The doping issue in cycling and how it affects its
stakeholders’ media reputation
A quantitative study on how the media framing of the doping issue in
the cycling sport affected the media reputation of main involved
stakeholders from 1998 to 2018
by Sanne van `t Ooster
s11994193 Master’s Thesis
Graduate School of Communication, University of Amsterdam
Master’s programme Communication Science: Corporate Communication Supervisor: Dhr. Dr. P.H.J. Schafraad
1
Abstract
The present study reviews the influence of media framing on media reputation in the cycling sport from 1998 – 2018 as depicted in the Guardian. This is done to gain a better understanding on how cultural changes in the cycling sport manifest themselves in the media. Five frames were constructed and five stakeholders were identified through a preliminary analysis. In the next step, a content analysis was conducted over a sample of 367 articles and followed by testing the variables media framing and media reputation. The most important finding was that, to a certain extent, media framing in the cycling sport has an influence on the media reputation of the main stakeholders. Future research should focus on creating a more thorough understanding of the nature of the relationship between media framing and media reputation in the sport industry.
2
Introduction
Cycling, people in the Netherlands do it almost every day although worldwide it is mostly considered as a professional sport. There are different forms of professional cycling: road, mountain bike, cyclo-cross and indoors. For this study, the focus will be on road cycling because this branch of the cycling sport is most popular (UCI, n.d.a). Each road cycling season starts with the Tour Down Under in January and ends with the Tour of Lombardia in October (Touretappe.nl, 2019). In each race, several teams consisting of 8 cyclists compete. Examples of big races are: Giro d’Italia, Tour de France, and La Vuelta Ciclista a España; examples of famous cyclists are: Eddy Merckx (1961-1978), Lance Armstrong (1991 – 2011), and Chris Froome (2007 – present). Cycling is considered a sport with a history of
extraordinary achievements and therefore has a large fan base from all over the world. The cyclists do something no one else is able to do: cycling for hours, hundreds of kilometres, pushing the limits of human capability. Sounds magical. With the exception of one artificial influence: doping.
Doping, otherwise known as performance-enhancing drugs, is defined as “the occurrence of one or more of the anti-doping rule violations” (World Anti-Doping Agency, 2018, p. 18). The anti-doping rule violations are centred around the concept of finding prohibited substances or its metabolites in an athlete’s sample or in the possession of an athlete (World Anti-Doping Agency, 2018). Athletes from all different types of sport can use certain substances to improve their performance, which is nowadays considered as cheating. Examples of prohibited substances that are used in cycling are blood transfusions,
erythropoietin (EPO), and testosterone.
It is generally known that doping has been an important factor in the cycling sport for decades (Jones, 2010). What is the reason that doping, in any sort or form, has been present in cycling’s past and present? Moller and Hoberman (as cited in Jones, 2010) provide a plausible
3 explanation stating that because of the physical demands of cycling athletes have been
pursuing medically innovative ways to meet the challenges ever since the beginning of the sport. Doping, as pointed out by Moller and Hoberman (as cited in Jones, 2010) provide the cyclists with a way to maintain their bodies to be able to function at a professional level. The boundaries of what is morally right and wrong are, through the use of cyclists, constantly being tested.
As stated by Boyle (2012) sport, at its most basic level, offers an insight into human character, nature, and creativity. It allows people to relate their own feelings and emotions through a cultural activity. Sport rather connects with the emotional part of human nature than the rational one. The cycling sport has changed between 1998 – 2018 due to the connection with this emotional part because of the changing environment and moral opinions regarding the use of doping (Pfister & Gems, 2015). With the change in moral opinions towards a more anti-doping environment came the advancement of technical developments. For example, in the area of doping tests, which have had an influence on the cycling sport ever since.
These changes indicate that sport is still a changing cultural form that goes through every single domain in our society such as politics and economics (Boyle, 2012). It is
generally known that our society is being discussed in the media. Zhang (2016) points out that the effects of media on society have been widely researched, however research on the effects of media on businesses is lacking. In addition, research into the effects of media on a sport industry is even less present. It is therefore relevant to see whether the aforementioned changes in the cycling sport, in regards to doping, manifest themselves in the media. From a theoretical point of view it is interesting to see in which manner the stakeholders and frames mentioned in the media are being used in regards to the doping issue. This can be determined through the use of the theoretical issue arena concept. This concept has been developed by Luoma-aho and Vos (2010) but has not yet been applied in a sport industry context, nor been
4 linked to the framing concept and been compared over time. Therefore, it is interesting to see what the influence of the media is on the cycling sport industry with specific regards to the issue of doping. Concluding, the following research question has been formulated:
To what extent has the media framing of the doping issue in the cycling sport affected the media reputation of main involved stakeholders from 1998 to 2018?
Theoretical background
The following concepts are identified from the research question: crisis, issue arena, media framing, and media reputation.
Crisis
Crisis is defined in line with Coombs (2007) as an event, which occurs unexpected, that threatens an organization’s or stakeholder’s reputation by disrupting its operations. The general known definition of doping is the use of a substance or technique to illegally improve an athlete’s performance. The usage of doping can be seen as a crisis in the cycling sport because this activity threatens the loyalty of one of the biggest stakeholders in the sport, the fans (Coombs, 2007). The loyalty of these fans is one of the reasons that the sport is able to exist. However, because the doping is a continuous issue in the cycling sport and has regular outbursts, e.g. the case of Lance Armstrong (Guardian, n.d.), the main concept in this research will be issue arenas.
Issue arena
According to Luoma-aho and Vos (2010) issues form the centre of communication in today’s society. This idea is in contrast with the stakeholder model developed by Freeman
5 (1984) which is based on the idea that the organization is in control of and in the centre of all the stakeholder interactions. However, over the years this situation has changed for
organizations which has resulted in corporate communications becoming less predictable (Luoma-aho & Vos, 2010). To be able to ensure organizational survival a more dynamic stakeholder model is necessary, the issue arena (Luoma-aho & Vos, 2010).
Doping is an issue because it is continuous, as stated above, and because it is at the centre of communication in the cycling sport. Luoma-aho and Vos (2010) state that these issues are being discussed in an issue arena. Issue arenas are defined as “places of interaction where an issue is discussed by stakeholders and organizations” (p. 316). In this arena, the organization is no longer the centre but the issue is and the exchange of views on this issue take place there (Vos, Schoemaker & Luoma-aho, 2014). It therefore describes the current environment in which organizations operate better than the well-known stakeholder model from Freeman (1984). Different forms of media are the platform of this issue because the debate in the issue arena between different stakeholders takes place there. For this study, the focus will be on a traditional form of media: the online edition of a traditional newspaper. According to Luoma-aho and Vos (2010) the concept of issue arenas theoretically has many starting points and in this research the focus is on three approaches: stakeholder thinking, network theory, issue management. This theoretical background will be structured following these three approaches.
Stakeholder thinking and the network of the doping issue
Rather than, the perspective that all the stakeholders have a shared stake in the organization the issue arena concept states that, in the current environment, parties involved rather have a stake in a certain issue instead of a stake in the whole organization (Vos, Schoemaker & Luoma-aho, 2014). It is, therefore, important for any organization to identify
6 and understand the stakes each party has (Wu, 2007) and to distinguish the interrelations that exist between these stakes and their holders (Luoma-aho and Paloviita, 2010). These
interrelations operate within vast and complex networks with each holder having different interests and priorities (Roloff, 2008). The doping issue in the cycling sport is such a complex network with stakeholders having different, and sometimes conflicting, interests. Therefore, the following definition is used for stakeholder thinking in the doping issue: a stakeholder is any group or individual that can be affected by the doping issue as arisen in the network surrounding that group or individual (Roloff, 2008; Vos, Schoemaker & Luoma-aho, 2014). The stakeholders, identified in this research, have a concrete stake in the doping issue and can influence or be influenced by the doping issue (Fassin, 2009). The stakes these holders have in the doping issue are defined as “some form of investment, interests or affect related to the organization” (Luoma-aho & Vos, 2010, p. 317) and cannot be separated from the holder.
By scanning the arena of the doping issue, in this case the media, the researcher was able to distinguish five of the most important stakeholders involved in this issue: Union Cycliste Internationale (UCI), World Anti-Doping Agency (WADA), professional teams, cyclists, and race organizers. Firstly, the Union Cycliste Internationale (UCI) is the world governing body of the cycling sport which develops and oversees all forms of cycling (UCI, n.d.). The UCI (n.d.b) states that it is committed to leading the development of cycling as a competitive sport while maintaining outstanding performance and integrity. Not only manages the UCI nine cycling disciplines but it also manages its own events such as the World
Championships. For these reasons, it can be established that the UCI has several stakes that are connected to the doping issue. Examples of these stakes are reputation and future of the sport and the UCI, and funds.
Secondly, the World Anti-Doping Agency (WADA) is an international independent agency which monitors the anti-doping policies in all sports and all countries that operate with
7 the World Anti-Doping Code (WADA, n.d.). It is funded by sport movements and
governments from all over the world. Their vision is to create a world in which all athletes can compete in a doping-free environment (WADA, n.d.). The ultimate stake that the WADA has in the doping issue is that they want to ensure that doping use is punished and ultimately eliminated through which they create a better future for the sport. Therefore, their existence is directly linked to the presence of doping.
Thirdly, in the last 20 years of cycling history there have been many teams that have entered and left the competition. Not all teams have been linked to the doping issue but still too many to discuss individually. To be identified as a team it should be one of the teams that were participating in the professional cycling races of that year with at least eight cyclists. Usually these teams carry the name of their sponsor such as Team Sunweb or Team Astana. Not only cyclists are part of a team but also team doctors, managers, and coaches. A stake that the teams have in the doping issue is the choice of using doping or not for gaining a better end result in the race. Other examples are reputation and revenues.
Fourthly, the athletes in the cycling sport who are the cyclists. They are the ones training on the bike and ultimately have to deliver a certain performance. Winning for cyclists is one of the most important objectives. When doping can enhance that chance the stakes get higher for cyclists, especially in professional sports. For them it is not only about economical stakes such as money but about their entire career. Examples of cyclists are Lance Armstrong, Thomas Dekker, Bauke Mollema, and Eddie Merckx.
Lastly, not only teams, cyclists, the UCI and the WADA are involved in a race. The last identified stakeholder is the race organizers. There are many races in a season and therefore the focus will be on four races: Tour de France, La Vuelta Ciclista a España, Giro d’Italia, and the World Championships. The actual organizations behind these races, other than the Amaury Sport Organization (ASO), are often not mentioned in the media. Therefore,
8 the actual races and the ASO are considered as the race organizers. They have economical stakes with the doping issue because if cyclists are not allowed to race anymore this will immediately affect the amount of people that are going to watch. In the following paragraph the interrelations between these stakeholders will be explained.
Surrounding the doping issue, a network of different stakeholders exists. Network theory is focused on the relations between the different stakeholders within a network (Vos, Schoemaker & Luoma-aho, 2014). A network is ‘a set of interconnected nodes’ (Castells, 2000, p. 152). The nodes in the doping issue are the stakeholders previously discussed. The relations between stakeholders within a network can influence the actions of these
stakeholders. One part of this doping network has not yet been discussed: the media. The media are, however, not a stakeholder in the doping issue. Previously, it was defined that a stakeholder has a concrete stake in an issue. The media, however, do not have a concrete stake in the issue and are thus no stakeholder. But even without a stake the media are able to
influence the issue. They are therefore labelled as stakekeeper (Fassin, 2009) and they provide the platform where the doping issue is being discussed. This network is depicted in Figure 1 and inspired by Fassin’s refined stake model (2009). The figure shows that the UCI, WADA, teams, cyclists and race organizers influence the doping issue but are also influenced by the issue. The media influences the whole network surrounding the issue and the issue does not have an influence on the media. This displays that the media is the platform or arena in which the issue is being discussed. Following the figure is a further explanation of the context of the doping issue.
9
Figure 1. Network model doping issue. Issue management
Luoma-aho and Vos (2010) pointed out that for a stakeholder not only the stake is of importance but also the broader issue. An issue becomes an issue when “the battle is joined and conflict occurs” (Schattschneider, 1960, p. 74). Hence, issue management is the proactive identification and subsequent defusing of problems before they escalate into crises (Luoma-aho & Vos, 2010, p. 316). It is about proactively knowing the early warnings and
understanding signals (Luoma-aho & Vos, 2010). When such a warning is identified a
stakeholder needs to monitor and, in some circumstances, participate in the public discussion. To be able to prevent crises, or manage issues, it is important to be involved in the public discussion and proactively address the issue (Coombs, 1992). By monitoring the public discussion in the media, the stakeholder can identify what issues are being discussed or occurring and to which severity because in the current environment issues are competing for the public’s attention. For an issue to be noticed by the public the concept of framing is used (Hilgartner & Bosk, 1988).
10
Framing
Framing is “to select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described” (Entman, 1993, p. 52). For framing to occur it is necessary that a communication source presents and defines an issue (De Vreese, 2005). By presenting an issue frames occur through defining problems, diagnosing causes, making moral judgements and suggesting remedies (Entman, 1993). They highlight a certain piece of information and therefore call attention to certain aspects of the issue (Entman, 1993).
Through the process of framing the news media have found an influential way to shape the public opinion and news framing is therefore an important tool (De Vreese, 2005). The power of framing lies in communication. Communication is a dynamic process that involves frame-building and frame-setting (De Vreese, 2005; Scheufele, 2000). Frame-building refers to the process of developing a frame through the influence of certain factors. The outcomes of this building process, as stated by De Vreese (2005), are the frames that manifest in a text. Frame-setting refers to the process of interaction between an individuals’ prior knowledge and predispositions and the news media frames (De Vreese, 2005).
Ultimately the news frames may alter attitudes on an individual and societal level (De Vreese, 2005) through frame-building and frame-setting.
Media and individuals rely on this frame process to be able to interpret and evaluate information (Neuman, Just, & Crigler, 1992, in Van Gorp, 2007). These frames can be encountered on several locations in the communication processes: in the minds of media makers and the audience, in media content, and in culture (Entman, 1993). Van Gorp (2007) states that frames seem to be everywhere and no individual can state where they begin and end. He therefore follows the idea that frames have an important connection with culture.
11 Entman (1993) states that each culture has a set of commonly used frames that have the ability to explain important issues to that culture. It is thus logical to follow Van Gorp (2010) in that the frame-building process as stated by De Vreese (2005), in the news media, “is at its core process in which cultural values and norms are reproduced” (p. 88). Sport is a part of our culture (Boyle, 2012) and its cultural values and norms are therefore reproduced in the frame-building process of the news media. It can therefore be said that the frames used for sport are culturally embedded frames. Sport is highly discussed in the news media and thus the
culturally embedded frames, identified by Van Gorp (2007), are part of the frame-setting process that De Vreese (2005) discusses.
Culturally embedded frames are difficult to distinguish according to Van Gorp (2007). Van Gorp (2007) shows that they are possible to reconstruct by looking at the frame packages in a text. A frame package is “a cluster of logical organized devices that function as an
identity kit for a frame” (Van Gorp, 2007, p. 64). Such a package manifests itself through framing devices and reasoning devices. Framing devices are manifest elements, such as metaphors, that function as indicators of the frame (Van Gorp, 2010). Reasoning devices are the defining functions of a frame as identified by Entman (1993) which are described in the previous paragraph. The latter does not have to be explicitly present in a text. The
reconstructing of the culturally embedded frames of the doping issue is described in the Method section and can be seen in the frame matrix included in Appendix A. In the following paragraph the frames that have been identified are summarized.
The five frames that were constructed:
1. Doping wins the race (frame 1). The problem definition in this frame is that doping is
necessary to be able to win cycling races. This follows the causal interpretation that all big winners in cycling use it. The moral evaluation behind this is that winning is
12 important for athletes. Suggested solution is that if you want to win you have to use doping.
2. Two speed cycling (frame 2). The frame problem is that there is no longer a level
playing field in professional cycling. This is caused by the fact that some cyclists use doping while others do not. The moral evaluation that is present is that cheating is wrong. A solution for this problem is that the policy in place should be stricter.
3. Scapegoat (frame 3). The problem is defined as cyclists are the ones to blame for the
doping issue because they are the ones taking the doping. The moral evaluation here is that the cycling sport should be completely transparent and honest even though there are a few individuals that mess up. A suggested solution is to catch the cyclists that are using doping.
4. Crooked world of professional cycling (frame 4). Proposed problem definition is that
doping is ingrained in the culture of cycling because there is a social acceptance of doping in the peloton. Moral evaluation in this frame is that doping is part of the history of the sport. Everyone knows it but does not speak out, there is an omerta present. Suggested solution here is to see if it is possible to solve something that is so ingrained in the culture.
5. Progress (frame 5). Science and laws have solutions for the doping problem but there
is a lack of faith in the possibility of science to catch up with doping. The sport should morally strive for a future without doping which can be realised through creating a greater trust in tests and laws.
The essence of framing lies in social interaction (Van Gorp, 2007). Key events in social interaction can influence the news frames in the sense that they can activate alternative frames, certainly if an event has become a part of collective memory (Van Gorp, 2007;
13 Scheufele, 2004). In the doping issue the event of the confession of Lance Armstrong could be seen as such a key event. Within the issue arena of doping news media interact with their sources in cycling, fans of cycling interact with the media and each other, and key events happen. Therefore, it is important to include the aspect of time. Over time the use of the five identified frames could change, due to the occurrence of key events. Concluding, the
following sub-questions are asked:
Q1. What frame(s) have been used the most for the doping issue between 1998 - 2018? Q2. How has the usage of the frames changed over time?
Q3. Which frames are used by which stakeholders and do they differ over time?
Media Reputation
Through the use of frames by the media a certain reputation surrounding a stakeholder can be formed in the public sphere; this reputation is called media reputation. Media
reputation is a concept constructed by Deephouse (2000). It is considered to be the “overall evaluation of a firm presented in the media” (Deephouse, 2000, p. 1091). This study will be using the concept of media reputation because Deephouse (2000) considers media reputation to be an important form of reputation and something that, in the current environment, can have large implications for stakeholders of a certain issue. This leads to the following sub question:
Q4. Does media framing influence the media reputation of the stakeholders involved in the doping issue?
14 As Deephouse (2000) points out media reputation connects the stakeholder, the media, the sources, and the reader. All of these stakeholders do not develop an opinion at one point in time but media reputation develops over time through a complex social process including these stakeholders (Deephouse, 2000). This is an important characteristic of reputation in general (Fombrun & Shanley, 1990). Therefore, it is possible that media reputation changes over time. This leads to the following sub question:
Q5. Does the media reputation of the UCI, WADA, teams, cyclists, and race organizers change over time?
Method
To be able to answer the research question and sub-questions the method content analysis was used.
Data and Time Period
This study was focused on news articles from the English newspaper ‘The Guardian’. The reason for choosing this newspaper was that The Guardian is generally known as one of the biggest English newspapers that cover cycling extensively. The time period chosen was from 1998 to 2018. This time period is a total of 20 years because doping issues have been occurring continuously through this period of cycling’s history. A major event happened in 1998 when a whole team, the Festina team, was excluded from the Tour de France because one of the team doctors was caught with doping (Guardian, 2018). Therefore, this will be the start for this study and 2018 will be the end to include full years and doping has been a relevant issue till today.
15
Sample characteristics
The articles from The Guardian were downloaded from the LexisNexis database with the search string ‘doping’ AND ‘cycling’ from 1998 to 2018. This search string was chosen to keep the population of articles as general as possible. This search resulted in 1,000 articles and to reduce this several criteria were applied. The final sample of articles had to contain both the words ‘doping’ and ‘cycling’, and this topic should be discussed in more than one paragraph. In addition to these criteria a time criterium was added. Articles in the dense period from the 1st of June to the 31st of August were all used. This dense period was chosen because in this period the largest cycling competition takes place: the Tour de France. This would result in more media coverage on the topic of cycling and the recalling of certain doping scandals.
These criteria resulted in a total sample of 241 articles. However, because of this dense period certain important topics in the doping issue were excluded. To solve this problem certain periods were added: September 2007, May 2010, September 2010, May 2011, October 2012, January 2013, March 2015, December 2017 and October 2018. In Table A3 the reasons for adding these periods are described. After adding the articles in these periods to the total sample this resulted in 367 articles. No additional sampling methods were applied to this total sample. All the articles, with the correct criteria, were used. Each whole article was a
recording unit. The period of 20 years was recoded into three time periods of six years for the analyses: period 1 from 1998 – 2004, period 2 from 2005 – 2011, and period 3 from 2012 – 2018.
Media Frames
The independent variable frames that have been used in this study, were applied through two phases following the method of Van Gorp (2010). Firstly, the inductive phase
16 was set up. The researcher analysed a random sample of 47 articles and open coded these articles. This means that the articles were analysed without using a predetermined coding tool (Van Gorp, 2010). After having open coded the articles the researcher looked for patterns within these codes by linking them to overarching ideas. Examples of these ideas were: cyclists use the doping and doping is ingrained in the culture of the sport. These patterns were used to fill out a frame matrix (which can be seen in Appendix A1). Each frame was based on the four elements of a frame constructed by Entman (1993): problem definition, causal
interpretation, moral evaluation, and suggestions for solutions.
In the next step, this populated frame matrix was tested to evaluate the suitability of the frames. The quality of the frames was assessed through the process of reading new articles and determining whether in these articles the frames also occurred. In this test each frame was recognized. The frames that were determined have been discussed in the previous section. After the inductive phase these frames were used in the deductive phase: the codebook. For each frame, four questions were constructed to be able to determine the presence of a frame in an article. For each stakeholder, the question was added whether this stakeholder could be linked to this specific frame through one of these four questions. This resulted in a total of nine questions per frame. Which questions were asked for which frame can be found in the codebook in Appendix A2.
Media Reputation
The dependent variable media reputation was coded in accordance with the method of Deephouse (2000). Firstly, the question was asked how many stakeholders were mentioned in the article. Following this, for each identified stakeholder, the question was coded: Is this recording unit favourable, unfavourable or neutral towards stakeholder X? For example,
17 identifying the media reputation of the UCI as a stakeholder the following recording unit outcomes can apply:
If the recording unit is favourable towards the UCI then the UCI is praised for its actions regarding the doping issue or is linked to past actions that have been indicated to increase the organization’s reputation.
If the recording unit is unfavourable then the UCI is criticized for its actions regarding the doping issue or is linked to past actions that have been indicated to decrease the organization’s reputation.
Neutral is considered to be a balanced view on the doping issue without any evaluative modifiers.
Coding Procedures
All the articles were coded by one coder, the researcher herself. To establish intercoder reliability ten percent of the total sample, 37 articles, was coded by one additional coder. These two coders have tested the developed coding procedures and categories through this double-coding. For the intercoder reliability it was decided to use the percentage agreement scores and not the Kappa scores. The reason for this decision was that the Kappa scores were extremely low due to lack of coder training and the random sample of 37 was too small and unequally distributed over the years. As a consequence, the problem of overcorrection for the Kappa scores occurred (Schafraad, 2009). When comparing these scores to the percentage agreement a relatively different conclusion could be drawn. The percentage agreements ranged from 46.2% to 100%. The intercoder reliability scores can be found in Appendix B in Table B1 for the frames and Table B2 for media reputation. Below 50% was considered an inadequate percentage agreement. Only three scores did not make this benchmark these are
18 indicated by an asterisk in the tables. Even though, they were included in the analysis because it could be explained with lack of coder training.
Results
In the following section the five sub-questions earlier discussed were analysed through a variation of tests. To be able to analyse these sub-questions the computed frame variables (which have been discussed in the previous method section) were used. All tables mentioned, that are not in the text, can be found in Appendix B.
Q1. What frame(s) have been used the most for the doping issue between 1998 – 2018?
For the first sub-question a descriptive test was chosen to determine the mean and standard deviation of each computed frame. These scores can be found in Table 1 below. For all five frames the range of the scores was between 0 and 1. As can be seen in the table, it can be said that the frame most used, in regard to the doping issue, is the “Progress” frame (M = 0.39, SD = 0.28). The least used frame is the “Doping wins the race” frame (M = 0.09, SD = 0.24).
Table 1. Descriptive scores of the five frames.
Mean (M) Standard Deviation (SD)
F1. Doping wins the race
0.09 0.24
F2. Two speed cycling 0.38 0.29
F3. Scapegoat 0.35 0.31 F4. Crooked world of professional cycling 0.34 0.36 F5. Progress 0.39 0.28 Total (N) 367
19
Q2. How has the usage of the frames changed over time?
To be able to analyse sub-question two, a One-way Analysis of Variance (ANOVA) was used. This test was used to compare the effect of time on the five different frames. Each frame will be discussed separately. All results of the ANOVA can be found in Table B3. Levene’s test for equality of variances was significant for frame 1 (F (2, 364) = 17.56, p < .001) and frame 2 (F (2, 364) = 9.55, p < .001). Therefore, for these two frames equal
variances cannot be assumed. However, non-significant results were found for frame 3 (F (2, 364) = 1.87, p = .156) , frame 4 (F (2, 364) = 0.91, p = .402), and frame 5 (F (2, 364) = 1.56,
p = .211). Concluding that, for these three frames equal variances could be assumed. The tests
were performed either way. All trends in frame usage over time are depicted in Figure 2 in percentages. In the following paragraphs the significant differences are discussed.
Figure 2. Trends in frame use in percentages shown over time.
As shown in Figure 2 usage of frame 1 over time decreases. This decrease was found with a small significant effect (F (2, 364) = 7.78, p < .001, ᶇ2 = 0.04). Specifically, between 1998 – 2004 and the other two time periods the occurrence of frame 1 was higher. However,
0,00% 10,00% 20,00% 30,00% 40,00% 50,00% 60,00% 70,00% 80,00% 90,00% 100,00%
Period 1 (1998 - 2004) Period 2 (2005 - 2011) Period 3 (2012 - 2018)
20 period 2 did not differ significantly from period 3. Post hoc comparisons using the Bonferroni post hoc test indicated that the mean score of frame 1 in period 1 (1998 – 2004; M = 0.22, SD = 0.34) was significantly different from the mean score of frame 1 in period 2 (2005 – 2011;
M = 0.11, SD = 0.25 , p = .019) and the mean score of frame 1 in period 3 (2012 – 2018; M =
0.06 , SD = 0.19, p < .001). However, this significance did not occur between period 2 and 3. Taken together, these results indicate that time does have a small effect on frame 1.
For frame 2 the highest usage is in period 2, as seen in Figure 2. A small significant effect of time on frame 2 (F(2, 364) = 4.87, p = .008, ᶇ2 = 0.02) was found. More specific, the difference takes place between period 2 and period 3 which means that frame 2 was used more in period 2 than in period 3. Bonferroni post hoc test was performed and indicated that the mean score of frame 2 in period 2 (2005 – 2011); M = 0.43, SD= 0.30 ) was significantly different from the mean score of frame 2 in period 3 (2012 – 2018; M= 0.34 , SD= 0.36, p = .012). The mean score of frame 2 in period 1 (1998 – 2004); M = 0.43, SD = 0.43) did not significantly differ from period 2 and period 3. Concluding, these results show that time does have a small effect on frame 2.
For frame 3 the same trend is shown in Figure 2. As for frame 2, the results indicate that the occurrence of frame 3 in period 2 was higher than in period 3. A small but significant difference in the use of frame 3 was found between period 2 and 3 (F (2, 364)= 6.73, p = .001, ᶇ2
= 0.04). Bonferroni post hoc test showed the exact place where this significant effect occurred. The mean score of frame 3 in period 2 (2005 – 2011; M = 0.42, SD = 0.33) differs significantly from the mean score of frame 3 in period 3 (2012 – 2018; M = 0.30, SD = 0.30, p = .001). The mean score of frame 3 in period 1 (M = 0.32 , SD = 0.28) did not conclude a significant result.
Frame 4 depicts a different trend with period 2 being the lowest point in usage. In period 1 the occurrence of frame 4 was higher than in period 2. Again, a small significant
21 difference was found between the usage of frame 4 and time (F (2, 364)= 3.76, p = .024 , ᶇ2 = 0.02 ). After looking at the Bonferroni post hoc test the specific place of significance was determined. The mean score of frame 4 in period 1 (1998 – 2004; M = 0.48 , SD = 0.37) differed significantly from the mean score of frame 4 in period 2 (2005 – 2011; M = 0.30 , SD = 0.34 , p = .020). Between the mean score of frame 4 in period 3 (M = 0.35, SD = 0.36) no significant results were found. Interpreting these results, it can be stated that the occurrence of frame 4 in period 1 differed compared to period 2 but not to period 3. Adding to that, period 2 and period 3 also did not differ from each other.
Lastly, for the trend shown for frame 5 in Figure 2 no significant result was found (F (2, 364 = 2.66, p = .071). Therefore, no relationship between the usage of frame 5 and time can be concluded.
Q3. Which frames are used by which stakeholders and do they differ over time?
The third sub-question was analysed through looking at the percentages of the amount of times each frame could be linked to one of the stakeholders. In the following table the results can be found.
Table 2. Percentages of use of frame by each stakeholder (N = 367).
F1 F2 F3 F4 F5 UCI 0.3% 19.9% 17.4% 10.9% 30.8% WADA 0.3% 9.3% 8.4% 4.1% 16.1% Teams 2.2% 15.0% 13.6% 8.2% 15.8% Cyclists 10.6% 36.2% 42.0% 22.6% 24.5% Race organizers 1.1% 11.7% 7.6% 7.1% 12.0%
22 As can be seen in Table 2 the UCI (30.8%), WADA (16.1%), teams (15.8%), and race organizers (12%) use frame 5, the “Progress” frame, the most. Cyclists, however, use frame 3, the “Scapegoat” frame, the most with 42%. All stakeholders use frame 1, the “Doping wins the race” frame, the least.
A next step is to determine whether the use of frames by the stakeholders differs over time and whether a trend could be determined. This was done through a chi-square test of independence with the independent ordinal variable time and the dependent dummy variable can the stakeholder (UCI, WADA, teams, cyclists, race organizers) be linked to at least one of the questions of the frames. For each stakeholder and each frame a separate cross tabulation was computed. In Tables B5 to B9 the results can be found of the chi-square tests and the measure of association of Somer’s d when significant results occurred. In Table B4 the possible trends over time are depicted in numbers. The following Table 3 shows these trends in a more concise manner.
Table 3. Trend per stakeholder for each frame over time.
F1 F2 F3 F4 F5 UCI - x - -/+ x WADA x x x x x Teams - x x x x Cyclists - x x x x Race organizers - +/- +/- - -
Note. x stands for not significant, + stands for significant and positive change over time, -
stands for significant and negative change over time. When a combination is shown the periods differed from each other in the direction from left to right.
23 To examine the relation between time and the use of the five frames by the
stakeholders five chi-square tests of independence were conducted. For the frame “Doping wins the race” (frame 1) several significant trends occurred. As can be seen in Table 3 all these significant trends were negative. The relation between the use of this frame by the UCI and race organizers over time showed significant results (X2 (2) = 8.68, p = .013; X2 (2) = 8.28, p = .016. The relation between the use of frame 1 by the teams and cyclists over time displayed significant results as well (X2 (2) = 14.73, p = .001; X2 (2) = 7.29, p = .026). The use of frame 1 by WADA over time did not provide significant results. For the significant results associations were tested with Somer’s d to determine the strength. Between the use of frame 1 by the UCI and race organizers and time small, negative associations occured. However, this association was for both stakeholders not significant (d = -.008, p =.315; d = -.024, p = .058). The use of frame 1 by the teams and cyclists resulted in small, negative significant
associations (d = -.040, p = .027; d = -.074, p = .017). A general trend is shown that the use of frame 1 decreases over time, as shown in Table 3 and in Appendix B in Table B4.
The frame “Two speed cycling” (frame 2) provided one significant result. Only the use of the frame by race organizers over time was a significant relationship (X2 (2) = 22.41, p = .016). The significant association between time and the use of the frame by race organizers was relatively small and negative (d = -.024, p < .001). The use of frame 2 by other
stakeholders could not be related to the variable time. It is, nevertheless, interesting to see (in Table 4) that only WADA shows an increase in the use of frame 2 after 2011, whereas the other stakeholders show a decrease from that point onwards.
Two significant results were found for the frame “Scapegoat” (frame 3). The use of frame 3 by the UCI and race organizers showed a relation with time (X2 (2) = 14.59, p = .001;
X2 (2) = 23.34, p < .001). Both relationships had a relatively small, negative and significant association (d = -.135, p < .001; d = -.106, p < .001). Interesting to see is that between 2004
24 and 2011 only the UCI has used frame 3 significantly less compared to the others.
The second to last frame “Crooked world of professional cycling” (frame 4) also showed a relation between the use of the frame by the UCI and race organizers and time (X2 (2) = 11.74, p = .003; X2 (2) = 12.05, p = .002). The association between the use of the frame by the UCI and time was small and positive but not significant (d = .045, p = .160). In contrast to that, the association between the use of the frame by the race organizers and time was negative and significant (d = -.083, p = .001). It can be seen that from 1998-2004 four out of five stakeholders (except WADA) used frame 4 the most in comparison to the other
periods.
The last frame, “Progress” (frame 5), resulted in one significant outcome as shown in Table 3. A relationship between the use of the frame by race organizers and time was
determined (X2 (2) = 20.34, p < .001). The association between these two variables was relatively small, negative and significant (d = -.134, p < .001). For all stakeholders a decrease is seen after 2011 for the use of this frame.
Q4. Does media framing influence the media reputation of the stakeholders involved in the doping issue?
Before the relationship between media framing and media reputation is analysed a table is shown depicting the descriptive analysis of the media reputation per stakeholder in percentages.
25 Table 4. The descriptive analysis of the media reputation per stakeholder in percentages.
Unfavourable Neutral Favourable
UCI 36.6% 49.4% 14%
WADA 13.3% 54.7% 32%
Teams 38.5% 51.1% 10.4%
Cyclists 48.3% 43.6% 8.2%
Race organizers 14.6% 48.3% 37.1%
To analyse whether there is a relationship between media framing and the media reputation of the stakeholders an ordinal logistic regression was performed. It was used to see whether several independent interval variables (frames) predicted the ordinal dependent variable media reputation. Media reputation was divided per stakeholder which resulted in five separate ordinal logistic regressions. Tests to see if the data met the assumption of collinearity indicated that multicollinearity was not a concern, as is shown in Table B10 to B14.
At first, the UCI was analysed. The ordinal logistic regression between the five frames and the UCI’s media reputation gives better predictions than if marginal probabilities were just guessed, X2 (5) = 51.41, p < .001. The goodness-of-fit provided a non-significant result (X2(235)= 259.66, p = .129) and therefore the observed data is consistent with the fitted model. The five frames explain a relatively large proportion of the variation between media reputation as is indicated by Nagelkerke, which is 31.2%. The test of parallel lines is not significant, X2(5)= 1.92, p = .860, and therefore the assumption of proportional odds is met. Frame 3 (b = 1.960, SE = 0.570, OR = 2.752 , p = .001), frame 4 (b = -2.172, SE = -.497, OR = 0.195, p < .001), and frame 5 (b = 1.839, SE = 0.636, OR = 2.820, p = .004) predicted the media reputation of the UCI. Each point of increase in frame 3 and 5 was associated with
26 respectively 2.752 and 2.820 increase in the media reputation of the UCI compared to the lower categories. In contrast, each point of increase in frame 4 was associated with 0.195 decrease in the media reputation of the UCI compared to the lower categories. So, by using frame 3 and 5 the chance of getting a higher reputation is significantly more likely. By using frame 4, however, the chance of getting a lower reputation is significantly more likely.
Secondly, WADA was analysed. The model fit is accepted because of the following significant result, X2 (5) = 11.51, p = .042. As was the same for the UCI, the goodness-of-fit provided a non-significant result (X2(123)= 131.05, p = .293) through which can be
concluded that the data is consistent with the fitted model. Nagelkerke is somewhat lower at 16.7%, which means that the five frames explain a relatively small proportion of the variation between media reputation. The assumption of proportional odds is again met with a non-significant result, X2(5)= 3.21, p = .668. Only frame 5 (b = 2.763, SE = 1.068, OR = 6.811, p = .010) predicted the media reputation of the WADA. Each point of increase in frame 5 was associated with 6.811 increase in the media reputation of the WADA compared to the lower categories. So, by using frame 5 the chance of getting a higher reputation is significantly more likely.
Thirdly, the model fit for teams is accepted because of a significant result, X2 (5) = 11.73, p = .039. Again the goodness-of-fit provided a non-significant result (X2(213)= 221.72,
p = .327) which shows that the data fits the model. A very small proportion of variation
between media reputation is explained by the five frames for teams with Nagelkerke being 9.8%. The assumption of proportional odds is met because the test of parallel lines is not significant, X2(5)= 3.62, p = .605. Only one frame predicted the media reputation of the teams, frame 4, b = -1.085, SE = 0.543, OR = 0.351, p = .046. Each point of increase in frame 4 was associated with 0.351 decrease in the media reputation of the teams compared to the lower categories. Concluding that by using frame 4 the chance of getting a lower reputation is
27 significantly more likely.
The fourth stakeholder that was analysed is the cyclists. The model fit is again accepted through a significant result, X2 (5) = 23.82, p < .001. A different goodness-of-fit result is found here with a significant result (X2 (413) = 472.66, p = .022) which means that the observed data is not consistent with the fitted model. Therefore, the results of this test should be observed cautiously. When looking at Nagelkerke it can be seen that 8.6% of the five frames explain the proportion of the variation between media reputation which is
relatively small. The proportional odds assumption is met with a non-significant result, X2(5)= 8.92, p = .112. Frame 3 (b = -1.219, SE = 0.406, OR = 0.667, p = .003), frame 4 (b = -1.058,
SE = 0.338, OR = 0.475, p = .002), and frame 5 (b = 0.958, SE = 0.433, OR = 1.186, p = .027)
predicted the media reputation of the cyclists. Each point of increase in frame 3 and 4 could be linked to 0.667 and 0.475 decrease in the media reputation of cyclists. In contrast to these results, frame 5 was approximately 18.6% more likely to increase the media reputation of the cyclists.
Lastly, the race organizers was analysed. In this case the model fit provided a non-significant result, X2 (5) = 8.26, p = .143. This means that the model does not give better predictions than when marginal probabilities were guessed. Therefore, the results of this test should be observed cautiously. In contrast to that, the goodness-of-fit did provide the non-significant result needed, X2 (155) = 162.86, p = .317. When looking at Nagelkerke, it can be stated that a small portion of 10.2% of the five frames explain the proportion of variance between media reputation. The assumption of proportional odds was also met in this test with a non-significant result, X2 (5) = 3.69, p = .595. Only frame 4, b = -1.372, SE = 0.63, OR = 0.180, p = .029, predicted the media reputation of race organizers. Each increase in frame 4 resulted in a 0.180 decrease in the media reputation of race organizers.
28 Table 5. Prediction of media reputation of each stakeholder through the use of the five
frames. F1 F2 F3 F4 F5 UCI x x + - + WADA x x x x + Teams x x x - x Cyclists x x - - + Race organizers x x x - x
Note. x stands for not significant, + stands for an increase in media reputation, - stands for a
decrease in media reputation.
One overall observation that can be made from Table 5 is that “Doping wins the race” (frame 1) and “Two speed cycling” (frame 2) provided no significant results for media reputation. Therefore, it can be stated that at least the framing of these frames did not have an influence on the media reputation of the stakeholders. What is also shown in Table 5 is that “Crooked world of professional cycling” (frame 4) only predicts the media reputation of the stakeholders in a negative manner. In contrast to this, is “Progress” (frame 5) which only predicts the media reputation of the stakeholders in a positive manner.
Q5. Does the media reputation of the UCI, WADA, teams, athletes, and race organizers change over time?
To determine whether an association is present between the ordinal independent variable time and the ordinal dependent variable media reputation for each stakeholder a cross tabulation was used with a chi-square test of independence. This resulted in five different cross tabulations. Each chi-square test resulted in a non-significant result. These numbers can
29 be found in Table B15. This means that no association is present between time and media reputation. It should be noted that for two of the tests the assumption, that the cells should not have more than 20% expected count less than 5, was violated. A reason for this could be that a stakeholder was not always mentioned in an article and therefore the level of empty cells could have been higher.
Discussion & Conclusion
In the following section a summary of the results is presented and the research question will be answered. Furthermore, the link to the theory, limitations, practical implications, and recommendations for future research will be discussed.
Discussion
To be able to determine whether the media framing of the doping issue affected the media reputation of the main involved stakeholders over time several sub-questions were constructed and analysed. To begin, the frequency of the frames was analysed and the conclusion was made that the “Progress” frame was used most often in the period from 1998 – 2018 and the “Doping wins the race” frame the least.
Secondly, it was analysed whether the use of the frames changed over time. For four out of five frames, it can be stated that they changed over time, only the “Progress” frame did not. This depicts the unstable form that sport is according to Boyle (2012). Sport is constantly changing due to moral opinions and the changes in usage of frames over time show this unstable form.
This can be explained more in detail by Van Gorp (2007), who stated that key events do have the power to change the framing used. The “Doping wins the race” frame occurrence was higher between 1998 – 2004 than from 2004 onwards. This can be explained through the
30 key event of the Festina affair in 1998 and the following discoveries surrounding the amount of doping that was used. The “Two speed cycling” frame and “Scapegoat” frame occurred more often between 2004 – 2011 than after 2011. In the years from 2004 onwards the amount of doping scandals that came to light due to cyclists getting caught was high. Cyclists were usually the ones that got blamed and the comparison was more often made between cyclists that apparently did use and the ones that did not. The “Crooked world of professional cycling” frame appeared more often from 1998 – 2004 than from 2004 – 2011 but did not differ from the third period. Again, in 1998 the world got to know the extent of the doping issue in the cycling sport due to the Festina affair. The reason for this was that more evidence came out about what actually went on in the culture of the cycling sport.
Following, which frames were used by which stakeholders was given a more detailed look. The UCI, WADA, teams, and race organizers used the “Progress” frame the most, whereas cyclists used the “Scapegoat” frame the most. The “Doping wins the race” frame was the least used by every stakeholder.
Adding to this was the possible change over time in usage of frames by the
stakeholders. The usage of the “Doping wins the race” frame decreased over time for all the stakeholders except the WADA. The “Two speed cycling” frame only showed a relationship with time for the race organizers with a decrease of the usage after 2011. A relationship for the “Scapegoat” frame and time was found with the UCI and race organizers. Only for the UCI the use of this frame decreased between 2004 – 2011. For race organizers, this decrease only occurred after 2011. The “Crooked world of professional cycling” frame showed a relationship with time for the UCI and race organizers. For the UCI, there was first a decrease and after 2011 an increase occurred. For race organizers, a constant decrease was shown. A decrease over time only occurred with race organizers for the “Progress” frame. The usage of the five frames by the WADA did not change over time.
31 To be able to understand these differences in usage of frames by the stakeholders and changes over time it is critical to turn to the issue arena dynamics described by Vos,
Schoemaker and Luoma-aho (2014). As explained by these authors in the issue arena it is important to understand the network of the doping issue. Each stakeholder, as discussed in the theoretical background, has different stakes in the issue. The UCI, WADA, teams, and race organizers are all important parts to the culture of cycling and therefore one of their important stakes is the future of the sport. The “Progress” frame is the frame that considers the future of the sport as important and it is therefore logical that these stakeholders use this frame the most. However, it is surprising to see that after 2011 the use of this frame decreased for all stakeholders. Cyclists are the ones that are racing and are considered the ones taking the doping. They are, therefore, considered the culprits and often have the opinion that they are used as scapegoats for the broader doping issue.
The changes over time can be explained through the network theory part of the issue arena (Vos, Schoemaker & Luoma-aho, 2014). The doping network consists of several nodes, e.g. the stakeholders and the media, that are connected and interact with each other and the issue in general. The stakeholders participate in the public discussion of the issue as well (Luoma-aho & Vos, 2010). This public discussion about doping has changed between 1998 – 2018 (Pfister & Gems, 2015). In accordance with this it can be seen in the results that over time the usage of the frames by the stakeholders changes because of the difference in
interaction of nodes in the doping issue and the changing public opinion of the doping issue. After determining the changes over time in the use of frames the possible influence of media framing on media reputation was analysed. A surprising result was that the “Doping wins the race” and “Two speed cycling” frame did not predict the media reputation of any of the stakeholders. One explanation for this could be that the “Doping wins the race” frame was the least used frame and the “Two speed cycling” frame depicts a more balanced view by
32 stating the doping and anti-doping side.
The “Scapegoat”, “Crooked world of professional cycling”, and “Progress” frame all predicted the media reputation of the UCI and the cyclists. For the UCI only the “Crooked world of professional cycling” did this in a negative manner whereas for the cyclists this was the “Progress” frame. It is an expected result that for the UCI the “Crooked world of
professional cycling” provided a negative result. This specific frame focused on the negative side of the culture in cycling and the UCI makes the rules for the culture of cycling so when this is put in a negative light the UCI is put in that light as well. Whereas, in the “Scapegoat” frame the blame is put on the cyclists and not the culture, and the “Progress” frame talks about cleaning up the sport. Therefore, a positive influence was found in relation to the UCI for both of these frames. The “Scapegoat” frame has a negative influence on the media reputation of cyclists because in this frame they are blamed for the doping issue and as a consequence are put in a bad light. The “Progress” frame is negative, for the media reputation of cyclists, probably because they are usually stated as the ones using the doping and
therefore not relatable to the changes that have to be made.
The “Crooked world of professional cycling” frame predicted the media reputation of teams and race organizers in a negative manner. An explanation for this result could be that the teams and race organizers are an important aspect within the culture of cycling. In this specific frame, the culture of cycling is criticized and with that its main aspects.
These results show that frames have an influence on media reputation (Deephouse, 2000). A relation has been proven between framing and media reputation of stakeholders. Hence, it gives further empirical evidence how media reputation behaves from a different perspective, namely the sport industry.
It was also interesting to determine whether the media reputation of the stakeholders changed over time. However, no significant results were found which means media reputation
33 did not change over time. Deephouse (2000) indicated that media reputation is a concept that develops over time because of a complex social process. These results, however, indicate that such a change of media reputation does not occur. A possible explanation for this result could be that the doping issue is such a normal concept in cycling that it cannot affect the media reputation of the sport in general and hence the media reputation of the stakeholders will not differ over time.
Concluding, to answer the main research question, the media framing of the doping issue in the cycling sport affects the media reputation of main involved stakeholders from 1998 to 2018 to a certain extent. However, results differed over the five frames and main stakeholders which shows that it is important to understand the relations within an issue. This research provided evidence that the issue arena theory is a theory that can be used to better understand the dynamics that occur within an issue such as doping.
Implications
The findings of this research have shown that the framing in the media of certain issues has a relative influence on the media reputation of stakeholders in the cycling industry. It shows the cycling industry that the media apparently can have an influence. An increase or decrease in usage of specific frames by a specific stakeholder may have an influence on the media reputation of that stakeholder. Concluding, this could indicate that stakeholders in the cycling industry should see framing as an important tool in issues such as doping when considering their media reputation.
Limitations
The findings of this research have to be considered while taking several limitations into account. Firstly, this research only used articles from the Guardian. Even though, this is
34 one of the main newspapers that reports cycling it is still important to note that when more newspapers would have been used the generalizability of the results would have been better. Secondly, the coder training that was done before the intercoder reliability should have been more extensive. Which means that a robust evaluation of the reliability of the coding was not possible and as this was an explorative analysis, future work should further validate the frames and their measurement. Thirdly, although this research contributes to a better understanding of the issue arena concept and the cycling industry the amount of empirical research on the specific topic discussed is low. Therefore, comparable material was lacking which could have resulted in less strong conclusions.
Future research
Future research should focus on broadening empirical research regarding the sport industry in general (so not only looking at cycling) and the issue arena concept. This research has shown the dynamics involved in the doping issue arena. It has shown evidence that certain frames are used differently by several stakeholders and change over time. These results show that the issue arena theory can be used to better understand issues in sport such as doping. Through the use of the issue arena concept a better understanding of the dynamics in the sport industry can be formed. Another point of interest for future research could be the relationship between framing and media reputation. More empirical evidence is needed to better
understand this relationship.
Conclusion
To conclude, this study expands the knowledge of media framing on media reputation. Specifically, looking at the cycling sport industry and its doping issue over a period of 20 years. The findings portrayed that the “Progress” frame was the most used frame and that all
35 frames were used differently by each stakeholder. The usage of these frames by the
stakeholders showed differences over time. In contrast, media reputation did not depict differences between 1998 and 2018.
Notwithstanding, the most important finding that media framing does have an
influence on the media reputation of the UCI, WADA, teams, cyclists and race organizers in the cycling sport. This implicates to the sport industry, and the cycling sport in particular, that when dealing with the doping issue the framing by stakeholders and/or the media can have an influence on the media reputation of the sport and its stakeholders. Nonetheless, more
research is needed with the use of more newspapers and better coder training to verify these results.
Future research should focus on the sport industry in relation to the issue arena
concept, framing, and media reputation. Finally, this research has expanded the knowledge of framing and media reputation in the field of sport.
References
Benford, R.D. & Snow, D.A. (2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26, 611-639.
Boyle, R. (2012). Reflections on communication and sport: On journalism and digital culture. Communication & Sport, 1(1/2), 88-99. doi: 10.1177/2167479512467978 Castells, M. (2000). The contours of the network society. The Journal of Future Studies,
Strategic Thinking and Policy, 2 (2), 151-157. doi: 10.1108/14636680010802591
Coombs, W.T. (1992). The failure of the task force on food assistance: a case study on the role of legitimacy in issue management. Journal of Public Relations Research, 4 (2), 101-122. doi: 10.1207/s1532754xjprr0402_03
36 Development and Application of Situational Crisis Communication Theory. Corporate
Reputation Review, 10 (3), 163 – 176. doi: 10.1057/palgrave.crr.1550049
Deephouse, D.L. (2000). Media Reputation as a Strategic Resource: An Integration of Mass Communication and Resource-Based. Journal of Management, 26 (6), 1091 – 1112. doi: 10.1016/S0149-2063(00)00075-1.
De Vreese, C. H. (2005). News framing: Theory and typology. Information design journal &
document design, 13(1).
Entman, R.M. (1993). Framing: Toward clarification of a fractured paradigm. Journal
of Communication, 43 (4), 51-58.
Fassin, Y. (2009). The stakeholder model refined. Journal of Business Ethics, 84, 113-135. doi: 10.1007/s10551-008-9677-4
Freeman, E. (1984). Strategic Management: A Stakeholder Approach. Pitman, Boston, MA. Fombrun, C., & Shanley, M. (1990). What’s in a name? Reputation building and corporate
strategy. Academy of Management Journal, 33 (2), 233–258.
Guardian (2018). Twenty years on the Festina affair casts shadow over the Tour de France. Retrieved from https://www.theguardian.com/sport/2018/jul/03/tour-de-france-festina- affair
Guardian (n.d.). The case of Lance Armstrong. Retrieved from https://www.theguardian.com/sport/lance-armstrong
Hilgartner, S., & Bosk, C.L. (1988). The rise and fall of social problems: a public arenas model. American Journal of Sociology, 94, 53-78. doi: 10.1086/228951
Jones, C. (2010). Doping in Cycling: Realism, Antirealism and Ethical Deliberation. Journal of the Philosophy of Sport, 37(1), 88-101. doi: 10.1080/00948705.2010.9714768
37 corporate communications. Corporate Communications: An International Journal, 15 (1), 49-67. doi: 10.1108/13563281011016831
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. doi: 10.1108/13563281011068159
Pfister, G. & Gems, G. (2015). Fairy tales? Marion Jones, C.J. Hunter and the framing of doping in American newspapers. Sport in Society, 18 (2), 136 – 154. doi:
10.1080/17430437.2013.854443
Reese, S.D., Gandy Jr., O.H., & Grant, A.E. (2001). Framing Public Life: Perspective
on media and our understanding of the social world. Mahwah, NJ: Lawrence
Erlbaum Associates
Roloff, J. (2008). Learning from multi-stakeholder networks: issue-focussed stakeholder management. Journal of Business Ethics, 82, 233-250. doi: 10.1007/s10551-007-9573-3
Schafraad, P. (2009). Controversial outsiders : a cross-national study of media attention to
the far-right 1986-2004 (Doctoral dissertation).
Schattschneider, E.E. (1960). The Semisovereign People: A Realist’s View of Democracy in
America, Holt, Rinehart and Winston, New York, NY.
Snow, D.A., Vliegenthart, R., & Corrigall-Brown, C. (2007). Framing the French riots: A comparative study of frame variation. Social Forces, 86(2), 385-415. Touretappe.nl. (2019). Wielerkalender. Retrieved from
https://www.touretappe.nl/uci/wielerkalender-2019/
UCI. (n.d.a). About road cycling. Retrieved from https://www.uci.org/road/about-road-cycling UCI. (n.d.b). Inside the UCI. Retrieved from https://www.uci.org/inside-uci
38 in. Journal of Communication, 57, 60-78. doi: 10.1111/j.1460-2466.2006.00329.x Van Gorp, B. (2010). Strategies to Take Subjectivity Out of Framing Analysis. In P.
D’Angelo & J.A. Kuypers (Eds.), Doing news framing analysis: empirical and
theoretical perspectives (pp. 84-109). New York, NY: Routledge.
Vos, M., Schoemaker, H., & Luoma-aho, V. (2014). Setting the agenda for research on issue arenas. Corporate Communications: An International Journal, 19 (2), 200-215. doi: 10.1108/CCIJ-08-2012-0055
WADA. (n.d.). Who we are. Retrieved from https://www.wada-ama.org/en/who-we-are World Anti-Doping Agency. (2018). World Anti-Doping Code. Retrieved from
https://www.wada-ama.org/sites/default/files/resources/files/wada_anti-doping_code_2018_english_final.pdf
Wu, X. (2007). Stakeholder identifying and positioning (SIP) models: from Google’s
operation in China to a general case-analysis framework. Public Relations Review, 33, 415-425. doi: 10.1016/j.pubrev.2007.08.016
Zhang, X. (2016). Measuring Media Reputation: A Test of the Construct Validity and
Predictive Power of Seven Measures. Journalism & Mass Communication Quarterly,
39
Appendix A
41
A2. Codebook
Online newspaper research project on the doping issue in the cycling sport
In the Frame Matrix examples can be found for when a certain frame is present.
Selection Criteria
- Online news outlet: For this study the Guardian was chosen as the source from which the sample was taken. The choice for this outlet was mainly based on the Guardian’s reputation towards the reporting of the cycling sport overtime. All the articles taken from this newspaper were downloaded from the database LexisNexis.
- Search terms: Doping AND Cycling
- Time frame: The time period used in this content analysis was from 1998 until 2018. From the major event of the Festina team in 1998 until the doubt about Chris
Froome’s use of a certain substance. Within this 20 year period a dense period for each year was chosen: 1st of June to the 31st of August. All articles from these periods were included in the sample. Adding to that were the articles from periods with major events: September 2007, May 2010, September 2010, May 2011, October 2012, January 2013, March 2015, December 2017 and October 2018.
42 - Further criteria: The chosen articles should talk about doping and cycling in at least
one paragraph.
Concepts
- Frame Doping wins the race: Doping is necessary to win races because all big winners have used it. Winning is important to athletes and therefore if you want to win you have to use doping.
- Frame Two speed cycling: There is no longer a level playing field in cycling because some cyclists use artificial assistance and others do not. Cheating is wrong and therefore the policy in place should become better.
- Frame Scapegoat: Cyclists are the ones to blame for the doping issue because they are the ones using it. These cyclists should be caught because the cycling sport has to become transparent and honest even if a few mess up.
- Frame Crooked world of professional cycling: Doping is ingrained in the culture of cycling because there is a social acceptance in the peloton. Doping is a part of the sport and of its history so is there even a solution to solve the doping issue?
- Frame Progress: Science and laws currently have solutions to solve the doping issue but there is a lack of faith in the power of these institutions. The sport should morally strive for a future without doping which can be done by creating greater trust in science and laws.
- Media Reputation: Media reputation is the “overall evaluation of a firm presented in the media” (Deephouse, 2000, p. 1091).
- UCI: Union Cycliste Internationale, which is the world governing body of the cycling sport (ICU: International Cycling Union).
