable impressions (Benoit, 1995). Yet, those held responsible are also seen
to have the empowerment to control the outcome (Iyengar, 1990). Responsible parties should thus focus on future problem-solving, with the ultimate aim of preventing the outcome to reoccur (Iyengar, 1990). Based on this the following hypotheses have been formulated:
H3: Profit-driven organizations’ framing of a crisis will focus primarily on treatment recommendation
H4: Profit-driven organizations’ framing of a crisis will focus primarily on human-interest frame.
However, in competitive environments individuals are continually exposed to frames from different parties (Chong & Druckman, 2011). Social movements which have
transformed into organizations through adopting values and entering global communication processes, now fight over shaping perceptions and opinions (Castells, 2007). These counter-power organizations confront the counter-power holders in the communication sphere (Castells, 2007) and are especially successful when there are salient audiences that disapprove of act or events (Benoit, 1995). Chong & Druckman (2011) introduce the term counter-framing, defined as ‘a frame that opposes an earlier effective frame’. These frames should (1) come later in time than the initial frame, (2) advocate an opposing perspective, and (3) have an incentive to counter the initial frame (Chong & Druckman, 2011). Based on this literature, the following hypotheses have been created:
H5: Social advocacy organizations’ framing of a crisis will focus primarily on causal interpretation
H6: Social advocacy organizations’ framing of a crisis will focus primarily on the attribution of responsibility frame
Different stakeholders initially also differ in their framing of the crisis (van der Meer et. al, 2014). However, as frames start to compete with one another and make the
communication surrounding the crisis more complex, consistency regarding the issue is needed to reduce uncertainty and doubt (van der Meer et. al. 2014). Marketing places, which serve as issues arenas becomes specifically relevant here. Marketing places act as places for enactment and facilitating interaction and stakes between different disputants (Luoma-aho & Vos, 2010). A prime example of such is the news media, as it plays a critical role in informing stakeholders (Coombs, 2006). The media does not hold the power, but they have a significant influence on where the power is decided (Castells, 2007). Through having the power to keep certain content absent and shape explicit messages, the media have the power to shape minds (Castells, 2007). Following the influence of the media, stakeholders in turn mutually borrow frames from one another, ultimately leading to an overlap in framing in the domain of the crisis (van der Meer, 2014). Chong & Druckman (2011) further support this notion by highlighting that frames over time cancel out one another and lose their effect. Based on these notions, the following hypotheses have been created:
H7: Framing efforts during the initial stages of a crisis between profit-driven organizations and social advocacy organization will be significantly different.
H8: There will be significant frame alignment between social advocacy organizations and the media
METHODOLOGY
DESIGN
As our aim with this study was to examine media content and organizational communication efforts during a crisis, we have chosen to conduct a quantitative content analysis. According to Berelson (1952) content analysis is "a research technique for the objective, systematic, and quantitative description of the manifest content of communication”, and was therefore the logical research method for our case study. Selected for the content analysis were two newspapers in the United States, namely the The Washington Post and The New York Times, press releases from BP as our multinational actor, and press releases from Greenpeace as our social advocacy actor. Furthermore, Facebook posts were used to supplement the press releases from Greenpeace.
For the creation of the final dataset, news articles, press releases, and the Facebook posts were coded according to codebook that was composed. The codebook examined each type of communication on its basic features, dominant framing perspective(s), and contextual relatability. In conducting our research, one-way MANOVA’s, cross tabulations, and multinomial regressions were used for the statistical analysis.
PROCEDURE
By placing focus on crisis communication as the topic, it was specifically chosen to further contribute to framing research. To further analyze specific framing differences between different actors, the BP Horizon Deepwater oil spill was chosen as our case for this investigation. This case showed clear interactions between different actors, specifically social advocacy organizations and multinationals.
During the offset of events during the crisis, not only did the crisis receive media coverage on a global scale, but also BP as an organization itself. This was specifically relevant as it provided not only significant insights into the framing efforts of the media, but also forced BP to communicate to its stakeholders and wider public, extending the potential for framing analyses. With difficulties in resolving the crisis, did BP not only suffer serious reputational damage but also allowed social advocacy organizations to challenge and act against the organization. In this way, the Deepwater Horizon oil spill served as a particularly suitable case study in that it (1) clearly highlights communication by three various actors, and (2) was time sensitive and thus allowed for changes in framing and frame alignment to occur.
CASE DESCRIPTION
The Deepwater Horizon oil spill marked the largest accidental oil spill in the history of the petroleum industry. Following the exploding and sinking of BP’s Deepwater Horizon oil rig, it was estimated that 210 million gallons of oil discharged in the Gulf of Mexico.
Following 87 days of free flowing oil, the blowout was capped on July 15th, 2010 and finally declared dead on the 19th of September, 2010. Within this time period, massive collective responses were put together yet significant adverse damages were measured along the US coastline. Ultimately, BP was banned from new contracts with the US government, was forced payed $42.2 billion dollars in criminal and civil settlements, and agreed to pay $18.7 billion dollars in fines in July 2015.
There are shared opinions on whether BP communicated successfully or
unsuccessfully during the crisis with its stakeholders (de Wolf & Mejri, 2013). However, it is evident that the crisis received much attention from the media, activists on social media as well as NGO’s throughout its 6-month life period. In highlighting this, the crisis shows the multifaceted scope of crisis communication, placing stakeholders at the center. In
communicating about the crisis, BP focused on direct formal communications through press releases, whilst communications by Greenpeace were both via formal direct channels, in the shape of press press releases, as well as via social media.
After accessing BP.com, press releases were accessed in the press section. Following access to all BP press releases a specific search was conducted between January 1, 2010 and December 1, 2010. Searching within this period of time yielded all of BP’s press releases issued that year. However, only all press releases related to the Deepwater Horizon oil spill between April 20th, 2010 and September 19th, 2010 were selected for investigation; which marked the beginning and the end of the crisis. Specific attention was placed on selecting all press releases related to the crisis such that press releases with political, economic, technical, social, and environmental topics had the potential to be represented. After specifying the time frame, a total of 170 press releases were found to be issued by BP. Placing focus on
Similar to BP.com, press releases from Greenpeace were accessed via the press section on Greenpeace.com. Without further specification possible, 1963 press releases were found released by Greenpeace in since March 2001. After specifically searching for the time frame in which the Deepwater Horizon oil spill occurred, press releases were found specifically relating Greenpeace to the crisis. Again, all press releases were selected relating to the crisis so that political, economic, technical, social, and environmental topics had the potential to be represented.
Furthermore, Facebook was used to retrieve communication efforts by BP and Greenpeace on social media. After accessing Facebook, the international pages of BP and Greenpeace were accessed, which allowed to choose specific years to research within. Interestingly, BP allowed no access to the year 2010. Following this, due to the low count of press releases by Greenpeace, Facebook served as a supplementary platform to specifically analyze their communication efforts to the public.
Articles from the Washington Post and the New York times were found using
LexisNexis. Using the terms ‘Oil spill’, ‘BP’, and ‘Deepwater Horizon’, a total of 235 articles were found within the Washington Post and 336 articles were found within the New York times. It was specifically chosen again to not filter any articles out to make sure that political, economic, technical, social, and environmental topics had to the potential to be represented, whether it be in a societal context or directly related to the crisis.
A non-probability sampling method know as quota sampling was conducted for newspaper articles characterized by months. More specifically, it was decided that 25 articles from each newspaper per month were to be selected to contribute to the research. However, it was found that the months April and September did not have 25 articles in each month, and thus all articles in these months were used. Even though April and September did not meet the minimum article count, articles from these months were still included provided that the crisis started and ended half way through the month as well as marked the most significant events within the crisis.
DATA
The BP Horizon Deepwater oil spill was as previously mentioned, analyzed in two newspapers from the United Stated, from press releases from two different types of organizations, and Facebook. The newspapers selected, The Washington Post and The New York Times, were specifically chosen as they had high number of circulation, which was mainly contributed to the fact that they were both national newspapers. They further served as our data due to the fact that they were both quality newspapers. The press releases were specifically selected from BP and Greenpeace as these organizations were actors highly involved in the crisis. These press releases were taken from both organizations’ international websites, rather than local or regional websites, for the main purpose that press releases would be in English and would cover all significant events during the crisis.
As mentioned previously, all press releases communicated by BP related to the Deepwater Horizon oil spill were used for investigation. The total amount of press releases issued by BP on this crisis totaled 117 between April 20th, 2010 and September 19th, 2010. Similarly, all press releases from Greenpeace related to the crisis were selected within the same time frame; a total of 3 press releases. Facebook postings by Greenpeace relating to the Deepwater Horizon oil spill between April 20th, 2010 and September 19th, 2010 totalled to 7. Ultimately, 10 units of data were gathered from Greenpeace in total about the Deepwater Horizon oil spill, marking a significantly low number of data and thus forming a limitation of this research. A total of 122 newspaper articles were selected for research from the New York Times and 116 newspaper articles were selected from the Washington Post. A random sampling method within the months May, June, July, and August to randomly select 25 articles, whilst, as previously mentioned, all articles within the months April and September were selected.
MEASURES
The variables used to measure framing differences between social advocacy organizations, multinationals, and the media in an organizational crisis were measured in a codebook, which contained 22 questions. The codebook first introduced general guidelines for coding, then presented 8 questions pertaining to general characteristics of the articles, press releases, and Facebook posts. Questions 9 and 10 specifically related to newspaper articles and were asked in to order to identify the prominence of the crisis in news media. Before answering questions 11 to 22, instructions were given to carefully read to article, press release, or Facebook post in order to enhance reliability amongst all data collected.
Origin and type of press
The variables origin and type of press served of specific interest in setting the basis for finding framing differences amongst the key actors within this research. The origin of the data (Q3 in the codebook) was measured by values ranging from 1 to 4. Coders were specifically asked to code where the article, press release, or Facebook post came from, ranging from values (1) the Washington Post, (2) the New York times, (3) British Petroleum, and (4) Greenpeace. An inter-coder reliability test was conducted to test whether the variable was reliable. Krippendorf’s alpha showed that α = 1.00, reflecting it to be coded completely the same amongst the coders. Furthermore, for the variable type of press (Q4) was shown to not have any differences either, highlighting again complete agreement amongst the coders.
Month of publication
Month of publication (Q5) of the articles, press releases, and Facebook posts served
the purpose to identify framing differences amongst the actors and to see whether frame alignment was present between actor at certain stages of the crisis. Month of publication was measured by values ranging from 1 to 12, highlighting each month in the year, with 1 representing January and 12 representing December. After conducting an inter-coder reliability test, Krippendorf’s alpha showed that α = 1.00, highlighting complete agreement amongst the coders.
Frame Functions
The Frame function (Q22) variable was measured to identify whether there were framing differences amongst the actors. The variable was measured by values ranging from 1 to 4, each representing a different frame function to be filled out after reading the whole
article, press release, or Facebook post. The different frame functions were (1) problem definition, (2) causal interpretation, (3) moral evaluation, (4) treatment recommendation, complimented by code 99, which represented the frame function was not identifiable. After conducting an inter-coder-reliability test to assess reliability of the variable, Krippendorf’s alpha was calculated to be α = 0.629. This thus passes minimum requirements and highlights that the variable was sufficiently clear.
Generic Frames
Derived from Semetko & Valkenburg (2000), 20 questions were developed in order to measure which news frames appeared in stories. Semetko & Valkenburg (2000) considered a minimum of 3 questions per item, highlighting 5 news frames in total; the attribution of responsibility frame, human-interest frame, conflict frame, moral frame, and economic frame. For each question used to measure the news frames the coder had to answer either (1) yes or (0) no. The attribution of responsibility frame was measured by 5 questions (Q21A-21E). An inter-coder reliability analysis was conducted to test the reliability of the variable after the 5 questions were computed into the level-of-responsibility frame, showing that Krippendorf’s alpha α = .638, which suggests it meets minimum requirements and thus found clear enough to use for further testing. The human-interest frame was similarly to the attribution of responsibility frame measured by 5 questions (Q21F-21J). After conducting an inter-coder reliability analysis, Krippendorf’s alpha was calculated (α = .636), showing that the variable met minimum requirements and was clear enough. The conflict frame was measured by 4 questions (Q21K-21-N). After calculating Krippendorf’s alpha (α = 1.00), the variable showed to be highly reliable as there was complete agreement between coders. The moral frame, measured by 3 questions (Q21O-21Q) had a calculated Krippendorf’s alpha of α = .015. This means this variable is not reliable, yet may be explained by the low observations of the questions measuring the variable. As the scale has been derived from Semetko &
Valkenburg (2000), it had been chosen to still incorporate the variable for further testing. The final frame, the economic frame was similarly measured by 3 questions (Q21R-21T). With a calculated Krippendorf’s alpha of α = .642, it was found to meet minimum requirements and considered clear enough to be used for further coding.
Furthermore, prior to testing our hypotheses, Cronbach’s alpha tests were conducted to test whether the subscales of the generic frames were internally consistent. The human-interest frame subscale consisted of 5 items (α = .67), proving to be strongly consistent. The conflict-frame subscale consisted of 4 items (α = .71), proving to be strongly consistent. The moral-frame subscale consisted of 3 items (α = .74), proving to be strongly consistent, as well as the economic-frame subscale (α = .72). Interesting to note is that the level-of-responsibility-frame was not internally consistent (α = .015). However, as the variable is of importance to this research in identifying framing differences between actors, it has been decided that the questions ‘Does the Story suggest solution(s) to the problem/issue’ and ‘Does the story suggest that an individual (or group of people in society) is responsible for issue-problem?’ will measure the attribution of responsibility frame. These questions have specifically been chosen as they are most frequently coded yes out of all 5 questions measuring the variable.
RESULTS
Generic Framing
The first analysis conducted was to determine whether the origin of the
communications about the Deepwater Horizon oil spill had a significant effect on what generic frames were used. As previously stated, it is believed that there are significant differences between the actors’ communications and generic frames used. Furthermore, it is
believed that BP will significantly focus on the human-interest frame whilst Greenpeace will significantly focus on the attribution of responsibility frame. A one-way MANOVA was run to determine whether the origin of the communication activities had a significant effect on generic frames used, with origin as the independent variable and the attribution of
responsibility, human-interest, conflict, moral, and economic frame as dependent variables. Looking at the origin and generic frames used it can be seen that the most prominent used frame by Greenpeace is the human-interest frame (M = 1.70, SD = 1.41), followed by the conflict frame (M = 1.20, SD = .918), the attribution of responsibility frame (M = 1.00, SD = .471), the economic frame (M = .100, SD = .316), and the morality frame (M = .000, SD = .000). In relation to BP, it can be seen that the most prominent generic frame in BP’s
communications is the conflict frame (M = 1.00, SD = 1.19), followed by the human-interest frame (M = .940, SD = 1.40), the economic frame (M = .877, SD = .992), the attribution of responsibility frame (M = .769, SD = .592), and the morality frame (M = .068, SD = .340). Looking at the US media, it can be seen that the most dominant generic frame in the Washington Post was the conflict frame (M = 1.23, SD = 1.21,) and in the New York times the economic frame (M = .877, SD = .992).
After conducting the one-way MANOVA, that there was a significant difference in generic frames used depending on the origin of the communications, F (15, 985) = 10.4, p < .001; Wilk’s = .666, partial 2
= .127. Taking an in-depth look at the generic frames themselves, it can be seen that origin of the communication had no significant effect on the attribution of responsibility frame (F (3, 361) = 1.22, p = .302; partial 2
= .010. However, there was a significant relationship between origin and the conflict frame (F (3, 361) = 31,5,
p < .001; partial 2 = 0.208, the human interest frame (F (3, 361) = 4.05, p < 0.01; partial 2 = .033), the morality frame (F (3, 361) = 2.97 p < .024; partial 2
= 0.02), and the economic frame (F (3, 361) = 21,4, p < .001; partial 2
A Tukey post-hoc test indicated that there was a significant difference between usage of the human-interest frame by the Washington Post and the New York times (Mdifference =
-.483, p < .046). In regards to the conflict frame, it can be seen that there are significant differences between the Washington Post and the New York Times (Mdifference = -1.17, p <
.001), BP and the New York Times (Mdifference = - .951, p < .001), and Greenpeace and the
New York Times (Mdifference = -1.14, p < .003). For the economic frame, similar results were
found, namely there are significant differences between the Washington Post and the New York Times (Mdifference = -.661, p < .001), BP and the New York Times (Mdifference = -.671, p <
.001), and Greenpeace and the New York Times (Mdifference = -.777, p < .010). No further
significant differences were found between actors concerning the human-interest, conflict, and economic frame, whilst similarly no significant results found concerning the morality and attribution of responsibility frame.
Additionally, Box’s M was found to be significant F(30, 391192) = 10.6, p < 0.001, showing that the assumption of equal variances in the populations has been violated. This is further supported by Levene’s. For the attribution of responsibility frame F(3, 361) = 3.58, p < .014, the human-interest frame F(3, 361) = 6.34, p < .001, conflict frame Levene’s F(3, 361) = 88.9, p < .001, the morality frame F(3, 361) = 11.5, p < .001, and the economic frame
F(3, 361) = 43.3, p < .001.
Following these results, we can state that we accept hypothesis 1. As previously mentioned, we can highlight that there are significant differences between actors concerning generic framing. In specific, its been highlighted that there is a significant difference between The Washington Post and The New York Times framing efforts of the human-interest frame. Additionally, The Washington Post, BP, and Greenpeace signficantly differ in their framing efforts when considering the conflict frame and the economic consequence frame. Even though we accept hypothesis 1, we reject hypothesis 4 and 6. It was hypothesized that BP’s
generic framing efforts will primarily focus on the human-interest frame, whilst Greenpeaces’ dominant frame is the attribution of responsibility frame. Results however highlight that the dominant frame from BP is the conflict frame, whilst for Greenpeace this was found to be the human-interest frame.
Frame Functions
The following tests were run to determine whether there were significant in differences in the communications about the Deepwater Horizon Oil Spill by BP and Greenpeace regarding frame functions. As previously stated, it is believed that the frame function most prominent in communications from BP is treatment recommendation and in communications from Greenpeace casual interpretation. After conducting a chi-square test, with origin as the independent variable and frame functions as the dependent variable, it was found that the origin of communications significantly determined which frame functions were used X2 (15, N = 365) = 69,5, p < .001. Cramer’s V further showed that there was a moderate effect (d = .437), suggesting that the origin of the communication activities had a moderate impact on the frame functions.
The most prominent frame function in communication from BP was seen to be moral definition (29.7%), followed by treatment recommendation (23.1%), problem definition (21.4%) and casual interpretation (17.9%). In the case of Greenpeace, the most prominent frame function was found to be moral definition (90%), followed by problem definition (10%). In relation to the US media, we can see that the most dominant frame function for the Washington Post is moral definition (37.1%), whilst for the New York times it is treatment recommendation (56.6%). Following these results we accept hypothesis 4, there are
significant differences between actors’ usage of frame functions, yet reject hypothesis 5 and 6 showing that for both BP and Greenpeace the dominant frame function is moral definition.
Figure 1. Graph highlights the framing functions used by the different actors during the BP Deepwater Horizon oil spill.
Frame Alignment
The follow analyses were conducted to see whether there was frame alignment amongst actors’ communication efforts. As previously stated, it is believed that there are significant framing differences between BP and Greenpeace at the beginning of the crisis. As the crisis prolongs it is then expected that there will be frame alignment between Greenpeace and the US media. A one-way MANOVA was conducted to determine whether there was generic frame alignment between actors and whether time had any significant effects on the frame alignment. For this specific test, the dependent variables were the attribution of
responsibility frame, human-interest frame, conflict frame, morality frame, and the economic frame, whilst the independent variables were month of publication and origin. Furthermore, a multinomial logistic regression tested whether there was frame function alignment between
BP and Greenpeace and to determine if there was frame alignment between Greenpeace and the US media as the crisis continued.
Looking at the usage of generic frames by BP in relation to month of publication, it can be seen that the BP’s most prominent generic frame was the human-interest frame in April (M = 1.25, SD = 1.89), the conflict frame in May (M = 1.48, SD = 1.35), the conflict frame in June (M = 1.12, SD = 1.05), the human-interest frame in July (M = .760, SD = .1.42), the conflict frame in August (M = 1.04, SD = 1.24), and again the conflict frame in September (M = 1.00, SD = 1.47)
Figure 2. Average means of generic frames used by BP
In regards to Greenpeace, it can be seen that the most prominent generic frames used are the human-interest frame in May (M = 2.50, SD = .707), the human-interest frame in June (M = 2.00, SD = ) and the attribution of responsibility frame (M = 2.00, SD = ) in June, the conflict frame in July (M = 1.60, SD = .894), and finally the conflict frame (M = 1.50, SD = .707) and the human-interest frame (M = 1.50, SD = 2.12) in August. No press releases or
Facebook posts were found to be communicated by Greenpeace in April and September, and thus provided no means.
Figure 3. Average means of generic frames used by Greenpeace
After conducting the one-way MANOVA it was found that there is no relationship between the the origin and month of publication in relation to generic framing F (65, 1605) = 1.06, p = .345; Wilk’s = 0.820, partial 2
= .039.
Additionally, Box’s M was found to be significant F(135, 20808) = 2.64, p < 0.001, showing that the assumption of equal variances in the populations has been violated.
After conducting a multinomial logistic regression, it can be seen from table 1, that the insignificant Pearson value and significant likelihood value highlights that the full model significantly explains usage of frame functions. Furthermore, table 2 highlights what frame functions can be predicts depending on month of publication and origin.
-2 log likelihood
Final 240***
Pearson 78.1
Nagelkerke R2 .24
Table 1. Model fit and variation predicted Note. * p < .05 ** p < .01 *** p < .001 Problem Definition Casual Interpretation Moral Definition Treatment Recommendation Month -.511** (.008) -.511* (.014) -.089 (.635) -.323 (.085) The Washington Post -14.5 (.987) .225 (1.00) -15.9 (.986) -.473 (.517) The New York
Times -14.4 (.987) .047 (1.00) -16.4 (.985) -.678 (.338) BP -13.8 (.987) (-.935) (1.00) -15.4 (.986) 1.27 (-) Greenpeace 0 - - 0 Intercept 19.0 (.983) 4.26 (.998 18.3 (.983) 4.02** (.005) N 62 45 115 123
Table 2. Multinomial logistic regression between frame functions and month of publication and origin
Note. * p < .05 ** p < .01 *** p < .001
Following these results, we can state that we reject both hypothesis 7 and 8. Hypothesis 7 highlights the expectation that there will be significant differences between
as seen above there are no significant frame differences between BP and Greenpeace in the months April and May. We can also reject hypothesis 8, which highlight that there is no observable frame alignment between Greenpeace and the media as the crisis reached its latter stages.
CONCLUSION
In investigating framing differences amongst various actors involved in a single crisis, it has been highlighted that there are significant framing differences amongst actors. It is interesting to note that these differences do not highlight differences between BP and Greenpeace. Instead differences in generic framing are emphasized between the New York Times and the other actors. The most prominent generic frame used by the New York times is the economic frame, followed by the attribution of responsibility frame, whilst the most prominent frames in for BP, Greenpeace, and the Washington Post are the conflict frame and the human-interest frame (not in the same order for every actor).
For frame functions, it can be observed for BP, Greenpeace, and the Washington Post that the most prominent frame function was the moral definition (BP 29.7%, Greenpeace 90%, and the Washington Post (37.1%), whilst for the New York Times the most prominent frame function was treatment recommendation (56.6%). This further supports that there are partial framing differences between the US media, multinationals and social advocacy organizations when communicating about the same organizational crisis.
Furthermore, results show no significant effect was found that showed whether month of publication and origin had any effect on generic frames and frame functions. Specifically relating this to BP and Greenpeace, it can be seen that that in the first two months BP’s dominant generic frames were the human-interest frame in April and the conflict frame in May, whilst in April Greenpeace had no data and in May was found to predominantly focus
on the human-interest frame. Notably for frame functions, it can be seen there is no significant effect either for BP and Greenpeace in usage of frame functions dependent on month of publication. This means that it cannot be necessarily shown there were framing differences between BP and Greenpeace in the beginning of the crisis.
It was furthermore believed that there would be frame alignment between the US media and Greenpeace as the crisis’ life was prolonged. In relation to generic framing and frame functions it can be seen that there are no significant differences between Greenpeace, the New York Times and the Washington Post and month of publication. Thus as the crisis prolonged its life, no frame alignment was noticeable. This highlights that in relation to social advocacy organizations and the US media, as an organizational crisis continues, keeping in consideration that both are not the instigator, no frame alignment happens between the two.
DISCUSSION
Our findings show that there are significant framing differences between the US media and, multinationals and social advocacy organizations in the case of the Deepwater Horizon oil spill. There was furthermore no significant frame differences found between BP and Greenpeace at the initial stages of the crisis, whilst no frame alignment was further found between social advocacy organizations and the media as the crisis’ life-prolonged.
When examining the framing differences, it was of particular interest that the hypothesized frames thought to be predominant in Greenpeace’s and BP’s communication were not met. It was believed that the predominant generic frame used by Greenpeace would be the attribution of responsibility frame, and predominant frame function would be causal interpretation. Results however showed that Greenpeace’s most dominant frames were the human-interest frame and moral definition. In regards to BP, it was expected that the most dominant frames be the human-interest frame and would focus primarily on treatment
recommendation. Instead, results highlighted that BP focused primarily on the conflict frame and moral definition.
This could imply that communications should not focus only on the crisis but should focus on appealing to stakeholders. Luoma-aho & Vos (2010) further support this through claiming that the media landscape have changed organizations to discuss ideas and issues that are of interest to its stakeholders. More specifically however, this may highlight that the timing of exposure and repetition of the counter-frame employed, in this case by Greenpeace, may not have sufficiently influenced individuals’ processing of information and (Chong & Druckman, 2011).
Furthermore, as previously mentioned it was found that there was no significant differences in framing in the initial stages of the crisis between BP and Greenpeace. As the crisis reached its latter stages months further, it was expected that there would be frame alignment between Greenpeace and the media, however this was not seen either.
This could imply various things, starting off by highlighting the independence of the media. As both media sources were considered quality newspapers, objectivity and factual reporting may have been a key reason for the media not align in frames with Greenpeace. Conversely, Greenpeace’s interests may extend beyond media in communicating with the public. It may further be explained through the claims made by Chong & Druckman (2011), whom emphasize that in competitive environments frames tend to cancel each other out and produce no effects.
LIMITATIONS
This research paper does contain limitations. Firstly, the most prominent limitation is that there was limited amount of data available from Greenpeace. In total, Greenpeace’s data consisted out of 7 Facebook posts and 3 press releases. Compared to the other actors this was
significantly lesser and had significant effects on the results. The tests run to determine framing differences amongst the different actors may have been more significant between Greenpeace and the other if more data had been available from Greenpeace.
Furthermore, as shown in the results the values of Levene’s test and Box’s M were found to be significant, highlighting that the assumption of equal variances in the population had been violated. This may again be attributed to Greenpeace’s significantly lesser data available for testing, but mainly highlights that the results need to be predicting with caution. Were these tests found to be insignificant, and thus highlight that assumption of equal
variances in the population hadn’t been violated, the results could have been interpreted with more certain making the conclusions more reliable.
Third of all, the content analysis conducted highlights some gaps in the research. Namely, besides Greenpeace data being significantly less, no Facebook posts were analyzed from BP. This is contributed to the fact that BP’s Facebook page in the year of the crisis, 2010, was not present. Due to this only press releases were analyzed which did provide a significant number. Furthermore, for US media only two newspapers were analyzed, representing the whole US media. With both being quality newspapers, certain other US media and their characteristics were largely ignored, which could provide interesting results.
FUTURE RESEARCH
This paper contributes to future research through the added dimensions of comparing multiple actors’ framing efforts within a single set crisis situation. However, the paper does leave room for future research to extend and further investigate the topic. More specifically, as the research is set within one organizational crisis situation, results of the framing between social advocacy organizations, multinationals, and the US media are focused on one specific
crisis. Future research into the same topic pertaining to different organizational crises could further enhance the ability to generalize results.
Furthermore, future research could contribute significantly by adding potential other social advocacy organizations and media outlets to determine framing efforts. As the research focused only the two newspapers of the US media, a further analysis of the media landscape within this research could prove to generalize media’s framing efforts. Similarly,
incorporating different social advocacy organizations would provide more data that may significantly enhance or differ current framing by social advocacy organizations.
PRACTICAL IMPLICATIONS
In concurrence with the finding that there are no significant framing differences
between BP and Greenpeace, it may be interesting to highlight the salience of issues for social advocacy organizations. In specific, this crisis received wide spread attention on a global scale and caused BP to fall under immense amount of scrutiny. Even though Greenpeace involved itself after the occurence of the crisis, the importance of widely communicating about this crisis may not have been a priority due to the widespread attention it was already receiving. It is often seen that social advocacy organizations raise awareness for issues and try to activate participants, whilst simultaneously scrutinizing organizations behaving against social standards. In realizing this, this may help for organizations responsible for crises in communicating about their crisis and use of framing.
This paper shows that there are significant differences in framing between the media and social advocacy organizations and multinationals. This is seen to be true for both generic framing and frame functions. In acknowledging this, it could be of strategic significance for multinationals to focus on framing more significantly different than social advocacy
viable than moral definition as focus on moral definition by social advocacy organizations may invalidate arguments on this level.
Furthermore, it is hard to imagine, in occurrence with our results, that frame alignment would occur between quality media and independent organizations. Especially when engaged, the independent organizations are at the different ends of the spectrum concerning
environmental salience. In this argument though, it could be of interest for both multinationals and social advocacy organizations to focus on media’s reporting and frame’s used to gain insights on public opinion concerning the crisis.
REFERENCES
An, S., & Gower, K. K. (2009). How do news media frame crises? A content analysis of crisis news coverage. Public Relations Review, 35(2), 107-112.
Benoit, W. L. (1995). Accounts, excuses, and apologies: a theory of image restoration strategies. Administrative Science Quarterly, 42(3), 584.
Benoit, W. L. (1997). Image repair discourse and crisis communication. Public
Relations Review, 23(2), 177-186.
Carroll, A. B. (1991). The pyramid of corporate social responsibility: toward the moral maangement of organizational stakeholders. Business Horizons, 34(4), 39-48.
Castells, M. (2007). Communication, power and counter-power in the network society.
International Journal of Communication, 1.
Chong, D., & Druckman, J. N. (2007). Framing theory. Annual Review of Political Science, 10(1), 103–126.
Chong, D., & Druckman, J. 2011). Counter-framing effects. 1-37.
Coombs, W. (1998). An analytic framework for crisis situations: Better responses from a better understanding of the situation. Journal of Public Relations Research, 10(3), 177-191.
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.
De Vreese, C.H. (2005). News framing: theory and typology. Identifying Information
De Wolf, D., & Mejri, M. (2013). Crisis communication failures: The BP case study.
International Journal of Advances in Management and Economic, 2(2), 48-56.
Druckman, J. N., Hennessy, C.L., Charles, K. S., & Webber, J. (2010) Competing rhetoric over time: frames versus cues. The journal of Politics, 72(1), 136-148.
Entman, R. M., (1993). Framing: Toward clarification of a fractured paradigm.
Journal of Communication, 43(4), 51-58.
Gatzert, N. (2015). The impact of corporate reputation and reputation damaging events on financial performance: empirical evidence from the literature. SSRN Electronic Journal
SSRN Journal, 33(6), 485-499.
Hallahan, K. (1999). Seven models of framing: Implications for public relations.
Journal of Public Relations Research, 11(3), 205-242.
Hayes, A. F., & Krippendorff, K. (2007). Answering the call for a standard reliability measure for coding data. Communication Methods and Measures, 1, 77-89.
Heider, F. (1958). The psychology of interpersonal relation. New York: Wiley. Iyengar, S. (1990). Framing responsibility for poltiical issues: the case of poverty.
Political Behavior, 12(1), 19-40.
Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunties of ssocial media. Business Horizon, 53(1), 59-68.
Kühberger, A. (1998). The influence of framing on risky decisions, a meta-analysis.
Organizational Behavior and Human Decision Processes, 75(1), 23-55.
Luoma-aho, V., & Vos, M. (2010). Towards a more dynamic stakeholder model: Acknowledging multiple issue arenas. Corporate Communications: An International Journal,
15(3), 315-331.
McCombs, M.E., & Shaw, D.L. (1972. The agenda-setting function of mass media.
Milburn, T. W., Schuler, R. S., & Watman, K. H. (1983). Organizational crisis. Part I: Definition and conceptualization. Human Relations, 36, 1141-1160.
Patriotta, G., Gond, J., & Schultz, F. (2011). Maintaining legitimacy: Controversy, Orders of Worth, and Public Justifications. Journal of Management Studies, 48(8), 1804-1836.
Rindova, V. P., Williamson, I. O., Petkova, A. P., & Sever, J. M. (2005). Being good or being known: An empirical examination of the dimensions, antecedents, and consequences of organizational reputation. Academy of Management Journal, 48(6), 1033-1049.
Schultz, F., Kleinnijenhuis, J., Oegema, D., Utz, S., Atteveldt, W. V. (2012). Strategic framing in the BP crisis: A Semantic network analaysis of associative frames. Public
Relations Review, 38(1), 97-107.
Seeger, M.W., Sellnow, T.L., & Ulmer, R.R. (2003). Communication and
Organizational Crisis. Westport, CT: Praeger.
Semetko, H., & Valkenburg, P. (2000). Framing European politics: A content analysis of press and television news. Journal of Communication, 50(2), 93-109.
Snow, D. A., Rochford, E.B., Worden, S. K., & Benford, R. D. (1986). Frame
alignment processes, micromobilization, and movement participation. American Sociological
Review, 51(4).
Valkenburg, P. M., Semetko, H. A., & de Vreese, C.H. (1999). The effects of news frames on readers’ thoguhts and recall. Communication Research, 26(5), 550-569.
Van der Meer, T.G.L.A., Verhoeven P., Beentjes, H., & Vliegenthart, R. (2014). When frames align: the interplay between PR, news media, and the public in tmes of crisis. Public
Relations Review, 40(5), 751-761.
Van der Meer, T.G.L.A. (2014). Organizational crisis-denial strategy: the effect of denial on public framing. Public Relations Review, 40(3), 537-539.
Weick, K. E. (1988). Enacted sensemaking in crisis situations. Journal of
Management Studies, 25(4), 305-317.
Zhou, Y., & Moy, P. (2007). Parsing framing processes: The interplay between Online Public Opinion and Media Coverage. Journal of Communication, 57(1), 79-98.
APPENDIX A
Codebook ‘BP Deepwater Horizon oil spill’
Thesis Erik Lauret
Mr G.L.A. (Toni) van der Meer MSc Department of Communication Science
Corporate Communication University of Amsterdam
Introduction
The goal of this study is to map the difference in framing between social advocacy
organizations and multinationals during a crisis and to identify what role the media plays in these situations.
With the speed with which news and information spreads today, adequate crisis
communication is probably more relevant than ever. Within this research we will be looking at the usage of framing by social advocacy organizations and multinationals, and identify the role of news media because ‘frames are powerful in that they can solve problems and shape public opinion’. (An & Gower, 2009) This study will examine the 2010 case of BP Deepwater Horizon oil spill caused. The choice for social advocacy organizations and multinationals has been made on the basis of expected differences in experiences and perspectives. It has further been chosen to use US-media to identify the role of the media due to the geographical
location of the Deepwater Horizon oil spill.
Sample
Country Newspaper # Sum Coders
United States The Washington Post … ... …
The New York Times … … …
Organizations
International British Petroleum (BP) … … …
Greenpeace … … …
Total sample …
Units of analysis
The units of analysis are either articles from newspapers of which the main topics is the Deepwater Horizon oil spill or press releases from BP or Greenpeace of which the topic is the Deepwater Horizon oil spill as well. The analysis will be conducted on article level, press
To start off with, a number of essential formal variables regarding publication and content are to be coded. Subsequently variables on the usage of different will be coded.
The coding structure is visualized in the following coding scheme:
article ID c02 c03 c04 c05 c06 c07 c08 c09 c10 001 xx xx xx xx xx xx xx xx xx 002 xx xx xx xx xx xx xx xx xx 003 xx xx xx xx xx xx xx xx xx 004 xx xx xx xx xx xx xx xx xx
General coding guidelines
- Start off by looking through the article or press release - Answer questions 1 to 9
- Read the instructions before question 10, then continue - Read the instructions before question 18, then continue - Continue to thoroughly read the whole article or press release - Answer questions from 22 onwards
- Only code what is written in the article. Do not code interpretations or assumptions and do not code associations that one can ‘read between the lines’
01 Article ID
The number of the article, starting from 001 for each coder
02 Coder ID
1 Erik Lauret
2 Karim Rabbani (for ICR purposes)
03 Origin
1 The Washington Post 2 The New York Times 3 British Petroleum (BP) 4 Greenpeace 04 Type of press 1 newspaper article 2 press release 3 Facebook post 05 Month of publication 1 January 2 February 3 March 4 April 5 May 6 June 7 July 8 August 9 September 10 October 11 November 12 December 06 Year of publication yyyy
07 Article/Press release/Facebook post size
Code the number of words of which the article is composed
08 Textual headline of the news item/press release/Facebook post
Copy-paste the exact headline of the news item
The following questions pertain specifically to the structure and focus of news articles. If not coding a news article, using code 99 not applicable. Upon completion of these specific questions, please continue down to question 12
09 Placing of the article
Code the number of the page on which the article was published
10 Section
Code in which section of the newspaper the article was published
1 Mixed international/national news 2 International news 3 National news 4 Regional/local news 5 Business/economy 6 Culture 7 Opinion 0 Other 99 Not Applicable
Now please read the whole article, press release, or Facebook post thoroughly and continue with the questions.
11 Scope
Code how broad the scope of the article, press release, or Facebook post is. Does it focus only on the actual event or does it place the event in a broader context?
1 The item focuses merely on the BP Deepwater Horizon oil spill
2 The item does not per se focus on the case of the BP Deepwater Horizon oil spill, but places it into context at societal level
99 Not identifiable / not applicable
12 Event categorization 1 Political Event 2 Environmental Event 3 Social Event 4 Economic Event 5 Technical Event 6 Other 99 Not applicable
13 What is the main topic in the title?
1 Clean-up efforts/reparations 2 Political affairs
3 The responsibility/guilt question
4 (Economic) damage to BP as a company
5 Environmental consequences on national or international scale 6 Economic consequences on a national or international scale 7 Other
99 Not identifiable
14 What is the main topic narrated in the article/press release/Facebook post?
1 Clean-up efforts/reparations 2 Political affairs
4 (Economic) damage to BP as a company
5 Environmental consequences on national or international scale 6 Economic consequences on a national or international scale 7 Other
99 Not identifiable
15 Does the press release/article/Facebook post quote external sources? 1 Yes
2 No code 99 for questions 16 and 17
16 What is the first quoted external source?
1 BP representative
2 Greenpeace representative 3 Expert
4 Activist 5 Politician
6 Other news source 7 Other
99 Not applicable
17 What is the second quoted external source?
1 BP representative
2 Greenpeace representative 3 Expert
4 Activist 5 Politician
6 Other news source 7 Other
99 Not applicable
18 Does the press release/article/Facebook post mention external sources?
1 Yes
19 What is the first mentioned external source? 1 BP representative 2 Greenpeace representative 3 Expert 4 Activist 5 Politician
6 Other news source 7 Other
99 Not applicable
20 What is the second mentioned external source?
1 BP representative
2 Greenpeace representative 3 Expert
4 Activist 5 Politician
6 Other news source 7 Other
99 Not applicable
21 Dominant frame in article
21A Does the story suggest that some level of government has the ability to alleviate the problem?
0 no
1 yes
21B Does the story suggest that some level of the government is responsible for the issue/problem?
0 no 1 yes
1 yes
21D Does the story suggest that an individual (or group of people in society) is responsible for the issue-problem?
0 no
1 yes
21E Does the story suggest the problem requires urgent action?
0 no
1 yes
21F Does the story provide a human example or ‘human face’ on the issue?
0 no
1 yes
21G Does the story employ adjectives or personal vignettes that generate feelings of outrage, empathy-caring, sympathy, or compassion?
0 no
1 yes
21H Does the story emphasize how individuals and groups are affected by the issue/problem?
0 no
1 yes
21I Does the story go into the private or personal lives of the actors?
0 no
1 yes
21J Does the story contain visual information that might generate feelings of outrage, empathy-caring, sympathy, or compassion?
0 no
21K Does the story reflect disagreement between parties-individuals-groups-countries?
0 no
1 yes
21L Does one party-individual-group-country reproach another?
0 no
1 yes
21M Does the story refer to two sides or more than two sides of the problem or issue?
21N Does the story refer to winners and losers?
0 no
1 yes
21O Does the story contain any moral message?
0 no
1 yes
21P Does the story make reference to morality, God, and other religious tenets?
0 no
1 yes
21Q Does the story offer specific prescriptions about how to behave?
0 no
1 yes
21R Is there mention of financial losses or gains now or in the future?
0 no
1 yes
21S Is there a mention of the costs/degree of expenses involved?
21T Is there a reference to economic consequences of pursuing or not pursuing a course of actions?
0 no
1 yes
22 Frame Functions
Code the frame function most dominantly identifiable in the article. After reading the text, what seems to have been the most predominant function of the article or the press release?
1 Problem definition
- Highlights what effect an entity has produced or what event or result it is responsible for in relation to the costs and benefits.
- Clarification of key facts related to the problem
2 Causal interpretation
Identifies the forces creating the problem.
3 Moral evaluation
- Evaluates entities and their effects.
- Judgements made of parties implicated in the problem
4 Treatment recommendation
- Offers and justifies treatments for the problems, and aims to predict their likely effects
- The proposing of solutions and the discussion of possible results 99 Not identifiable
APPENDIX B
Checklist for ethical permission for research conducted by students in the framework of education
1) Title of the research project:
Framing the Deepwater Horizon Oil Spill; A Comparative Analysis Between Social Advocacy organizations, Multinationals, and the Media
2) Component of the programme (Bachelor’s or Master’s/Name of module): Master’s Programme Communication Science (Corporate Communication) 3) Student(s) who will conduct the research:
Erik Lauret
4) Teacher(s) who will supervise the research: Dr. G.L.A. Toni van der Meer
5) Brief description of the research (max. 200 words)
Framing differences between social advocacy organizations, multinationals, and the media will be examined. This relationship will be researched through using the BP Deepwater Horizon oil spill as a case study. The key actors to be focused on are BP as multinational, Greenpeace as social advocacy organizations, and the Washington Post and the New York times as media. Through researching the crisis it will be sought not only to find significant differences in framing between actors but also to establish whether their frames align as the crisis’ life reaches its latter stages. In identifying these framing differences, framing will be investigated according to previously established approaches, namely generic framing (Semetko & Valkenburg, 2000) and framing functions (Entman, 1993).
6) Research method (max. 100 words):
A content analysis will be conducted to analyze framing differences between social advocacy organizations, multinationals. In the content analysis, specific attention will be placed on newspaper articles, press releases, and Facebook posts to highlight the communication activities by the different actors. Press releases will be gathered directly from the official websites of both Greenpeace and BP, whilst newspaper articles will be collected via LexusNexus.
7) Where will the research be conducted (e.g. online, location, through organisation): online
8) Duration of the research (from the start of recruitment until the close of data collection): 1 week
9) Who are the participants? How will they be recruited? N/A
10) Are all participants adult (18 years or older), mentally competent individuals? O yes
If no, explain how active or passive permission will be obtained from the parents. N/A
11) Number of participants to be recruited: N/A
12) How will the anonymity and privacy of the participants be guaranteed? Explain. N/A
13) Will participants receive compensation for participating in the research? O no
If yes, explain. N/A
14) Will any misleading occur? O no
