Running on dynamite: framing in the issue arena
of doping in cycling
A quantitative study on framing and intergroup bias in the issue arena
of doping in professional cycling between 1999 and 2019
by Martijn van Dorp
s12874191 Master’s Thesis
Graduate School of Communication, University of Amsterdam
Master’s Programme Communication Science: Corporate Communication Supervisor: Dhr. Dr. P.H.J. Schafraad
ABSTRACT
Professional road cycling has struggled with doping for almost a century. This thesis analyzes how this issue has been framed in the media with the help of the theory of issue arenas and literature on intergroup bias. To answer this question, a systematic content analysis was conducted with a sample of 327 newspapers articles from De Telegraaf and Het Nieuwsblad. Results found that foreign cyclists, the most active actors in the issue arena, were also evaluated the most negative in the process and that issue-related aspects and the course of the debate indeed play an important role in the issue arena. Furthermore, the study nuances previous studies on intergroup bias and valence framing that argue that domestic actors are by definition evaluated positively and foreign actors are evaluated negatively. Lastly, differences were found between the two newspapers, concerning the usage of frames, negative evaluations towards foreign teams and mentioning of domestic cyclists and teams. In the end, this thesis gives recommendations for future research and the field of communications, including studying framing in countries that have significantly been affected by doping and the potential for professional teams to become more active in the issue arena.
INTRODUCTION
In May 2020, former Tour de France winner Lance Armstrong confessed that he had been doping throughout his career. Whilst it was publicly known that the American cyclist had used performance enhancing drugs in the second half of his career, he now revealed that he used doping in the first half of his career as well (Dupuy, 2020). This added another footnote to the history of professional cycling, which has been haunted by doping for almost a century. One of the first published cases of doping can be traced back to 1924, when Tour de France participants Henri and Francis Pélissier confessed to “running on dynamite” to an investigative journalist (Mignon, 2003). In the decades that followed, the sport and its culture became more and more polluted by doping and the desire to push the human limits with the help of doctors and technology. This trend reached its peak in the 1990’s and 2000’s, with multiple large-scale doping scandals, of which the Festina trial and Operación Puerto are the most notorious, involving arrests, major suspensions and sponsors turning their back on the sport. Since those scandals, things have taken a turn for the better, as cyclists started to confess their illegal substance use and anti-doping tests and policies substantially improved.
In these hectic years, the media had a prominent role in supplying the public with news and information on doping in cycling. The way in which information is distributed is crucial, as it can be argued that media have an important responsibility when it comes to covering certain topics and highlighting specific aspects of those topics (McCombs & Shaw, 1972). Frames play a vital role in the distribution of information, as they are vehicles that carry a certain message or interpretation (Entman, 1993). It is therefore not surprising that numerous researchers have applied these theories to events in the history of sports (Mercurio & Filak, 2010; Messner & Solomon, 1993; Sherwood et al., 2020; Spencer & Limperos, 2018). However, there have not been any studies that looked into the frames that were used with regards to doping in professional cycling.
Hence, the following study will analyze the frames that have been used to describe the issue of doping in professional cycling in newspapers. This will be done using issue arena theory, which focuses on describing the dynamic between different actors in a particular issue (Luoma-aho & Vos, 2010). With such an approach, an overview can be created of the frames that those actors use and how frequently they are used overtime. Furthermore, in reaction to literature on potential intergroup bias in newspapers (Bie & Billings, 2013; Li, Stokowski, Dittmore, & Scott, 2016) and suggestions of historic bias towards French cyclists in the Tour de France (Schneider, 2006), the current research will also seek to investigate how domestic and foreign actors are framed in Dutch and Belgian newspapers. Lastly, since the foundation of the World Anti-Doping Agency (WADA) in 1999, Houlihan, Hanstad, Loland and Waddington (2019) conclude that “the fight against doping in sport is much stronger and more effective than in the 1990s” (p. 200). Similarly, Dimeo (2014) confirms that anti-doping tests have improved after the aforesaid doping scandals in cycling, becoming more effective against blood doping. Thus, this study will subsequently analyze if there are changes in the use of frames between 1999 and 2019. In sum, the overall research question is:
“How do Dutch and Belgian newspapers report on the issue of doping in professional cycling between 1999 and 2019 and what is the role of nationality in the framing of these issues?”
First of all, the theoretical framework will expand on the concepts and theories central to this study. Secondly, the methodology behind the study, including the method of analysis and the sample will be elaborated upon. Thirdly, the results of the study will be displayed. These results will consequently be discussed and put into perspective in the discussion. Last but not least, the research and its results will be summarized in the conclusion.
THEORETICAL FRAMEWORK
The following paragraphs will explore the concepts of issue arenas, intergroup bias and framing and subsequently connect them to literature on sports and doping. Lastly, several sub research questions will be presented.
ISSUE ARENAS
Vos, Schoemaker and Luoma-aho (2014) define the issue arenas as arenas that focus on the public debate about a particular issue among various actors in both traditional and virtual media. In particular, Luoma-aho and Vos (2010) suggest that issue arenas lead to a more dynamic stakeholder model, referring to the interaction of stakeholders regarding a certain topic in traditional or virtual media. The participants of interaction in a specific issue arena are the parties that hold a stake in the issue (Van Schendelen, as cited in Vos et al., 2014). Vos et al., (2014) state that an issue arena, above all, refers to the places where exchange of views on the particular topic takes place. Moreover, within the arena, the focus is on the issue itself and how it is portrayed in the public domain. “The arena is dynamic and characterized by a struggle of power between players with their own agendas and strategies, where the organization is only one player among others who vie for public attention” (Ravazzani, & Maier, 2017, p. 3). Vos et al., (2014) identify four different levels that play a role in an issue arena: issue-related aspects, the actors involved, places of interaction and course of the debate.
To illustrate these four levels, the coming paragraphs will serve as application of the issue arena to the world of sports. In the modern era, sports in general have been affected by scandals involving corruption (Numerato, 2009) or doping (Vlad, Hancu, Popescu, & Lungu, 2018) and these issues have become topics in the public debate (Starke & Flemming, 2016; Ncube, 2017).
First of all, Vos et al., (2014) note that issue arenas are different when it comes to their issue-related aspects. These characteristics take the lifecycle of an issue into account (Vos, van der Molen, & Mykkänen, 2017), as the discourse around a new issue can be different than discourse for an older and longer existing issue. This can be illustrated by the issue arena of doping in professional sports. Mazzeo, Altavilla, D'Elia and Raioloa (2018) describe a shift in the strictness of anti-doping policies and regulations. One could argue that, after many cases of doping, the actors have a different stance on how strict an anti-doping policy should be to keep the sport clean. Similarly, Christiansen (2005) notes that a doping scandal can also change an actor’s perception and public stance towards doping. In his analysis of the Festina trial in 1998, he concludes that before the scandal teams and managers were, to a certain extent, supportive of doping structures because that was in line with the culture in the sport. However, after the scandal, which had major repercussions for the cyclists and the sponsors, teams and managers were quick to speak out against all forms of doping. Although there is no research on the issue arena of doping in cycling, one can assume that the issue arena dynamic significantly changed after the scandal occurred in 1998.
Secondly, the involved actors play a key role in the issue arena. These actors, who possess characteristics such as credibility, legitimacy and power, have their own agendas and reasons to participate in the debate (Vos et al., 2014). Although it often depends on the issue, there are several actors involved in the issue arena of sports. Individual athletes or professional teams are often the subject of corruption or doping scandals and are therefore significantly involved in the issue arena. On a broader scale, there are global or national federations that represent their respective sports (FIFA, UCI, FIBA) and authorities that create regulations to maintain the level playing field (IOC, WADA). As regulators and representatives, they tend to be opinionated on the issue and can thus play a role of importance in the issue arena. Furthermore, sponsors, investors and spectators can involve themselves in an issue. Sponsors
and investors finance the athletes and professional teams, making their opinion on the matter crucial for the continuity of the sport. To illustrate, Reeth (2015) states that the majority of professional cycling teams are largely dependent on sponsors and investors, which increases their importance. Similarly, in some sports, the spectator is a vital stream of revenue, which makes their opinions and concerns relevant to the issue arena.
Thirdly, an issue can be discussed in different places of interactions, which can both be physical and virtual. Public debate can take place among actors on conferences or in governmental settings, but also in the virtual space and (social) media (Vos et al., 2014). The place of interaction can influence the number of actors that partake in the debate. Depending on if it is a small group at a conference or a large audience on social media, it impacts the manner in which this topic is communicated to the outside world – one actor voices their opinion in a (sports) newspaper or multiple actors voice their opinion on platforms such as Twitter (Hellsten, Jacobs, & Wonneberger, 2019). A study from Sherwood et al. (2020) illustrates this dynamic, when they observed the reaction to a Pride-themed football game on social media compared to that in the traditional media. Social media gives spectators a platform to express their opinion, whereas officials, athletes and other actors are more prominent in traditional media. In the end, this can create a different overall perception.
Lastly, Vos et al. (2014) note the importance of the course of the debate. As an issue develops, the issue arena could open itself to new actors or could expand itself by approaching a new topic or dimension of the issue. Hobbs (2020) adds that this includes “various factors that can influence possible outcomes or solutions, including communication strategies seeking to redefine “the public interest” within a debate” (p. 3). Actors often try out different strategies in the issue arena, in order to assess which strategy works best (Vos et al., 2017). The focal point of a debate can also change over time. Vos et al., (2017) give the example on risks caused by global warming. As the issue arena “became dominated by the issues of lowering CO2
emissions and use of fossil fuels, (…) other substances that contribute to global warming, such as methane, seem to be forgotten”. Similarly, in the world of sports, one scandal could be overlooked or even forgotten because of a more recent scandal.
Issue arena theory provides a dynamic perspective on the public debate regarding a specific issue, featuring various actors in both traditional and virtual media. Issue arenas have four different levels: issue-related aspects, the actors involved, places of interaction and course of the debate, which can be applied to various issue arenas in the world of sports. This study revolves around the issue arena of doping in professional road cycling in newspapers from 1999 to 2019, which means that the involved actors and the course of the debate are the two focal points of this study.
INTERGROUP BIAS
The origins of in-group and out-group biases can be found in the common in-group identity model, which “asserts that intergroup bias and conflict can be reduced by factors that transform members’ cognitive representation of the memberships from two groups to one group” (Gaertner, Dovidio, Anastasio, Bachman, & Rust, 1993, p.2). In this model it is argued that if somebody is categorized as part of in-group rather than part of an out-group, that person will be evaluated more positively (Hornsey & Hogg, 2000). Hewstone, Rubin and Willis (2002) add that the creation of in-group identities can create intergroup bias, which creates a situation in which an individual favors the in-group over the out-group and its members. To a certain extent, the idea of intergroup bias is present in ethnocentrism, which revolves around in-group-favoring attitudes and discriminatory behaviors based on group boundaries defined by characteristics that indicate common descend (Hammond & Axelrod, 2006). Such characteristics can include language, religion, ethnic national identity or civic identities (Hammond & Axelrod, 2006; Pehrson, Brown & Zagefka, 2011). However, it is important to
note that, in contrast to ethnocentrism, in-group bias does not inherently translate to out-group derogation (Brewer, 2007).
In-group identification is also present in the world of sports. In his study on football in the United Kingdom, Abell (2011) uses the example of football to showcase how supporting a national football team can be an expression of national identity. This is in line with findings from Mael and Ashforth (2001), who state that “spectator sports facilitate a sense of belonging, albeit in a vicarious way in the camaraderie of fellow fans” (p. 207). It is for such reasons that the literature has frequently pointed to the effects of large sport events, such as the Olympic Games or the FIFA World Cup, on national identities and the concept of nation-building (Farquaharson & Marjoribanks, 2003; Gassmann, Haut, & Emrich, 2019; Kersting, 2007; Mutz & Gerke, 2017). For instance, Mutz and Gerke (2017) argue that the experience of collective emotions in the course of mega sports events “foster in-group identification and add affectivity and meaning to groups and group symbols” (p. 618). Especially in regard to these large sport events, the media play a significant role in the process of fostering in-group identification. Billings, Brown and Brown (2013) observed that heavy viewers of the Olympic games reported higher levels of nationalism and patriotism than light viewers. Similarly, Mutz and Gerke (2017) found an increase in national identification among emotionally highly involved spectators after their national team had been victorious in the FIFA World Cup. Next to the effect of the sport itself, broadcast commentators can contribute to in-group identification by displaying patriotic tendencies. For instance, broadcast commentators can use pronouns such as “us” when referring to the in-group and “them” when referring to the out-group (Ličen & Billings, 2013; Scott, Hill, & Zakus, 2012) or maintaining a nation-centric approach to broadcasting (Angelini, Macarthur, Smith, & Billings, 2015; Billings, Brown, & Brown, 2013; Ličen & Billings, 2012).
However, it is important to note that patriotic tendencies are not reserved to intergroup biases and can promote fair positive patriotism (Horky, Baranovskaa, Grimmer, Jakubowska, & Stelzner, 2019; Kersting, 2007). This is in line with the findings from Brewer (2007): in-group bias does not necessarily translate to out-in-group derogation. In conclusion, literature shows that in-group biases are present in sports in the form of patriotic tendencies, by both spectators and the media.
FRAMING
Framing is an act of selection and salience (Entman, 1993) in which highlighting specific aspects of a perceived reality can promote specific interpretations in a text. Interpretations and meanings can be created by the connection of certain textual elements (Van Gorp, 2005), the distinct differences in the use of certain words and the context that provides meaning to these words (Hellsten, Dawson, & Leydesdorff, 2009). Van der Meer, Verhoeven, Beentjes and Vliegenthart (2014) combine these aspects with the element of interaction: “framing is a set of discourses that interact in complex ways within and among domains” (p. 752). Approaching the theory from a constructionist standpoint, Van Gorp (2007) adds that framing is facilitated by the interaction between the textual level, the cognitive level, the extramedial level and the stock of frames that is available in a given culture (p. 64). The aforementioned interaction can also be seen as a process consisting of building, frame-setting and the consequences (De Vreese, 2005). Frame-building refers to the factors that cause the emergence of specific frames, frame-setting describes the relationship between the frame and the prior knowledge of the reader. The consequences of framing can be altering attitudes about an issue on an individual level and processes such as political socialization on a societal level (De Vreese, 2005, p. 52).
fundamentally different, as its core resides in the meaning that is assigned to topics, whereas agenda-setting and priming focus on the topics that are discussed in public. In sum, the current study defines framing as a process of interactions between textual elements and their contexts, highlighting specific interpretations in a text.
Framing theory is frequently applied to the world of sports, as journalists can use frames to make a story more appealing or highlight a specific part in the story (Lewis & Weaver, 2013). Frames can be applied to athletes based on certain characteristics, such as race (Angelini & Billings, 2010a; Angelini & Billings, 2010b; Eagleman, 2011; Ličen, 2013; Mercurio & Filak, 2010), gender (Messner, Duncan, & Jensen, 1993), sexual orientation (Billings, Moscowitz, Rae, & Brown-Devlin, 2015; Sherwood et al., 2020), athletic performance (Knight, Macneill, & Donnelly, 2005) or actions off the field (Messner & Solomon, 1993; Spencer & Limperos, 2018). Research on frames in sports tend to focus on differences between groups, longitudinal changes or observing the framing of one particular occurrence. To illustrate, a study by Mercurio and Filak (2010) found differences in the characteristics that sport journalists attributed to black and white quarterbacks in the NFL Draft. Where black quarterbacks were seen as very athletic but mentally weak, white quarterbacks were described as mentally strong with weak physical attributes. This is a different approach from Sherwood et al. (2020), who captured the frames that were applied for the first Pride Football game in Australia. Lastly, journalists can also apply frames to a sport. For instance, whilst American football used to be framed as masculine and warrior-like, the focus now appears to be on the dangers and health risks of the sport (Anderson & Kian, 2012; Cassilo & Sanderson, 2016).
Framing in sports can also occur based on nationality or ethnic identity. Based on intergroup bias theory, it can be assumed that frames can be applied to domestic and foreign athletes, often favoring the ingroup athletes. Intergroup bias frequently occurs in sport broadcasting, as many national television channels adopt a nation-centric approach (Angelini,
Macarthur, Smith & Billings, 2015; Billings et al., 2013; Ličen & Billings, 2012). Research by Angelini et al., (2015) showcased that Canadian television attributed domestic athletic success to commitment and intelligence, whilst foreign athletic success was attributed to strength. Similarly, Ličen and Billings (2012) studied Slovenian sport broadcasting and concluded that athletes representing domestic teams were portrayed as outgoing, whereas their opponents were described in a more abstract manner, mentioning their athletic characteristics. Frames based on intergroup biases can also appear in print media, as Li et al., (2016) found that Chinese newspapers described Chinese athletes as dominant and superior to foreign athletes, who were labelled as less competitive and inadequate. Bie and Billings (2013) examined the case of Ye Shiwen, a Chinese swimmer, whose performances were framed as “too good to be true” by American newspapers and as outstanding by Chinese newspapers.
As expressed in the previous paragraphs, frames are applied in multiple different circumstances. However, it is possible to differentiate in the type of frame. This study makes use of two types of frames: thematic framing and valence framing (De Vreese, 2005).
THEMATIC FRAMING
Thematic framing places political issues and events in a general context (Iyengar, as cited in De Vreese, Peter & Semetko, 2001; Van Gorp, 2007) and provides trends and information on the issue (Hart, 2010). In the case of this study, thematic frames place the issue of doping in cycling in a broader context by covering the latest news and offering insights from actors in the field (Gross, 2008).
Literature on framing in doping scandals featured four frequently mentioned thematic frames: legality, health, morality and responsibility. First of all, in their research on anabolic steroids in the Spanish press, Calatayud, Cogollos and Zurián (2014) found that stories were often focused around the penalization of substance use, which meant that authorities in the form
of the police or judicial system were present actors. Similarly, Rutecki and Rutecki (2010) observed that the WADA was consistently mentioned when it came to the illegal aspect of doping. Secondly, actors can also approach the doping problem from a health-perspective, focusing on the harmful side of steroids instead of the “pseudo-benefits” that they can generate (Denham, 1999; Pfister & Gems, 2013). Thirdly, the use of doping is can be seen as unethical and immoral (Rutecki & Rutecki, 2010). For instance, Kozman (2016) elaborated that scientists, lawmakers and parents identified the negative impact on young players as the biggest problem of steroids. Comparatively, journalists have a tendency to highlight “idealistic considerations about integrity and justice” when doping or steroids are discussed (Denham, 2004b). Lastly, multiple studies focus on the responsibilities of the doping problem. In their research on responsibility, Starke and Flemming (2016) found that the system, the doping authorities and overarching federations, and the individual athletes were equally attributed with the causes of the doping problem. To a certain extent, this is in line with Laine (2006), who, in her research on Finnish doping scandals, observed that the doped athletes took responsibility for their actions, but also blamed the doping system for being wrong. This study will use five frames developed by Van ‘t Ooster (2019), which are tailored to the problem of doping in cycling and take the frames found in the literature into account. These frames will be elaborated upon in the method section.
Following the literature on thematic framing in the issue arena of doping and the understanding that the involved actors can play a key role, this study will focus on the thematic frames that are used by domestic and foreign cyclists and professional teams. Based on that, the following research question has been formulated:
Sub-question 1: “Are there differences in the use of thematic frames when it comes to domestic and foreign cyclists or cycling teams?”
VALENCE FRAMING
Valence framing focuses on the depiction of an issue as positive or negative (De Vreese & Boomgaarden, 2003; Schuck & De Vreese, 2006). A frame can also be regarded as neutral, which means that it contains elements of “good and bad” (De Vreese & Boomgaarden, 2003, p. 363) or elements of support or opposition (Bizer, Larsen, & Petty, 2010). Following literature on intergroup bias, the current study will delve into the differences in valence framing between domestic and foreign cyclists and professional teams in newspaper articles. Thus, the following research question has been formulated:
Sub-question 2: “Are there differences in valence when it comes to domestic and foreign cyclists or cycling teams?”
The following paragraphs will discuss the frames that are used in the issue arena of doping. As noted in prior paragraphs, an issue arena focuses on the interaction between various actors with different motives and stakes in the public debate. Based on agenda-setting theory and the idea that media can influence the opinions of their audience and political decisions (Calatayud et al., 2014; Quick, 2010), these actors can use frames in their strategies to try to fulfill their goals. For instance, Denham (2004a) emphasizes that a dramatic story on doping can help the topic to be discussed on the agenda and be taken more serious by policymakers.
Despite the growing body of literature, it is difficult to predict how framing in the issue arena of doping in professional road cycling has developed. Firstly, most studies that touch on the issue arena of doping in professional sports are centered around a single scandal or occurrence, whereas professional road cycling has experienced a multitude of scandals in the last decades (Christiansen, 2005; Dimeo, 2014). Thus, it is interesting to see whether frames change over time, as suggested by Entman (2007). Secondly, in line with Vos et al. (2014), issue arenas are dependent on the context and the course of the debate, which means that it is
difficult to make generalizations based on one occurrence in one arena. Based on the contextual nature of issue arenas, the following research questions were formulated:
Sub-question 3: “Which actors are the most active in the issue arena of doping in cycling and which frames do they use?”
Sub-question 4: “Are there differences in the use and users of frames between 1999 and 2019?”
The importance of the place of interaction in the issue arena of doping can be illustrated by a study from Pfister and Gems (2013), who researched the framing of track superstar Marion Jones in newspapers. They found a difference in focus, as American newspapers focused on the arguments between sports federations and the individual athletes, whereas Danish newspapers published articles that delved into the effects of drugs, medical and legal problems and insights from anti-doping regulations and institutions. Differences between media across countries can also be attributed to the media system that they are in. For instance, Hallin and Mancini (2004) argue that while Belgium and the Netherlands have similar media systems at their core, Belgium’s media system also shows similarities to southern European countries such as Italy and Spain in terms of the party-political character of its broadcasting system and the degree of fragmentation of their political party system. Hence it can be argued that the party-political character of the broadcasting system, which in general tend to involve biases against opposing parties, can influence Belgium’s print media. Thus, although there is no literature that explicitly supports differences between Belgian and Dutch print media in terms of the use and users of frames in the issue arena of doping, one can assume that differences between the two countries are possible. For that reason, the following study seeks to investigate their use of thematic frames and valence frames when it comes to professional road cyclists and cycling teams. Hence, the following research question has been formulated:
Sub-question 5: “Are there differences in the use and users of frames between Dutch and Belgian newspapers?”
METHODS
The following sections will elaborate on the methodological approaches that were used in this study. The method of analysis, the sample, codebook and the process of intercoder reliability will be touched upon.
METHOD OF ANALYSIS
To analyze framing in the issue arena of doping in professional cycling, a systematic quantitative content analysis will be conducted. A systematic content analysis strives to analyze data whilst taking into account the specific meanings that are given to it in a systematic manner (Krippendorff, 1989). This method was used by studies of similar nature (Bie & Billings, 2013; Li et al., 2016) and is therefore appropriate for this research.
SAMPLE
In line with other content analyses on doping frames in sports (Bie & Billings, 2013; Li et al., 2016), the current study will analyze articles from two newspapers, namely the Dutch daily De Telegraaf and the Belgian daily Het Nieuwsblad. The newspapers were chosen because of their similar journalistic styles, reputation and devotion to sport. Articles from Het Nieuwsblad were retrieved from the GoPress database, the articles from De Telegraaf were retrieved from the LexisUni database.
The final sample contains 327 articles from 1999 to 2019, almost evenly divided over De Telegraaf (N = 166) and Het Nieuwsblad (N = 161). Within the abovementioned time period, all articles from the 1st of May to the 31st of August were included in the sample.
Exploratory research showed that this time period generated the most articles and covers the three largest cycling races in the world: Il Giro d’Italia, Le Tour de France and La Vuelta a España (El Helou et al., 2010). Additionally, important events that did not fall into that time period were included. An overview of these events can be found in Appendix 1.
Further inclusion criteria involve mentioning keywords and terms related to “doping” and “cycling” twice, in order to avoid brief mentions of cycling in a global article about doping or brief mentions of doping in a cycling article, and a minimum article length of 150 words.
VARIABLES
For this study, an adjusted version of the codebook developed by Van ‘t Ooster (2019) will be used. Based on open coding, 5 frames were created: Doping Wins the Race, Two Speed Cycling, Scapegoat, Crooked World of Professional Cycling and Progress. Detailed descriptions of these frames can be found in Table A1. Using the method from Semetko and Valkenburg (2000), four questions were assigned to each frame in order to identify if the frame is present in the article.
Table A1
The frames that were used in the study (Van ‘t Ooster, 2019)
Frame Description
Doping Wins the Race 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 important for athletes. Suggested solution is that if you want to win you have to use doping.
Two Speed Cycling 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.
Scapegoat 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.
Crooked World of Professional Cycling
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 omertà present. Suggested solution here is to see if it is possible to solve something that is so ingrained in the culture.
Progress 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 realized through creating a greater trust in tests and laws.
Consequently, the codebook tests which actors make use of the aforementioned frames. Expanding on the researchconducted by Van ‘t Ooster (2019), the following actors have been identified: Cycling Federations, Doping Authorities, Domestic Cyclists, Foreign Cyclists, Domestic Cycling Teams, Foreign Cycling Teams and Race organizers. Detailed descriptions of these actors can be found in the codebook.
Lastly, in order to measure valence framing in the article, evaluative questions were created, inspired by evaluative questions for reputation by Coombs (2007). In this study, a
cyclist or team can be evaluated as positive, negative or neutral towards the doping issue. The full codebook can be found in Appendix 2.
INTERCODER RELIABILITY
Prior to starting the final coding procedure, intercoder reliability was established by having two coders, one of them being the main researcher, coded a selection that represented 10% of the final sample. This was done after coder training, in which the codebook was carefully evaluated and, where needed, descriptions were refined. Furthermore, the coders coded one article together and discussed their choices. To conduct the training, 35 articles were coded over the span of one week.
Calculating intercoder reliability was done with the help of the Reliability Calculator, an online tool developed by Deen Freelon (n.d.). This tool calculates several reliability statistics, one of them being Krippendorff’s Alpha, which examines the agreement between coders and evaluates the extent to which data can be “trusted to represent something real” (Krippendorff, 2011, p. 1). Whilst both the percentage of agreement (percent agreement) and Krippendorff’s Alpha were calculated, this study will use the percent agreement as the leading reliability measure. This decision was made on the assumption that there was an occurrence of overcorrection, based on the small sample size and the high percentage of agreement between the coders (Schafraad, 2009). As percent agreement is a liberal index, a higher benchmark will be used than for conservative indices such as Krippendorff’s Alpha (Lombard, Snyder-Duch & Bracken, 2002).
Overall, a value above 90% is considered to be reliable, whereas a value smaller than 70% can be considered unreliable. As can be seen in Appendix A, the majority of variables score more than 90% and are thus reliable. Similarly, several values are greater than 80% and are therefore acceptable. Only NB2, which measured valence for foreign cyclists, scored < 70%,
which means that conclusions based on that variable should be interpreted carefully. The full results of the intercoder reliability test can be found in Appendix A.
RESULTS
The following sections will showcase the results of the study and the performed tests per sub-question.
SQ1: Are there differences in the use of frames when it comes to domestic and foreign cyclists or cycling teams?
To gain insight into the differences in the use of frames, a frequency test and a crosstab analysis were conducted. The results of those tests can be seen in Table B1 and Appendix 3. It can be observed that foreign cyclists (M = .08, SD = .271). and domestic teams most frequently use the Scapegoat frame (M = .01, SD = .095). Similarly, domestic cyclists make most use of the Two Speed Cycling frame (M = .03, SD = .181) and foreign teams most often used the Progress frame (M = .03, SD = .164).
Table B1
SQ1: Crosstabulation for differences in frame use among domestic and foreign actors Frame Actor Doping Wins the Race Two Speed Cycling Scapegoat Crooked World of Pro Cycling Progress N Domestic Cyclists .03 (.164)z .03 (.181)z .03 (.172)b .01(.110)b .03 (.164)z 43 Foreign Cyclists .04 (.203)a .06 (.228)a .08 (.271)a .05 (.216)a .04 (.196)z 87 Domestic Teams .01 (.078)b .01 (.095)b .01 (.095)b .00 (.055)b .02 (.145)z 16 Foreign Teams .01 (.078)b .02 (.123)b .02 (.145)b .02 (.134)b .03 (.164)z 29
Notes. Values in parentheses represent standard deviations Adjustment for multiple comparisons: Bonferroni
a, b Column values with different subscripts were significantly different from each other at p < .05. Z No significant difference with other actors in the column.
Layout adapted from Semetko & Valkenburg (2000).
The differences in frame use between the groups were calculated with a One-Way Analysis of Variance (ANOVA). The results of the test are summarized in Appendix 4. It can be observed that there were significant differences in frame use among actors for the Doping Wins the Race (F (3, 1307) = 5.227, p = .001), Two Speed Cycling (F (3, 1307) = 5.115, p = .002), Scapegoat (F (3, 1307) = 9.331, p < .001) and Crooked World of Professional Cycling (F (3, 1307) = 6.466, p < .001) frames. No significant differences were found for the Progress (F (3, 1307) = .686, p = .561) frame.
For further analysis into the differences in frame use between groups, a post hoc Bonferroni test was conducted. Appendix 5 summarizes the results from this test. For the Doping Wins the Race frame, significant differences were found between foreign cyclists and domestic teams (Mdifference = .037, SD = .011, p = .006) and foreign cyclists and foreign teams
(Mdifference = .037, SD = .011, p = .006). In sum, it can be concluded that foreign cyclists make
more use of the Doping Wins the Race frame than domestic teams and foreign teams.
For Two Speed Cycling frame, significant differences were found between foreign cyclists and domestic teams (Mdifference = .046, SD = .013, p = .002) and foreign cyclists and
foreign teams (Mdifference = .040, SD = .013, p = .013). In sum, it can be concluded that foreign
cyclists make more use of the Two Speed Cycling frame than domestic teams and foreign teams. For the Scapegoat frame, significant differences were found between foreign cyclists and domestic cyclists (Mdifference = .049, SD = .014, p = .004), foreign cyclists and domestic
teams (Mdifference = .070, SD = .014, p < .001) and foreign cyclists and foreign teams (Mdifference
= .058, SD = .014, p < .001). In sum, it can be concluded that foreign cyclists make more use of the Scapegoat frame than domestic cyclists, domestic teams and foreign teams.
For the Crooked World of Professional Cycling frame, significant differences were found between foreign cyclists and domestic cyclists (Mdifference = .037, SD = .011, p = .006),
foreign cyclists and domestic teams (Mdifference = .046, SD = .011, p < .001) and foreign cyclists
and foreign teams (Mdifference = .031, SD = .011, p = .034). In sum, it can be concluded that
foreign cyclists make significantly more use of the Crooked World of Professional Cycling frame than domestic cyclists, domestic teams and foreign teams.
SQ 2: Are there differences in valence when it comes to domestic and foreign cyclists or cycling teams?
Firstly, a descriptive test and a crosstabs analysis were conducted. An actor’s valence was measured on a scale ranging from 1 to 3. If the actor was present in the article, their performance was evaluated and thus coded as either negative (1), neutral (2) or positive (3). The results of these tests are summarized in Table C1 and Appendix 6.
All four actors were mostly evaluated as neutral. However, in comparison to the other actors, foreign cyclists were more often evaluated in a negative way (M = 1.89, SD = .437). Furthermore, on average, domestic cyclists were evaluated more positively than other actors (M = 2.00, SD = .243).
Table C1
SQ2: Differences in valence among domestic and foreign actors Valence Actor Mean N Domestic Cyclists 2.00 (.243)a 136 Foreign Cyclists 1.89 (.437)b 270 Domestic Teams 1.96 (.296)z 90 Foreign Teams 1.94 (.278)z 187
Notes. Values in parentheses represent standard deviations Adjustment for multiple comparisons: Bonferroni
a, b Column values with these subscripts are significantly different from each other at p < .05. Z No significant difference with other actors in the column.
Layout adapted from Semetko & Valkenburg (2000).
The differences between the groups was measured with a One-Way Analysis of Variance (ANOVA). Appendix 7 summarizes the results of the tests. These results suggest that there are significant differences between the groups (F (3, 682) = 3.566, p = .014). Using a Bonferroni Post Hoc test, differences between the groups themselves could be analyzed. The results of the test are displayed in Appendix 8. There was a statistically significant difference between domestic cyclists and foreign cyclists (Mdifference = .11, SD = .036, p = .010). There
were no significant differences between the other actors.
SQ 3: Which actors are the most active in the issue arena of doping in cycling and which frames do they use?
A crosstab analysis was conducted to show how often each actor used each frame. Table C1 and Appendix 9 give an overview of the frames used by actors. It can be observed that the Doping Wins the Race, Two Speed Cycling, Scapegoat and Crooked World of Professional Cycling frames are most often used by foreign cyclists (M = .04, SD = .203; M = .06, SD = .228; M = .08, SD = .271; M = .05, SD = .216). Progress is most often used by cycling federations (M = .05, SD = .222). Furthermore, it can be concluded that cycling federations, doping authorities, domestic teams, foreign teams and race organizers made most use of the Progress frame (M = .05, SD = .222; M = .02 , SD = .155; M = .02, SD = .145; M = .03, SD = .164; M = .02, SD = .145). In contrast, domestic cyclists most often used Two Speed Cycling (M = .03, SD = .181). and foreign cyclists frequently used Scapegoat (M = .08, SD = .271).
In sum, the most active actors in the issue arena are the foreign cyclists, who mostly use the Scapegoat frame. The least active actors are the domestic teams, who make most use of the Progress frame.
Table C1
SQ3: Distribution of frame use across all actors
Actors Frame Cycling Federations Doping Authorities Domestic Cyclists Foreign Cyclists Domestic Teams Foreign Teams Race Organizers N Doping Wins the Race .01 (.078) .00 (.000) .03 (.164) .04 (.203) .01 (.078) .01 (.078) .01 (.078) 31 Two Speed Cycling .02 (.134) .01 (.078) .03 (.181) .06 (.228) .01 (.095) .02 (.123) .02 (.134) 51 Scapegoat .01 (.110) .00 (.055) .03 (.172) .08 (.271) .01 (.095) .02 (.145) .01 (.095) 54 Crooked World of Cycling .01 (.095) .02 (.145) .01 (.110) .05 (.216) .00 (.055) .02 (.134) .01 (.095) 40 Progress .05 (.222) .02 (.155) .03 (.164) .04 (.196) .02 (.145) .03 (.164) .02 (.145) 70 N 32 18 43 87 16 29 21
Note. Values represent means.
Values in parentheses represent standard deviations.
SQ4: Are there differences in the use and users of frames between 1999 and 2019?
Table D1 and Appendix 10 illustrate the use of each frame per time period. Progress was used the most in each time period, Doping Wins the Race was relatively used most often between 2013 and 2019 (22,0%) and that the Scapegoat frame was relatively most often used between 1999 and 2005 (29,5%). The table also shows that the Two Speed Cycling (29,9%), Crooked World of Professional Cycling (23,9%) and Progress (36,4%) frames were relatively mostly used between 2006 and 2012. Furthermore, the table shows that the Doping Wins the Race frame is increasingly used in terms of percentage over time. A similar process occurs for the Crooked World of Professional Cycling and Progress frames.
The effect of time on the five frames was measured with a One-Way Analysis of Variance (ANOVA). There are no significant differences between groups for Doping Wins the Race (F (2, 324) = .803, p = .449), Two Speed Cycling (F (2, 324) = 1.164, p = .314), Scapegoat (F (2, 324) = .866, p = .422), Crooked World of Professional Cycling (F (2, 324) = 1.754, p = .175) and Progress (F (2, 324) = .368, p = .693), meaning that there is no significant effect of time on the frames. The results of this test can be found in Appendix 11. In addition, a Bonferroni Post Hoc test did not yield any statistically significant effects.
To observe the relation between time and the presence of actors in the articles, a chi-square test of independence was conducted for all five frames. For the significant results, Somers’ d was calculated in order to determine the strength of the association. The results of these tests are displayed in Table D2. The chi-square test of independence yielded significant results for Doping Authorities (χ 2 (2) = 16.015, p < .001) and Race Organizers (χ 2 (2) = 9.923,
Table D1
SQ4: Differences in frame presence across time
Time Period Frame Period 1 (1999 – 2005) Period 2 (2006 – 2012) Period 3 (2013 – 2019) Doping Wins the
Race
16,4% 19,6% 22,0%
Two Speed Cycling 29,5% 29,9% 23,2%
Scapegoat 29,5% 23,9% 25,6%
Crooked World of Professional Cycling
14,8% 23,9% 23,2%
Progress 31,1% 36,4% 35,4%
Note. The percentages reflect the frame usage based on the total amount of articles in the distinct period
Doping Authorities (d = .196, p < .001) and a significant, weak and negative association between time and Race Organizers (d = -.149, p =.003). There were no statistically significant results for other actors.
Table D2
SQ4: Chi-Square Test of Independence for actor presence across time Statistics Actor χ2 df p d Cycling Federations 1.700 2 .428 .039 Doping Authorities 16.015** 2 .000 .196** Domestic Cyclists 2.817 2 .245 .015 Foreign Cyclists .138 2 .933 .018 Domestic Teams 3.178 2 .204 -.065 Foreign Teams .048 2 .976 -.010 Race Organizers 9.923* 2 .007 -.149* Note. * p < .05, ** p < .001
To observe the relation between time and the specific actors that used certain frames, five chi-square tests of independence were conducted for all actors across all five frames. For the significant results, a Somers’ d was calculated to measure the strength of the association. The results from these tests can be found in Appendix 12 up to Appendix 16.
For Doping Wins the Race, significant results were found for foreign cyclists (χ 2 (2) = 8.408, p = .015) and domestic teams (χ 2 (2) = 6.012, p = .049). These results were accompanied by a small, negative and insignificant association for foreign cyclists (d = -.024, p =.601) and a
small, positive but insignificant association domestic teams (d = .031, p =.154). Other actors did not have any significant results.
For Two Speed Cycling, a significant result was found for the domestic teams (χ 2 (2) = 9.046, p = .011). Consequently, Somers’ d was calculated and resulted in a small and insignificant association (d = .046, p = .079). Other actors did not have any significant results.
For the Scapegoat frame a significant result was found for the foreign cyclists (χ 2 (2) = 7.504, p = .023). Consequently, Somers’ d was calculated and resulted in a small and insignificant association (d = .008, p = .881). Other actors did not have any significant results.
For the Progress frame significant results were found for cycling federations (χ 2 (2) = 7.846, p = .020) and domestic teams (χ 2 (2) = 9.204, p = .010). These results were accompanied by a small but significant association for cycling federations (d = .098, p = .017) and a small and insignificant association for domestic teams (d = .011, p = .784). Thus, there is a significant relation between time and the cycling federations’ use of the Progress frame. Other actors did not have any significant results.
The tests for the Crooked World of Professional Cycling frame yielded no significant results.
SQ5: Are there differences between Dutch and Belgian newspapers?
Firstly, in order to analyze the differences in frames, a chi-square test of independence was carried out.
Significant differences were found between De Telegraaf and Het Nieuwsblad for the Doping Wins the Race (χ 2 (1) = 5.602, p = .018), Two Speed Cycling (χ 2 (1) = 5.698, p = .017) and Progress frames (χ 2 (1) = 5.781, p = .016). The strength of the associations was measured using Somers’ d, which was significant and moderate for all three frames – Doping Wins the Race (d = .127, p = .017), Two Speed Cycling (d = .131, p = .016) and Progress (d = .133, p =
.015) frames. In sum, Het Nieuwsblad makes significantly more use of the Doping Wins the Race, Two Speed Cycling and Progress frames. For the other frames, no significant differences were found. The results of the chi-square test can be found in Table E1.
Table E1
SQ5: Differences in frame presence across newspapers Presence in newspaper Chi-Square Test Frames De Telegraaf Het Nieuwsblad χ2 df p d F1 14,5% 24,8% 5.602* 1 .018 .127** F2 22,3% 34,2% 5.698* 1 .017 .131** F3 24,1% 26,7% .294 1 .587 .030 F4 18,7% 25,5% 2.195 1 .138 .081 F5 28,9% 41,6% 5.781* 1 .016 .133**
Note. The percentages reflect the presence based on the total amount of articles per newspaper (De Telegraaf, N=166; Het Nieuwsblad, N=161)
* χ2 is significant at p < .05
** Somers’ d is significant at p < .05
To analyze the differences in actor presence, a chi-square test of independence was carried out. Significant differences were found for domestic cyclists (χ 2 (1) = 4.120, p = .042) and domestic teams (χ 2 (1) = 9.833, p = .002). Somers’ d was consequently calculated to determine the strength of the association. For domestic cyclists, a small and significant association was found (d = .112, p =.041). For domestic teams, a moderate and significant association was found (d = .172, p =.001). In sum, domestic cyclists are featured more often in Het Nieuwsblad than in De Telegraaf and domestic teams are present more often in De Telegraaf than in Het Nieuwsblad.
For the other frames, no significant differences were found. The results of the Chi-Square test can be found in Table E2.
Table E2
SQ5: Differences in actor presence across newspapers Presence in newspaper Chi-Square Test Actors De Telegraaf Het Nieuwsblad χ2 df p d Cycling Federations 49,4% 40,4% 2.690 1 .101 -.091 Doping Authorities 25,9% 28,6% .294 1 .588 .588 Domestic Cyclists 35,5% 46,6% 4.120* 1 .042 .112** Foreign Cyclists 86,1% 79,5% 2.540 1 .111 -.085 Domestic Teams 36,1% 20,5% 9.833* 1 .002 -.172** Foreign Teams 62,0% 53,4% 2.497 1 .114 -.087 Race Organizers 14,5% 17,5% .526 1 .468 .038
Note. The percentages reflect the presence based on the total amount of articles per newspaper (De Telegraaf, N = 166; Het Nieuwsblad, N = 161)
* χ2 is significant at p < .05
** Somers’ d is significant at p < .05
To analyze differences in valence across the actors, a chi-square test of independence was carried out. Significant results were analyzed with Somers’ d to calculate the strength of the association. Table E3 displays the valence across the different actors in the newspapers in percentages. There was a significant result for foreign teams (χ 2 (2) = 6.045, p = .049), there is a significant difference between newspapers when it comes to the valence attributed to foreign teams. The strength of the association was moderate, negative and significant (d = -.136, p = .035). In sum, articles in Het Nieuwsblad are more negative towards foreign teams than articles in De Telegraaf. For the other actors, no significant differences or associations were found. The results of the Chi-Square test can be found in Appendix 17.
Table E3
SQ5: Distribution of valence across newspapers
Valence
Actor Negative Neutral Positive
De Telegraaf Domestic Cyclists 3.3% 93,4% 3,3% Foreign Cyclists 15,2% 80,7% 4,1% Domestic Teams 8,5% 88,1% 3,4% Foreign Teams a b 2,9% 96,2% 1,0% Het Nieuwsblad Domestic Cyclists 2,7% 94,7% 2,7% Foreign Cyclists 16,8% 78,4% 4,8% Domestic Teams 3,2% 96,8% 0,0% Foreign Teams a b 12,0% 86,7% 1,2%
Note. The percentages are based on the number of articles in which the actor is present.
a Effect is significant at p < .05
DISCUSSION
The following paragraphs will discuss the results from the study and provide answers to the research questions. Furthermore, potential implications and limitations will be highlighted and recommendations for future research will be provided. Firstly, the results of the analyses will be put into perspective to see if they are in line with the literature.
SQ1: Are there differences in the use of frames when it comes to domestic and foreign cyclists or cycling teams?
There are differences in the use of frames when it comes domestic and foreign cyclists or cycling teams. Foreign cyclists made significantly more use of the Doping Wins the Race and Two Speed Cycling frames than domestic and foreign teams. Furthermore, foreign cyclists more often used the Scapegoat and Crooked World of Professional Cycling frame in comparison to the other actors.
As stated by Calatayud et al., (2014), actors can use frames for their own benefit or to push for a change. Hence, it is not difficult to understand why foreign cyclists use the Doping Wins the Race and Two Speed Cycling frames more often than teams do, as those frames discuss how doping affects the competitiveness and fairness of the sport. This point of view is clearly illustrated in one of the articles, as a foreign cyclist refers to doping users as thieves and that their cheating is the reason that clean cyclists do not get paid as much and do not have the accolades they deserve (“Politieke samenzwering”, 2001).
Similarly, the avid use of the Scapegoat and Crooked World of Professional Cycling frame by foreign cyclists is not controversial in any regard. By using these frames, a cyclist can re-emphasize the fact that they are clean or superior athletes. For example, one cyclist argued that his cheating colleagues did not have “as much passion as I [the cyclist] did” (“Laurens ten Dam: Ik voel me bedonderd”, 2008). Similarly, a doped cyclist could have blamed the toxic
doping culture of cycling for their own doping use, which would correspond with the findings from Laine (2006), who found that whilst doped athletes took responsibility, they also blamed the doping system.
Thus, this study contributes to literature on issue arenas in sports by showing that foreign cyclists make more use of frames than the other actors, which implies that foreign cyclists tend to be more outspoken about doping in cycling, likely because they want to prove that they are clean athletes.
SQ2: Are there differences in valence when it comes to domestic and foreign cyclists or cycling teams?
Foreign cyclists were more often criticized for their actions or standpoint regarding the doping issue or were said to have had a negative impact on the reputation of their team. In similar fashion, reports on domestic cyclists were more neutral and provided a balanced view on the doping issue. To a certain extent, these findings are in line with research on framing by Li et al., (2016), who, amongst other things, found that domestic athletes were evaluated as dominant and foreign athletes were framed as inadequate and a study from Bie and Billings (2013) who found that newspapers had a tendency to be critical about the performances of foreign athletes. However, the findings deviate from the literature when it comes to domestic cyclists, who are not viewed more positively than foreign cyclists. This does not correspond with literature on in-group bias, which argues for the tendency to evaluate the in-group more positively (Hornsey & Hogg, 2000), not derogating the out-group (Brewer, 2007).
Thus, the findings in this study adds nuance to previous ideas that newspapers are by definition positive towards domestic athletes or teams and negative towards foreign athletes or teams (Bie & Billings; 2013, Li et al., 2016). Furthermore, it challenges the findings from Brewer (2007), who argues that in-group bias does not imply out-group derogation. However,
it must be noted that because of a low score during the intercoder reliability trajectory, findings regarding the valence of foreign cyclists should be interpreted carefully.
SQ3: Which actors are the most active in the issue arena of doping in cycling and which frames do they use?
Following the analyses, it was concluded that foreign cyclists are the most active actors in the issue arena and that they are the most frequent users of most frames. This can be explained by the fact that foreign cyclists are simply frequently involved in doping scandals, which is a reason for a journalist to give them a platform to voice their opinion on the matter. The Progress frame is an exception to this trend, as this frame is mostly used by cycling federations. Cycling Federations making frequent use of the Progress frame is not surprising either, as that frame highlights the status and quality of policies and anti-doping regulations, which are partially determined by the cycling federations. It is interesting to note that, apart from foreign cyclists, all actors most often used the Progress frame. This aligns with findings from Rutecki and Rutecki (2010), who observed that the WADA, an anti-doping authority, was frequently mentioned when the illegal aspects of doping were discussed in an article.
It is possible to expand on the previous finding that foreign cyclists are more active in the issue arena than other actors because they want to prove they are clean athletes; this study shows that foreign cyclists are most often given a platform by the journalist to comment on the issue of doping in cycling. The manner in which they comment on the issue is something that has not been written about in previous studies. Following up on the finding that newspapers are the most negative towards foreign cyclists, it can be argued that foreign cyclists tend to use frames against other foreign cyclists, in order to emphasize that they are clean athletes themselves whilst condoning the behavior of other foreign cyclists. This re-confirms that actors use framing strategically and for their own benefit (Calatayud et al., 2014).
It is surprising that cycling teams, and domestic teams in particular, were not as active as other actors in the issue arena. In his research, Christiansen (2005) found that managers and teams frequently spoke out against doping after the Festina trial in 1998, which is when sponsors started to face repercussions. Following the theory of issue arenas, it could be the case that cycling teams were not active in the traditional media but rather active in other places of interaction, such as TV or social media.
SQ4: Are there differences in the use and users of frames between 1999 and 2019?
The analysis compared three time periods: 1999 to 2005 (Period 1), 2006 to 2012 (Period 2) and 2013 to 2019 (Period 3). During these periods it was found that race organizers became less present over time, whereas doping authorities were mentioned more often. This corresponds with the findings from Dimeo (2014) and Houlihan et al. (2019), who observed that after 1999 anti-doping policies became stricter and the quality of anti-doping tests improved. This increase in prominence and involvement explain the increase in presence in newspaper articles. The decrease in presence of race organizers could be related to the increased importance of doping authorities. Prior to the shift towards anti-doping policies, race organizers played a key role in expelling doped cyclists and were therefore relevant for journalists to include in their articles, which indicates why their presence decreased over the years.
When it comes to the usage of certain frames, it is interesting to observe that domestic teams increased their use of Doping Wins the Race, Two Speed Cycling and Progress over time. This can be attributed to the fact that domestic teams started to appear more often in articles with time but could also be because of retrospection. As the period between 2013 and 2019 is defined as the start of “clean” cycling (Sefiha, 2016), one can assume that a significant number of articles look back on previous doping cases and evaluate them using the three abovementioned frames.
Furthermore, foreign cyclists used the Doping Wins the Race and Scapegoat frames significantly less between 2006 and 2012. It is difficult to point to an exact reason for this finding, but it could be argued that due to the course of the debate in the issue arena, foreign cyclists were not eager to incriminate their colleagues. This is also historically accurate, as the omertà, a code of silence, was still ingrained into cycling culture during that point in time (Dimeo, 2014).
Lastly, cycling federations made significantly more use of the Progress frame between 2013 and 2019. This could be related to the findings from the aforementioned analysis by Dimeo (2014), as the policies created by cycling federations became stricter and for that reason cycling federations spoke out their trust in the policies.
These findings illustrate the importance of issue-related aspects and the course of the debate in an issue arena (Vos et al., 2014). The changes in actor presence and frame use that were seen in the analysis aligned with the historical overview written by Dimeo (2014) on multiple occasions, confirming that the theory of issue arenas provides a valuable perspective to framing studies and that frames can change over the course of time (Entman, 2007).
SQ5: Are there differences between Dutch and Belgian newspapers?
Several differences between the Dutch and Belgian newspapers were found. The Doping Wins the Race, Two Speed Cycling and Progress frames were significantly more present in Het Nieuwsblad, which gives insight into the type of news that is valued by their target group. In succession of Pfister and Gems (2013), who stated that US newspapers paid more attention to conflicts and Danish newspapers focused on providing information on drugs from authorities, it can be concluded that Belgian newspapers focus on anti-doping tests and policies – focal points of the Progress frame – and the effect of doping on the competitiveness and fairness of the sport, which are at the core of the Doping Wins the Race and Two Speed Cycling frames.
Domestic cyclists are significantly more present in Het Nieuwsblad and domestic teams are significantly more present in De Telegraaf. In addition, Het Nieuwsblad is more negative towards foreign teams. This finding is in line with earlier indications on the party-political character of Belgium’s media system (Hallin & Mancini, 2004), as there is a negative bias towards foreign teams and an ethnocentric focus on domestic cyclists. The finding also adds to the presence of the Doping Wins the Race and Two Speed Cycling frames, as it could be argued that the domestic cyclists are using those frames to criticize the doping use of foreign teams.
This study provides an interesting outlook on Belgian media, which, similar to Danish newspapers, emphasize the legal and medical aspects of doping in cycling, but also frequently comment on the negative impact of doping on the competitiveness and fairness of the sport. Furthermore, they display ethnocentric tendencies by focusing on Belgian cyclists and evaluating foreign teams in a negative way.
RECOMMENDATIONS
In sum, there are theoretical recommendations for future studies and the field of communications based on this study. Firstly, future research on doping in issue arenas should focus on the discourse in newspapers based in countries that were significantly impacted by doping scandals. The Netherlands and Belgium, despite having different athletes involved in doping scandals in professional cycling, were not impacted as much as nations such as Spain or Germany. The former had many of its professional cyclists and other athletes involved in Operación Puerto, whereas the latter banned professional road cycling from television until the sport was clean again (Solberg, Hanstad & Thøring, 2010). Secondly, future studies should continue to explore valence framing in the context of sports. Exposing negative biases towards foreign athletes or ethnocentric approaches towards domestic athletes and teams as seen in this study could spark discussions on whether or not these features are appropriate for modern sport
reporting. Thirdly, research on issue arenas should consider the role of the journalist as an actor. Whilst agenda-setting theory implies that the journalist provides a platform for other actors to voice their opinions, journalists themselves can transcend that role by voicing their opinions through columns or personal conclusions. Gaining more insight on the journalist as an opinion maker and user of frames could provide compelling results in media comparisons. Lastly, although the theory of issue arenas provides an appropriate angle to this study, this study was unable to provide a complete overview because it was restricted to traditional media. As there is a possibility that actors that are not active in newspapers are active on other platforms, future research on doping in cycling should incorporate different platforms (TV, social media) to find differences or similarities between actors on platforms.
Similarly, this study also presents a practical implication for the field of communications. It was found that professional cycling teams were not as active as other actors in the issue arena. Using frames and being active in the issue arena could be valuable for cycling teams, because the issue arena contains stakeholders that are important to them and their sponsors, which are vital for the existence of cycling teams (Reeth, 2015). Thus, it is recommended for cycling teams to be more active in the issue arena to emphasize their stance against doping.
LIMITATIONS
The following paragraph will present the limitations found in this study. First of all, the generalizability of the study is limited by the degree of specificity of the subject. Whilst the analyses did indicate that intergroup bias could have occurred, the notion that Belgian newspapers partake in intergroup bias in general cannot be made. Whilst the results of the study are of added value to research on the issue arena of doping (in professional cycling), it cannot be applied to other areas of framing. Additionally, whereas the choice for two newspapers with
similar reporting styles provides ground for a comparison between Dutch and Belgian media, it also restricts itself in only giving insight on framing in popular media. Articles from quality newspapers could have been included in the sample, in order to broaden the findings.
In terms of the research design itself, the specificity of the codebook and the lack of time and room for practicing and applying it to articles caused one item to be scored insufficiently. In the future, the codebook and the matching questions should be re-evaluated and tested again to optimize results.
