The Agenda Building Success Of NFL Teams During The National Anthem Protests
University of Amsterdam
Graduate School of Communication Corporate Communication
Master’s Thesis
Eliane Goslinga
Student number: 11722274 E-mail: e.goslinga95@gmail.com Supervisor: Pytrik Schafraad
1
Abstract
This thesis focuses on the agenda building success of sports organizations. The study looks at how sports organizations use press releases to gain desirable news coverage around a politically charged public debate, in which they are involved. A content analysis of 85 press releases and 753 news articles showed that the rank of a sports organization does not influence the amount and tone of news coverage, and that incentives only influence the frequency once a press release has already been picked up. Sports organizations do not have a lot of second-level agenda building power, since neither the salience of affective nor substantive attributes was transferred from press releases to news articles. Furthermore, the political affiliation of news outlets does influence the transfer of salience of attributes, implying that sports organizations need to work extra hard to get positive coverage from news outlets with whom they disagree ideologically. This study shows that a lot remains unclear about what factors influence the agenda building power of sports organizations and that further research is needed to obtain more generalizable results.
Introduction
Media coverage is an important asset for organizations that can give organizations competitive advantages (Deephouse, 2000). It helps organizations maintain positive
reputations (Coombs, 2007) and is needed during key events that result in a lot of publicity (Westphal & Deephouse, 2011). For this reason, many organizations have communications departments dedicated to generating desired media coverage by sending out press releases. Press releases are an important form of information subsidy that organizations use to
influence the media’s agenda (Lewis, Williams, & Franklin, 2008). In this way, organizations are involved in agenda building, by trying to get their information into the news. Agenda
2
building “refers to the sources’ interactions with gatekeepers, a give-and-take process in which sources seek to get their information published and the press seeks to get that information from independent sources” (Ohl, Pincus, Rimmer, & Harrison, 1995, p. 90). Agenda building can be divided into two aspects. First-level agenda building looks at if press releases are picked up by the media and how often (Kiousis, Popescu, & Mitrook, 2007). Schafraad, van Zoonen, and Verhoeven (2016) showed that elements within press releases can make press releases more or less successful in generating media coverage. Second-level agenda building looks at which attributes of organizations are highlighted or hidden, which can result in a more positive or negative image of the organization (Kiousis, Mitrook, Wu, & Seltzer, 2006).
As Reich (2011) points out, the genre of the news and the organization’s industry play an important role in whether an organization will be placed on the media’s agenda. Schafraad et al. (2016) suggest looking at other fields to see if the same mechanisms work as they found for Dutch corporations. One field that has been researched insufficiently when it comes to media relations is sports organizations (Boyle, 2006). Even though professional sports teams need the media to cover their matches and other events, not much is known about how sports organizations generate media coverage (Sugden & Tomlinson, 2007). The news beat of sports coverage is different, because sports is covered constantly, even though it is not always newsworthy (Lowes, 1999). This could mean that the agenda building of sports organizations works differently than that of organizations researched in previous studies. It would be interesting to see how successful sports organizations are in first and second-level agenda building, since there seems to be a lot of practical and theoretical knowledge to gain from this kind of research.
There is plenty of media coverage of sports organizations, because sport is popular among vast audiences and is very valuable to advertisers, and big sports events and scandals
3
often become front-page news (Boyle, Rowe, & Whannel 2009). A highly-covered and recent event in sports has been the protests of American football players in the NFL during the national anthem. Just last month, the NFL’s decision to force players to stand during the anthem reached the front pages of major US newspapers such as The New York Times (Futterman & Mather, 2018). This shows that sport is not its own section and that the boundaries between sport and politics are blurred. This thesis will focus on the
communication of the 32 NFL teams relating to the national anthem protests and the resulting media coverage.
By focusing on the politicized issue of players protesting police brutality, this thesis tries to gain insight into the agenda building power of sports organizations involved in a political issue. The political nature of the issue gives us the opportunity to look at the
moderating effect of the political affiliation of the news outlets. Media outlets affiliated with the political ideology of the government might report political issues in a way that is in line with the government’s stance on the issue (Liu & Yao, 2015). As a result, sports organizations might have a hard time getting media coverage about a political topic in media outlets that they disagree with ideologically. Gaining empirical knowledge about which characteristics of press releases by sports organizations are effective, can help to reach the sports organization’s goal of getting media coverage about a politicized issue, and can lead to practical
recommendations for communication professionals in sports organizations.
This thesis aims to expand on Kiousis et al.’s (2007) study of American companies and Schafraad et al.’s (2016) study of Dutch corporations by looking at press releases and agenda building of sports organizations. They have not been researched extensively before, and therefore it is unclear how sports organizations can influence the media coverage they receive. This thesis aims to fill this gap in literature by looking at press releases published by teams of the NFL in the US. The aim of this study is to investigate how successful sports
4
organizations are in their agenda building practices and what factors influence this successfulness, which leads to the following research question: How do press releases by teams of the NFL influence the amount and tone of media coverage about these teams when involved in a national political debate?
Theoretical Framework
Background informationEven though many argue that politics have no place in sports, it seems impossible to separate the two in practice (Long & Spracklen, 2010). Sport has been used as a successful means to highlight political standpoints, because of the general consensus that sports are above the political realm (Hassan, 2018). Sport has a high socio-cultural prominence and is an important institution within human societies (Rowe, 2007). For this reason, political issues often infiltrate the realm of sport and sport has been used to promote political movements (Bloomfield, 2003). In some cases, sport has been the single outlet through which political debates have been given a platform (Hargreaves, 2000). Countless examples exist of athletes using their platform to give voice to social and political issues. For example, the ‘Black Power Salute’ (McNeal, 2018), and Muhammed Ali’s outspokenness about his political views
(Kaufman & Wolff, 2010).
On 14 August 2016, American football player Colin Kaepernick took a knee during the national anthem to protest the oppression of black people and people of color (Hauser, 2016). Even though Kaepernick’s intention was to raise awareness for police brutality towards black people, many Americans thought Kaepernick was disrespecting the US flag and
military. Because of this, the protests became a hot topic in the media and a big national debate was sparked about the appropriateness of social issues and politics in sports, in which people supported Kaepernick’s freedom of expression, and other heavily critiqued him for not
5
respecting the country (Jennings, 2017). The debate became more heated when President Trump called for the NFL players who did not stand for the anthem to be fired by owners of the teams (Webber, 2017). Because of Trump’s comments, the media increased their coverage about the topic, as the issue was now also political. A social movement arose, and many NFL teams participated in the protests during the anthem. All this media attention led to the teams publishing press releases to make clear what their standpoint was on the issue and Trump’s disapproving words (Gottlieb & Maske, 2017).
Whenever sports organizations are faced with much media coverage and are expected to participate in public debates, they have to know how to react. However, there is a lack of research into what sports organizations should do in such situations (Rowe, 2007).
Communicating in a way that is deemed wrong, inappropriate, or insufficient can damage the sport organization’s reputation (Coombs, 2007). Issue management and appropriate
communication are thus of vital importance to sports organizations. Issue Management
Organizations have to deal with crises that involve social issues at an increasing rate (Heath, & Palenchar, 2009). Nowadays, organizations have less control over their
communication practices, since organizations are no longer sending out their communications to stakeholders who passively receive the information (Luoma-aho & Vos, 2010). If an organization is involved in a public debate in the way the NFL teams were, communication practitioners can facilitate the public debate by identifying the right issue arenas, which are places of interaction where organizations and stakeholders discuss issues in which they have equal stakes (Luoma-aho & Vos, 2010). By managing issues in these arenas, which involves balancing the interests of stakeholders and the organization (Heath, 1997), the organization’s reputation can be upheld (Luoma-aho & Vos, 2010). If organizations are unable to manage their communication in these arenas, other stakeholders might become the dominant voice in
6
the discussion and convince the public to turn against the organization (Karagianni &
Cornelissen, 2006). However, since sports organizations are not expected to be participants in political arenas, sports organizations might have difficulties in engaging with the media and the mass public about political issues.
One way of making sure that the NFL teams’ voices and views are heard is by being covered in the media. By releasing press releases, the teams try to influence what the media talked about and in what way. Using public relations efforts to get on the media’s agenda is called agenda building (Kiousis, et al., 2006).
Agenda Building
The media’s evaluation of an organization can provide organizations with a competitive advantage (Deephouse, 2000). However, news rooms are flooded with information on a daily basis. News organizations do not have the capacity and budget to dedicate time to all the stories going on. This means that journalists have to make decisions on what to report in their news articles. Based on characteristics of the story such as the source and type of information, journalists have to decide which stories they pick up and reach the masses (Schafraad et al., 2016).
Sports organizations compete with one another for media attention and can try to influence this selection process by making their stories more newsworthy. Because of the scarcity of media attention, organizations work hard to make sure that they are present in the media, primarily by sending out press releases (Schafraad et al., 2016). The public relations departments of the NFL teams play a key role in the successfulness of agenda building (Kiousis et al., 2007). Sports organizations have to fight even harder to get the media’s attention around political topics, since a survey showed that only 3% of sports news coverage was related to politics (Horky & Nieland, 2013).
7
Agenda building can be split into two different aspects that each influence media coverage in different ways. The first aspect of agenda building is called ‘first-level agenda building’ and focuses on the prominence of topics in the news media (Carroll & McCombs, 2003). ‘Second-level agenda building’ is concerned with in what way media report about a certain topic (Kiousis et al., 2006). This thesis will look at what NFL teams are successful in gaining desired media attention about the national anthem protests through first and second level agenda building by analyzing their published press releases.
First-Level Agenda Building
As described above, first-level agenda building is the practice of getting an
organization or person coverage in the media. Most agenda building studies have looked at how often and in what way political candidates are portrayed in the news (Carroll &
McCombs, 2003). Since the public becomes aware of sports primarily through the media in the same way it learns about political candidates, first-level agenda building research can also be extended to sports organizations. If the press releases by NFL teams make it into the media and thus generate attention for that team, their agenda building practices are successful. Knowing what press releases result in media attention and why can be valuable information for sports organizations that want to obtain a competitive advantage and dictate the discourse around a specific issue. The first question that this thesis aims to answer is:
RQ1: What percentage of press releases published by NFL teams relating to the national anthem protests make it into the news outlets selected for this study?
However, it is unlikely that all teams receive the same attention based on their press releases (Schafraad et al., 2016). The characteristics of press releases and the teams
8
look at elements that could explain the difference in what press releases receive more media attention than others and thus explain the difference in agenda building success.
Elite Teams
Schafraad et al. (2016) showed that the success of first-level agenda building is influenced by several factors. One factor is the sender of the press release (Schafraad et al., 2016). The phenomenon that organizations receive more media coverage the higher their influence in society is called the ‘elite organizations’ news factor (Eilders, 2006; Harcup & O’Neill, 2001; Maier & Ruhrmann, 2008), or in this study the ‘elite teams’ news factor. The eliteness of teams is defined as the team’s implied societal importance and economic
influence based on their performance and is measured by the rank of the team on the official 2016 NFL Standings. Schafraad et al. (2016) found that Dutch corporations that ranked higher on a list of influential Dutch corporations were more likely to generate media attention based on their press releases than corporations lower on that ranking. Since sports organizations also enjoy importance and influence in our societies based on their achievements and
performances, it is likely that that NFL teams that perform better are seen as more elite in the same way. The ‘elite teams’ factor could thus play a role in NFL teams’ first-level agenda building success. Transferring the findings of Schafraad et al. (2016) on ranking and applying them to the 32 NFL teams, it is expected that:
H1a: The higher the team’s ranking on the 2016 standings, the higher the chances of its press releases being picked up.
H1b: A higher ranking on the 2016 standings is positively related to the number of times a press release is picked up by the media outlets.
Incentives
Apart from characteristics of the sender of the press release, several studies found that different stimulants, sometimes called ‘incentives’, can be added to press releases with
9
different effects on increasing media coverage (Schafraad et al., 2016; Aronoff, 1975). For example, linking press releases to current affairs and adding extra information like quotes and photos have been found to be effective (Seletzky & Lehman-Wilzig, 2010; Berkowitz and Adams, 1990). Schafraad et al. (2016) looked at the effects of the following incentives: Direct quotations, sensation, current affairs, research, and events, and found that the presence of these incentives positively impacted the chances of the press releases being picked up.
Including these incentives might increase the chances NFL teams’ press releases being picked up, and it is predicted that:
H2a: The presence of incentives in the press releases is positively related to the likelihood of a press release being picked up.
H2b: The presence of incentives in the press releases is positively related to the number of times a press release is picked up by the media outlets.
Second-Level Agenda Building
The way sports organizations are represented in the media provides the public with frames of reference and constructs social realities (Whannel, 1992). The emphasis on certain characteristics and the tone that the media use when talking about the NFL teams will thus influence the way people think about the teams. Agenda building theory is not only focused on the amount of media coverage, but also on how organizations are mentioned in the media (Carroll & McCombs, 2003). Second-level agenda building says something about whether positive or negative attributes of an organization are emphasized in the coverage or ignored. Kiousis et al. (2007) define an attribute as ‘a property, characteristic, or quality that describes an object’ (p. 151). Ohl et al. (1995) showed that news media mentioned attributes of the organization that were also highlighted in their press releases. This implies that by using certain attributes in their press releases, the NFL teams can shape the media’s coverage about them. Second-level agenda building involves two different types of attributes, affective and
10
substantive attributes (McCombs, Lopez-Escobar, Llamas, 2000). Affective attributes deal with how attributes are evaluated and are important for the perceptions of corporate reputation (Kiousis et al., 2007). This view acknowledges that news media do not merely report facts, but also convey feeling and tone (McCombs & Ghanem, 2001). The affective attribute that will be analysed in this thesis is ‘tone’, which covers whether attributes in the press releases and news articles are evaluated in a positive, negative, mixed, or neutral light (Deephouse, Carroll, & McCombs, 2001). Looking at how tone is transferred from press releases to news articles can be valuable information from a business perspective (Kiousis et al., 2007). On the other hand, substantive attributes refer to characteristics of objects, such as the experience and ideologies of political candidates (Carroll & McCombs, 2003). The substantive attributes that are included in this study are support and criticism towards the protesting actions of players.
This thesis will investigate whether the transfer of affective and substantive attributes applies to sports organizations and could help sports organizations manage their future communication strategies. Based on previous work by Kiousis et al. (2007) the following hypotheses will be tested to see how second level agenda building affects media coverage of NFL teams:
H3: The higher the team’s ranking on the 2016 standings, the more positive the tone of the media attention based on the press releases.
H4: The salience of the tone in press releases will be positively related to the salience of the tone in media coverage.
H5a: The salience of the substantive attribute ‘support’ in press releases will be positively related to the salience of the substantive attribute ‘support’ in media coverage.
H5b: The salience of the substantive attribute ‘criticism’ in press releases will be positively related to the salience of the substantive attribute ‘criticism’ in media
11 coverage.
The Political Affiliation of News Outlets
Public protests and social movements often use the news media to spread their
message among the mass public (Smith, McCarthy, McPhail, & Augustyn, 2001). The media frame news and make sense of events for the audiences. Many protest organizers accuse the media of incorrectly portraying the cause and meaning behind their protests (Smith et al., 2001). News organizations are often seen as ‘gatekeepers’ and are expected to strive for a standard of objectivity (Reese, 2001). News organizations are shaped by journalists
individually, media routines, organizations’ routines, extramedia forces, and ideology (Reese, 2001). And even though many outlets claim to be balanced in their reporting and selecting topics because of their newsworthiness, they are often accused of bias in this selection process by wanting to either appease their owners or their intended audiences (Niculae, Suen, Zhang, Danescu-Niculescu-Mizil, & Leskovec, 2015). Niculae et al. (2015) propose that bias occurs on different levels and along different dimensions, and that ideological stand bias affects how issues are commented on and that the political ideology dimension is also often perceived. They showed that conservative news outlets portray persons differently than liberal outlets by using different quotes. This implies that when reporting about a political issue, news outlets affiliated with political ideologies on different ends of the spectrum might cover political issues differently, which means that the success of agenda building is influenced by the outlet’s ideology. For this reason, four news outlets were chosen that differ in who they cater to ideologically, based on research by Mitchell, Gottfried, Kiley, & Matsa (2014). Since the NFL teams have been attacked by President Trump for their protests, it seems plausible that news outlets who support Trump might be less eager to (positively) cover the protests of the NFL players. This leads to the following hypotheses:
12
articles are based on press releases by the NFL teams, in a way that the more left leaning a news outlet is the more likely it is to base their news article on a press release.
H7a: The political affiliation of media outlets will have a moderating effect on the relationship between press releases and the tone of the media coverage, in a way that right-leaning outlets will cover the NFL teams in a less positive tone than center and left leaning media and will be more positive about Trump.
H7b: The political affiliation of media outlets will have a moderating effect on the relationship between press releases and the salience of substantive attributes in the media coverage, in a way that right-leaning outlets will cover the NFL teams more critically and left-leaning outlets will express more support.
Methodology
This section describes the quantitative content analysis that was conducted on press releases and news articles to test the hypotheses and answer the research question. This method was used to investigate the agenda building power of NFL teams. Additionally, the political affiliation of the news outlets was included in the analysis as a potential moderator. The design of this study is a quantitative content analysis of press releases by NFL teams pertaining to players’ protests during the national anthem and news articles of four US news outlets about the NFL teams within a set timeframe.
Sample and Data Collection
The press releases were retrieved using two different methods to ensure the highest inclusion of relevant press releases. The first method was to collect the press releases from the news sections on the websites of the 32 NFL teams. To be included in the sample, the press release had to address the players’ protests during the American national anthem. This was
13
determined by looking at the titles of the press releases. If the topic of the press release could not be determined by looking at the title, the press release was opened and scanned to
determine the topic. The timeline for the press releases started on 14 August 2016, the day Colin Kaepernick first kneeled during the national anthem. Because of how recent this topic is, January 31 2018 was chosen as the end of the timeline. All press releases within this timeframe that addressed one or more NFL players protesting was included in the sample. Since not all press releases were accessible on the teams’ websites, the sample was expanded by including all the press release collected by the National Public Radio. Press releases that were collected by the NPR for their article (Gottlieb & Maske, 2017) and that were not available through the official websites were included in the sample. In total, 85 press releases were included in the sample and were subjected to the content analysis. The press releases in their entirety were used for the content analysis, but only the text of the press release (including the title) was analyzed; images and videos were excluded because of time constraints.
Subsequently, news articles from four US news outlets were collected using
LexisNexis or the outlets’ official websites. To be able to investigate the moderating effect of the political affiliations, news outlets that differed in political affiliation were selected based on research by the Pew Research Center (Mitchell et al., 2014). News outlets were also selected based on online access and circulation numbers. USA Today is one of the most read US news outlets and can be categorized on the center of the ideological spectrum. The New York Times has the highest circulation numbers of the left-leaning newspapers. Fox news was chosen as the right leaning media outlet, because it is the most popular media outlet among right wing Americans. Additionally, ESPN.com was chosen, because of its dedication to sports news and its leading position as a sports outlet (Wong, 2008). To sample the news articles, Schafraad et al.’s (2016) sampling method was used. This method consisted of
14
searching for news articles that mentioned at least one of the 32 NFL teams on the day the press released was published and six consecutive days after the publication. The timeframe of six days was chosen because of the news’s fast pace. Press releases usually only make it into the news within the first few days of their release, because after a while they will be
considered old news and will lose their newsworthiness. In total, 753 news articles were analyzed. Of these 753 news articles, 377 were excluded because they were not about the national anthem protests. The remaining 376 articles were coded to see if they were based on press releases. In the end, 133 news articles were included in the sample because they were based on a press release and were further analyzed.
Coding Instrument
All press releases and news articles were given a code for identification purposes and were coded on general information: date, team, and sender. Press releases and news articles were also coded for the following variables:
Amount of media coverage looked at whether the first level agenda building process of NFL teams was successful (Carrol & McCombs, 2003). This variable was operationalized in two parts, first by analyzing the data and seeing if a press release was picked up, and second by counting how often a press release had been picked up. This variable was used in
hypotheses 1 and 2 and was the binary dependent variable in the first aspect of media coverage and the continuous dependent variable for the second aspect.
The rank of NFL teams had 32 options in the codebook, and was later recoded into an interval level scale in a way that the team that came in first received a count of 32 and the team that ended up last received 1 point.
Press releases were checked for the presence of incentives (Schafraad & Spitteler, 2014). The incentives that were checked for presence were: quotes by key persons, quotes by relevant others, quotes by laypersons, sensation, current affairs, research, and events.
15
Descriptions for these seven incentives can be found in table 1. An incentive was present when an element as described in the codebook (see appendix I) could be identified as having been added to the text of the press release. More than one incentive could be present in the press release. An eighth option was included in the codebook that read ‘no incentives are present’. This was treated as a categorical independent variable with eight different options for hypothesis 2.
Table 1 Descriptions of the seven incentives (based on Schafraad & Spitteler, 2014)
Incentive Description
Quotes by laypersons
A quote from someone within the organisation is included in the text. Examples are representatives and employees
Quotes by key persons
A quote that is ascribed to an important person within the organization. These key figures are CEO, Coach, player(s), and owners.
Quotes by relevant others
A quote from someone outside the organization that sent the press release is included in the press release.
Research The press release is based on research requested by (a trade organization of) the sender of the press release.
Sensation The title of the press release consists of ‘hyperbole’, ‘cynicism’, ‘satire’ and/or the title is in capital letters, underlined, bold, or ending in an exclamation mark.
16
Current affairs An explicit link to current affairs is included in the press release.
Events The press release contains an explicit link to an event that was organized by the sender of the press release. For example, a press conference.
The tone of press releases and media coverage refers to whether the press releases and news articles talked about (a) the sports organization, (b) the protesting actions, and (c) Trump in a positive, negative, mixed, or neutral way (or not at all) (Deephouse, 2001). This variable was recoded into an interval variable where ‘not applicable’ was recoded into
missing variables, and a scale was created where ‘negative’ was 1, ‘neutral’ and ‘mixed’ were put together as they are both not overwhelmingly positive or negative and were given the code 2, and ‘positive’ was coded as 3.
To see whether the protesting actions of the players were evaluated positively or negatively, the variables support and criticism were analyzed. Support refers to the press releases mentioning acceptance or support for the players’ protests and was measured on an interval scale ranging from ‘no support at all’, to ‘to some degree’, and finally ‘strong support’. Criticism was measured by looking at whether disapproving words were used in relation to the protests, and a similar ordinal scale was used as for support. The second level agenda building variables support and criticism were coded in press releases and news articles.
The political affiliation of news outlets was treated as a categorical variable with four options in the codebook, namely The New York Times, Fox News, USA Today, and ESPN. It was used as an independent variable for hypothesis 6 and as a moderator in hypothesis 7. For the analyses, the outlets were recoded into an ordinal scale ranging from left-wing to
right-17
wing to see what influence political affiliation had. ESPN was coded as missing values, since it was seen as having no political affiliation. The News York Times received the code ‘1’ in the scale, while USA Today and Fox News received 2 and 3 respectively.
Inter-coder Reliability
All coding for this study was done manually by one student and to facilitate the coding an extensive codebook was developed. To ensure that the codebook and the coding was reliable, 10% of the sample was coded by another master thesis student and inter-coder reliability tests were run. As can be seen in Table 2, Krippendorff’s Alpha (Krippendorff, 2004) was calculated for the six main variables (incentives only for press releases). This test yielded the following result for the variable incentives: Krippendorff’s Alpha of .795. The KAlpha for tone towards team was .817 in press releases and .755 in news articles. The KAlphas for tone towards actions were .667 and .714 in press releases and news articles respectively. The KAlpha tone towards Trump in press releases was .838 and .887 in news articles. Support had a Kalpha of .685 in press releases and .789 in news articles. Criticism had a Kalpha of .722 in press releases and .892 in news articles. The Alphas of tone actions and support in press releases were just above the accepted limit of .667 (Krippendorff, 2004). Tone team and tone Trump in press releases, and tone Trump and criticism in news articles had satisfactory reliability (Krippendorff, 2004). Incentives and criticism in press releases, and tone team, tone actions and support in news articles are considered to have moderate reliability (Krippendorff, 2004).
Table 2 The inter-coder reliability measurements for six variables
Variable Press Releases News Articles
18 Incentives .795 - Tone Team .817 .755 Tone Actions .667 .714 Tone Trump .838 .887 Support .685 .789 Criticism .722 .892
Results
Hypotheses testingOut of the 85 press releases analyzed, 56 were picked up by the news outlets, which is 65.9%. Hypotheses 1 and 2 were split into two different tests. First, the influence of rank and incentives on the succession of press release in news articles was investigated. Second, I looked at the relationship between rank and incentives and how often a press release was picked up by the news outlets. To test the first part, a multiple logistic regression test was used with rank and presence of incentives as independent variables and press release is picked up yes/no as binary dependent variable. The model chi-square test for succession of press release is 7.57 (8) (p = .477), this means that the main effect in this model was not significant.
To see if rank and presence of incentives influenced how often a press release found its way into news articles, a multiple regression test was conducted with ranking as interval independent variable, each incentive as independent dichotomous variables, and how often a press release was picked up as continuous dependent variable. On average, a press release was
19
picked up 2.62 times (M = 2.62, SD = 4.61). There was a statistically significant main effect for incentives [F (7, 33) = 3.74, p = .004]. No post-hoc test could be performed, because at least one group had fewer than two cases. The influence of ranking on how often a press release is picked up was not significant [F (31, 33) = .80, p = .732]. Based on the two tests that were conducted, we can say that hypothesis 1a and b and 2a were rejected and that
hypothesis 2b was supported, since incentives did have a positive impact on how often a press release was picked up, but not on the likelihood of succession itself. Rank does thus not influence the agenda building success, and incentives only once the press release has already been picked up.
Hypothesis 3 predicted that rank would positively influence tone, which meant that the higher the NFL team ended on the 2016 NFL Standings, the more positive the news articles would become towards the team and the team’s protesting actions. Rank and tone were treated as interval variables, with the dependent variable tone being divided into tone towards team and tone towards protesting actions. Simple linear regression tests were applied two times to the 85 press releases and 133 news articles to investigate the relationship between rank of NFL team and tone of news articles. A non-significant regression equation was found for tone towards team (F(1, 176) = 1.124, p = .291), with an R2 of .006. The rank of the NFL team did not impact the tone of the news articles towards the teams. The data for tone towards the players’ actions also yielded a not significant equation (F(1, 210) = 2.520, p = .114), with an R2 of .012. The position on the 2016 NFL Standings did not influence the tone of news articles towards the protesting actions of players. Hypothesis 3 is thus rejected.
Hypothesis 4 investigated the relationship between the tone used in press releases towards the NFL teams, protesting actions, and Trump, and the salience of these affective attributes in news articles. To test this, three separate linear regression tests were used,
20
because both the dependent and independent variables were interval. Table 3 shows the means and standard deviations of the tone of press releases and news articles.
Table 3: The salience of affective attributes in press releases and news articles
Tone Towards
n M SD
Team in Press Release Team in News Article Actions in Press Release Actions in News Article Trump in Press Release Trump in News Article
73 178 70 212 27 158 2.89 2.37 2.57 1.97 1.04 1.39 .393 .538 .672 .537 .192 .526
The press releases were most positive about the teams (M = 2.89, SD = .393), followed by the protesting actions of the players (M = 2.57, SD = .672), and negative towards Trump (M = 1.04, SD = .192). If we look at the news articles, we see that they have become slightly less positive about the teams and the tone is now more neutral or mixed (M = 2.37, SD = .538). The news outlets are more negative towards the protesting actions of the players, with a mean just under neutral/mixed (M = 1.97, SD = .537). The news articles are also mostly negative towards Trump, which is in line with the press releases (M = 1.39, SD = .526).
The first regression test showed that there is not a significant effect (F(1, 64) = 1.131, p = .718), with an R2of .002 between the salience of the tone towards the NFL team in press releases and in news articles. Similarly, the tone towards the protesting actions of the players was also not carried over from press release to news article: (F(1, 68) = 1.390, p = .242), with an R2 of .020. The test for the salience of the tone towards Trump(‘s remarks) also gave a not
21
significant negative result (F(1, 23) = .920, p = .347), with an R2 of .038. Since the three different tests yielded no significant results, hypothesis 4 was rejected.
Hypotheses 5a and 5b looked at whether the substantive attributes support and
criticism were picked up by the news outlets in their articles. Table 4 gives an overview of the means and standard deviations.
Table 4: The salience of substantive attributes in press releases and news articles
Support & Criticism
n M SD
Support in Press Release Support in News Article Criticism in Press Release Criticism in News Article
85 214 85 214 2.02 2.21 1.19 2.30 .740 .747 .500 .753
The press releases on average mention support to a certain degree (M = 2.02, SD = .740), similarly to news articles (M = 2.21, SD = .747). News articles (M = 2.30, SD = .753). are on average more critical than press release (M = 1.19, SD = .500).
A linear regression test was done with support in news articles as the interval
dependent variable and support in press releases as an interval independent variable. The data showed that the relationship between support in press releases and news articles is not
significant (F(1, 83) = .501, p = .481), with an R2of .006. The results of the linear regression test for the attribute criticism were also not significant (F(1, 83) = .306, p = .582), with an R2 of .004. Therefore, Hypothesis 5a and 5b were rejected.
To test hypothesis 6, a chi-square test was conducted. The 376 analyzed news articles were subjected to a chi-square test to see if the news outlet significantly influenced whether a
22
news article was based on a press release or not. The news outlet was treated as a categorical independent variable with 4 options and ‘based on press release’ was the dependent
dichotomous variable. As can be seen in Table 5, 35.4% of news articles were based on press releases. 37.6% of New York Times articles were based on press release, 37.2% of USA Today articles, 37.2% of Fox News articles, and 25% of ESPN articles. The chi-square test did not yield significant results regarding the relationship between news outlet and based on press release yes or not: χ2 (3) = 3.34; p = .34. ESPN had a standardised residual of -1.4. Since
the chi-square test was not significant, no Lambda test for the direction of the association was conducted and hypothesis 6 can be rejected.
Table 5: The likelihood of using a press release for each outlet
Based on
Press Release News Outlet
New York
Times USA Today Fox News ESPN Total
No 73 (62.4%) 27 (62.8%) 98 (62.8%) 45 (75%) 243 (64.6%)
Yes 44 (37.6%) 16 (37.2%) 58 (37.2%) 15 (25%) 133 (35.4%)
Total 117 43 156 60 376
Notes: N = 376, χ2 = 3.37, p = .34
Hypothesis 7a proposed that the political affiliation of the news outlets would have a moderating effect on the relationship between tone in press releases and in news articles. Three multiple linear regression analyses were used to predict the moderating effect of news outlet on the relationship between the tone in press releases and news articles. A
non-significant regression equation was found for tone towards team (F(3, 54) = .142, p = .934), with an R2 of .008. A non-significant regression equation was also found for tone towards
23
regression equation was found for tone towards Trump (F(2, 21) = 1.214, p = .317), with an R2 of .104. Hypothesis 7a is rejected.
Hypothesis 7b predicted that the political affiliation would act as a moderator for the salience of substantive attributes in press releases and news articles. Two multiple regression tests were run to see if political affiliation was a moderator for ‘support’ and ‘criticism’. A significant regression equation was found for ‘support’ (F(2, 71) = 6.501, p = .001), with an R2 of .215. Results indicated that the more right-wing news outlet (b* = -.40, t = -3.82, p <
0,001) was weakly associated with less support in news articles. The interaction between news outlet and support in press release was also significant (b* = .251, t = 2.28, p = .026), suggesting that the transfer of support from press release to news article weakly depended on the news outlet. Left leaning outlets were more likely to highlight the support attribute in their articles that was also present in the press releases compared to right-leaning.
A significant regression equation was also found for ‘criticism’ (F(3, 71) = 3.584, p = .018), with an R2 of .132. Results indicated that the more right-wing news outlet (b* = .36, t = 3.21, p = 0,002) was moderately associated with more criticism in news articles. However, the interaction between news outlet and criticism in press release was not significant (b* = -.02, t = -.15, p = .879), suggesting that the transfer of criticism from press release to news article did not depend on the news outlet. Hypothesis 7b is partly supported since political affiliation of news outlet was a moderator for support, but not for criticism.
Conclusion
The aim of this thesis was to add to literature of public relations and agenda building theory by extending the research to sports organizations. A content analysis was performed on press releases by NFL teams and news articles by four US news outlets in order to answer the
24
question of how the publication of press releases influences the amount and the tone of media coverage these teams receive.
65.9% of press releases were picked up by news outlets. This study found that it does not matter who sends out the press release, since higher-ranking teams do not get covered significantly more (positive) than lesser teams. It also showed that adding incentives only partially influences media coverage, as it did not increase the likelihood of a press release being picked up, but it did increase the number of times a press release was picked up once it had already been used. The second level agenda building efforts of the NFL teams were not very successful, seeing how both affective and substantive attributes were not copied from press releases to news articles. The results also showed that when involved in a political matter, the political ideology of the news outlets does not impact the likelihood of using press releases. However, the political ideology of newspaper does affect the salience of substantive attributes. Even though moderation was not found for the affective attribute tone, the analyses did show that the ideology impacted the salience of the positive attribute support and the negative attribute criticism.
Discussion
Theoretical implications and Contributions
This thesis contributed to literature on agenda building processes and public relations practices in several ways. First, it followed Schafraad et al.’s (2016) suggestion to investigate the success of first-level agenda building in the field of sports. By looking at the
communication practices of sports organizations, this study also tried to add to the lacking research into the media relations of sports organizations (Boyle, 2006). The high percentage of press releases being picked in this study (cf Schafraad et al, 2016) might be explained by the news beat of sports news, which consists of consistent coverage even if it is not
25
newsworthy (Lowes, 1999), and the controversial nature of the politicized topic as it has been shown before that the news factor controversy is positively influences the chances of media coverage (Schafraad et al, 2016). This unique combination of constantly covered sports news and the political controversy of this case could explain the high first-level agenda building success.
The combination of looking at both levels of agenda building also gives a more
complete picture of organizations’ agenda building success. This study also tried to contribute to issue management and crisis communication theory and showed that when sports
organizations are involved in an issue as widely covered as the national anthem protests that the rank of the sports organization does not matter. This is contrary to what Schafraad et al. (2016) found and implies that there are other factors that influence an organization’s agenda building success and that a higher ranking does not automatically lead to more or more positive coverage. This thesis also added to the theories of incentives, since it was proven that for the NFL teams, the presence of incentives was only relevant once a press release had already been picked up by the newspapers. This finding is in line with Schafraad et al.’s (2016) and confirms their results for the effectiveness of adding incentives to press releases. On the other hand, the results of this study did not correspond with Kiousis et al.’s (2007) finding for the transfer of affective and substantive attributes. The salience of attributes in press releases of NFL team was not copied in the news articles, implying that wanting positive coverage in the news media is not achieved by solely stressing these attributes in press releases, but that other factors play a role in which attributes will be most salient. Additionally, this study found that the political affiliation of the news outlets might be one of these factors influencing the salience of substantive attributes in news articles and that NFL teams thus might have to be afraid that they will receive more negative coverage from news outlets with whom they disagree ideologically. This way, this research also adds to the theory
26
of how sports organizations should employ their communications tactics when they are involved in political debates, as it showed that sports organizations will have to find other tactics to get positive media coverage from news organizations that they disagree with ideologically, and that adding positive substantive attributes to press releases will help them get positive coverage from outlets that they agree with.
Practical implications
Communications practitioners of NFL teams can take away some practical
implications from this study. Contrary to what was hypothesized, the rank of NFL teams does not influence the agenda building successes. Coming in first does not guarantee more and more positive coverage than ending up at the bottom of the NFL Standings. This means that lesser teams do not need to work extra hard by sending out more press releases than others, and that teams at the top should not become lazy in their public relations efforts and expect preferential treatment from the media. Furthermore, practitioners can add stimulants to their press release in the form of direct quotations and other incentives if they want to have their press release picked up several times, but it does not improve the likelihood of a press release being picked. Lastly, communication professionals should keep in mind that there are many factors that contribute to what attributes will be highlighted in the media, and that adding extra positive affective and substantive attributes to their own press releases will not lead to more positive coverage, since news outlets are unlikely to simply transfer the tone and
attributes like support and criticism. On the other hand, the organizations should keep in mind that political ideologies of news outlets play a role in how they will transfer substantive attributes.
27
A number of limitations of this thesis have to be addressed. First, all the data was collected and coded by one single coder and the intercoder-reliability scores were all above the accepted threshold, but not all could be deemed sufficient. The likelihood of bias and errors is higher if there is only one coder and this could make the results invalid.
Additionally, because of time constraints, the sample of this study is fairly small since just four US news outlets were investigated and no other media such as Twitter and TV. Because of this lack of diversity in studied media and corporate media and news media, the results might not be generalizable to a larger population. Also, all the press releases and news articles were centered around a big issue in sports that generated a lot of coverage, maybe not making it the best fit to do an analysis of the agenda building. If a topic is so widely covered, sports organizations might need to do less effort, since they will already receive much
coverage, because it is a controversial topic and current affairs.
Several reasons could be brought up to explain why most of the hypotheses were rejected. The high success of first-level agenda building that was found might mean that when an issue is already as widely covered, that other factors such as rank and incentives do not influence this. The fact that President Trump was involved might explain why the attributes were not transferred, because outlets often chose to include Trump’s negative words, which already steer the tone of a news article in a certain direction.
Future research should look at agenda building successes of sports organizations on a larger scale. The unique nature of this highly politized case within sports makes it harder to generalize these findings to sports organizations in general. More press releases, news articles, and media should be analyzed in order to get more reliable data that can be applied to sports organizations in a more general way. Similarly, future research could look at the publication of press releases during a longer period of time and not only relating to a hotly debated topic,
28
with the intent of getting more useful knowledge of what is effective when sports organizations need to put more effort into being heard.
References
Aronoff, C. (1975). Credibility of public relations for journalists. Public Relations Review, 1(2), 45-56.
Berkowitz, D., & Adams, D. (1990). Information subsidy and agenda-building in local television news. Journalism & Mass Communication Quarterly, 67(4),
723–731.
Boyle, R. (2006). Sports journalism: Context and issues. Thousand Oaks, CA: Sage Publications.
Boyle, R., Rowe, D. and Whannel, G. (2009). Delight in trivial controversy? Questions for sports journalism. In: Allan, S. (Eds.), The Routledge Companion to News and Journalism (pp. 245-255). New York, NY:Routledge.
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(1), 36-46.
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.
Deephouse, D. L. (2000). Media reputation as a strategic resource: An integration of mass communication and resource-based theories. Journal of management, 26(6),
109111112.
29
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 Competitiveness, Paris.
Eilders, C. (2006). News factors and news decisions. Theoretical and methodological advances in Germany. Communications, 31(1).
Futterman, M., & Mather, V. (2018, May 24). Trump Supports N.F.L.’s New National Anthem Policy. The New York Times. Consulted on 03-06-2018. Retrieved from
www.nytimes.com
Gottlieb, J., & Maske, M. (2017, September 23). Roger Goodell responds to Trump’s call to ‘fire’ NFL players protesting during national anthem. The Washington Post. Consulted on 10-04-2018. Retrieved from www.washingtonpost.com
Harcup, T., & O’Neill, D. (2001). What is news? Galtung and Ruge revisited. Journalism Studies, 2(2), 261–280.
Hargreaves, J. (2000). Heroines of Sport: The Politics of Difference and Identity. London: Routledge.
Hassan, D. (2018). Sport and politics in a complex age. Sport in Society, 21(5), 735-744. Hauser, C. (2016, August 27). Why Colin Kaepernick Didn’t Stand for the National Anthem.
The New York Times. Consulted on 10-04-2018. Retrieved from www.nytimes.com Heath, R. (1997). Strategic Issues Management. Thousand Oaks, CA: Sage.
Heath, R.L. and Palenchar, M.J. (2009). Issues management and crisis communication, in Heath, R.L. and Palenchar, M.J. (Eds.), Strategic Issues Management: Organizations and Public Policy Challenges (2nd edn, pp. 273–308). Thousand Oaks, CA: Sage Publications.
Horky, T., & Nieland, J. U. (Eds.). (2013). International sports press survey 2011 (Vol. 5). BoD–Books on Demand.
30
Jennings, P. (2017, October 11). Colin Kaepernick: From one man kneeling to a movement dividing a country. BBC. Consulted on 10-04-2018. Retrieved from www.bbc.com
Karagianni, K. S., & Cornelissen, J. (2006). Anti-corporate movements and public relations. Public Relations Review, 32(2), 168-170.
Kaufman, P., & Wolff, E. A. (2010). Playing and protesting: Sport as a vehicle for social change. Journal of Sport and Social Issues, 34(2), 154-175.
Kiousis, S., Mitrook, M., Wu, X., & Seltzer, T. (2006). First-and second-level agenda building and agenda-setting effects: Exploring the linkages among candidate news releases, media coverage, and public opinion during the 2002 Florida gubernatorial election. Journal of Public Relations Research, 18(3), 265-285.
Kiousis, S., Popescu, C., & Mitrook, M. (2007). Understanding influence on corporate
reputation: an examination of public relations efforts, media coverage, public opinion, and financial performance from an agenda-building and agenda-setting perspective. Journal of Public Relations Research, 19(2), 147–165.
Krippendorff, K. (2004). Content analysis: An introduction to its methodology. Thousand Oaks, CA: Sage.
Lewis, J., Williams, A., & Franklin, B. (2008). A compromised fourth estate? UK news journalism, public relations and news sources. Journalism studies, 9(1), 1-20. Liu, J., & Yao, Z. (2015). How to play the game of intellectual property rights in China: the
impact of party affiliation and media location on the use of media frames. Asian
Journal of Communication, 25(4), 371-392.
Long, J., & Spracklen, K. (Eds.). (2010). Sport and challenges to racism. Springer.
Lowes, M.D. (1999). Inside the Sports Pages: Work Routines, Professional Ideologies, and the Manufacture of Sports News. Toronto, ON, Canada: University of Toronto Press. Luoma-aho, V., & Vos, M. (2010). Towards a more dynamic stakeholder model:
31
acknowledging multiple issue arenas. Corporate Communications: An International Journal, 15(3), 315-331.
Maier, M., & Ruhrmann, G. (2008). Celebrities in action and other news factors of German TV news 1992–2004. Results from a content analysis. Human Communication, 11(1), 197–213.
McCombs, M. and Ghanem, S.I. (2001) The convergence of agenda setting and framing. In Reese, S.D., Gandy, Jr O.H. and Grant, A.E. (Eds.), Framing Public Life:
Perspectives on Media and our Understanding of the Social World (pp. 67–81). Mahweh, NJ: Lawrence Erlbaum Associates.
McCombs, M. E., Lopez-Escobar, E., & Llamas, J. P. (2000). Setting the agenda of attributes in the 1996 Spanish General Election. Journal of Communication, 50, 77–92.
McNeal, L. R. (2018). From Hoodies to Kneeling During the National Anthem: The Colin Kaepernick Effect and Its Implications for K-12 Sports. Louisiana Law Review, 78(1), 8.
Mitchell, A., Gottfried, J., Kiley, J., & Matsa, K. E. (2014, October 21). Political Polarization & Media Habits. Pew Research Center. Consulted on 16-04-2018. Retrieved from
http://www.journalism.org/
Niculae, V., Suen, C., Zhang, J., Danescu-Niculescu-Mizil, C., & Leskovec, J. (2015). Quotus: The structure of political media coverage as revealed by quoting patterns. In Proceedings of the 24th International Conference on World Wide Web (pp. 798- 808). International World Wide Web Conferences Steering Committee.
Ohl, C. M., Pincus, J. D., Rimmer, T., & Harrison, D. (1995). Agenda building role of news releases in corporate takeovers. Public Relations Review, 21(2), 89-101.
Reese, S.D. (2001). Understanding the Global Journalist: a hierarchy-of-influences approach.
32
Reich, Z. (2011). Comparing reporters' work across print, radio, and online: Converged origination, diverged packaging. Journalism & Mass Communication Quarterly, 88(2), 285-300.
Rowe, D. (2007). Sports journalism: still the ‘toy department’ of the news media? Journalism, 8(4), 385-405.
Schafraad, P., & Spitteler, R. (2014). Dit Wordt Het Nieuws. Amsterdam: ASCoR
Schafraad, P., van Zoonen, W., & Verhoeven, P. (2016). The news value of Dutch corporate press releases as a predictor of corporate agenda building power. Public Relations
Review, 42, 451- 458.
Seletzky, M., & Lehman-Wilzig, S. (2010). Factors underlying organizations' successful press release publication in newspapers: Additional PR elements for the evolving “Press Agentry” and “Public Information” models. International Journal of Strategic Communication, 4(4), 244-266.
Smith, J., McCarthy, J. D., McPhail, C., & Augustyn, B. (2001). From protest to agenda building: Description bias in media coverage of protest events in Washington, DC. Social Forces, 79(4), 1397-1423.
Sugden, J., & Tomlinson, A. (2007). Stories from planet football and sportsworld: Source relations and collusion in sport journalism. Journalism Practice, 1(1), 44-61. Webber, T. (2017, September 25). How Every NFL Team Responded To Trump's National
Anthem Protest Comments. National Public Radio. Consulted on 10-04-2018. Retrieved from www.npr.org
Whannel, G. (1992). Fields in vision: Television sport and cultural transformation.
Psychology Press.
Westphal, J. D., & Deephouse, D. L. (2011). Avoiding bad press: Interpersonal influence in relations between CEOs and journalists and the consequences for press reporting
33
about firms and their leadership. Organization Science, 22(4), 1061-1086.
Wong, G. (2008). The comprehensive guide to careers in sports. Burlington, MA: Jones & Bartlett Learning.
34
Appendix 1: Codebook
The aim of this thesis was to see how press releases by NFL teams can influence the amount of coverage they generate and the tone of this coverage. A quantitative content analysis was conducted of press releases published by the 32 NFL teams of the 2016 season, and of news articles published by four US media outlets. This codebook formed the basis for the content analysis.
General coding rules
- Read the press release in its entirety once before coding - Read the news article once in its entirety once before coding - Code all the text: include title, lead, and body of text
- Exclude images and other embedded media that are not text
- Code what is written, and not your own interpretations and associations
- If you are not familiar with a concept or word, use Google to understand the concept/word to be able to code it
- Use this coding form:
https://uvacommscience.eu.qualtrics.com/jfe/form/SV_eG6B3JO5KWkVCfP
1. Coding Unit: Press releases by NFL teams relating to National Anthem protests and news articles from four news outlets about the NFL teams.
V1. Is the coding unit a press release or news article? (1) Press release (2) News article V2. Sender (1) Player(s) (2) CEO (3) Owner (4) Coach (5) PR/Communications Department (6) New York Times
(7) USA Today (8) Fox News (9) ESPN (10) Other (11) Unknown
V3. Is the news article based on one of the press releases published by the NFL teams?
Skip if you are coding a press release
The succession of a press release in the news is the goal of the sports organization sending the press release. It is important to make sure that the news article actually includes (parts of) the press release and does not just talk about the team and/or the protests. To accurately code the presence of (parts of) the press release follow these guidelines:
1. The coding answers for this variable are binary: 1 = no and 2 = yes.
2. The press release does not have to be quoted in its entirety to be coded as ‘yes’
3. If any part of the press release is quoted, code it as ‘yes’
4. If (part of) the information in the press release is paraphrased, code it as ‘yes’
5. The next sentence is an example from a news outlet that should be coded as ‘yes’: ‘In a
statement Sunday morning, owner Robert Kraft said he was "deeply disappointed" by Trump's comments.’
(1) No
(2) Yes
35 → If ‘no’ stop coding
V4. Date of press release or news article
Write the date in the following format: DD-MM-YYYY
V5. Team
Indicate which NFL team published the press release or the news article is about.
If the news article is about more than one team, fill out several coding forms, one for each team. >dropdownmenu
V6. Code of press release or news article
Press releases receive a code for identification purposes. The codes consist of 2 numbers, starting with 01, and then 02, 03, etc. The code for the news articles start with the code of the press release it is based on, followed by a ‘-’, and ending with a number with 3 figures. For example, 24-097. News articles that are not based on a press release only receive the news article’s number, so from 001 upwards.
Fill in the code of the press release or news article.
V7. Incentives
Based on Aronoff (1975), Seletzky & Wilzig-Lehmann (2010), Schafraad & Spitteler (2014), and Schafraad, Van Zoonen, & Verhoeven (2016). The press releases contain one of the following characteristics that are not explicitly part of the core of the message. Multiple answers are possible.
(1) Quotes by laypersons
A quote from someone within the organisation that sent the press release is included in the text. Examples are representatives and employees.
(2) Quotes by key persons
A quote is included and clearly ascribed to an important person within the organization. These key figures are CEO, Coach, player(s), and owner.
(3) Quotes by relevant others
A quote from someone outside the organisation that sent the press release is included in the press release.
(4) Research
The press release is based on research requested by (a trade organization of) the sender of the press release.
(5) Sensation
The title of the press release consists of ‘hyperbole’, ‘cynicism’, ‘satire’ and/or the title is in capital letters, underlined, bold, or ending in an exclamation mark.
(6) Current affairs
An explicit link to current affairs is included in the press release. Do not code if the press release is about current affairs.
(7) Events
The press release contains an explicit link to an event that was organized by the sender of the press release. For example, a press conference.
(8) No incentives present
If none of the above-mentioned incentives are present in the press release.
V8. What is the overall tone towards the team?
Based on Lefevere (2011) and Evers (2016). Coding tone is difficult and relies mostly on the judgments of the individual coder. To make these judgments easier, ask yourself the question: ‘Is this news article positive or negative news for the NFL team?
(1) Negative
The words that are used to describe the main subject evoke negative associations. The main subject is exposed in a bad way. For instance: misbehaving, unthankful, unsatisfied, causing troubles or unethical.
(2) Positive
The words that are used to describe the main subject evoke positive associations. The main subject is exposed in a good way. For instance: helping, honest, satisfied or (suggesting) solutions.
36 (3) Neutral
The main subject is described in an objective manner and the story is based on facts. The words used to describe the main subject evoke no specific positive or negative associations.
(4) Mixed
The words that are used to describe the main subject evoke both negative and positive associations. Negative attributes of the main subject are mentioned, as well as positive ones. (5) N/A
No words in the press release describe an evaluation of the team.
V9. What is the overall tone towards the protest actions of the players?
Based on Lefevere (2011) and Evers (2016). Coding tone is difficult and relies mostly on the judgments of the individual coder. To make these judgments easier, ask yourself the question: ‘Is this news article positive or negative news for the NFL team?
(1) Negative
The words that are used to describe the main subject evoke negative associations. The main subject exposed in a bad way. For instance: misbehaving, unthankful, unsatisfied, causing troubles or
unethical. (2) Positive
The words that are used to describe the main subject evoke positive associations. The main subject is exposed in a good way. For instance: helping, honest, satisfied or (suggesting) solutions.
(3) Neutral
The main subject is described in an objective manner and the story is based on facts. The words used to describe the main subject evoke no specific positive or negative associations.
(4) Mixed
The words that are used to describe the main subject evoke both negative and positive associations. Negative attributes of the main subject are mentioned, as well as positive ones. (5) N/A
No words in the press release describe an evaluation of the actions that players took to protest.
V10. What is the overall tone towards President Trump('s remarks)?
Based on Lefevere (2011) and Evers (2016). Coding tone is difficult and relies mostly on the judgments of the individual coder. To make these judgments easier, ask yourself the question: ‘Is this news article positive or negative news for Trump?
(1) Negative
The words that are used to describe the main subject evoke negative associations. The main subject is exposed in a bad way. For instance: misbehaving, unthankful, unsatisfied, causing troubles or unethical.
(2) Positive
The words that are used to describe the main subject evoke positive associations. The main subject is exposed in a good way. For instance: helping, honest, satisfied or (suggesting) solutions.
(3) Neutral
The main subject is described in an objective manner and the story is based on facts. The words used to describe the main subject evoke no specific positive or negative associations.
(4) Mixed
The words that are used to describe the main subject evoke both negative and positive associations. Negative attributes of the main subject are mentioned, as well as positive ones. (5) N/A
No words in the press release describe an evaluation of President Trump(‘s remarks)
V11. To what extent is explicit support for the protesting players/team mentioned?
If an organization comes across as a unified front, it will be seen in a more positive light by its stakeholders. The media are important stakeholders and the NFL teams will want to be seen in positive light by the media. If the teams express that they accept and support their players’ action, they will be seen as a unified organizations that
