|UFDC Home||myUFDC Home | Help|
1 PERCEIVED CREDIBILITY OF SPORTS ARTICLES AND ATTITUDES TOWARD SPORTS SOURCES AND MEDIA: THE ROLE OF SPORT FAN IDENTIFICATION By SEAN SADRI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PA RTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 Sean Sadri
3 To my family and friends for the constant support throughout my life
4 ACKNOWLEDGMENTS I would like to tha nk my advisor Dr. Cory Armstrong for providing me three amazing years of guidance as a doctoral student. I would not be where I am today without her support. I also thank my committee members, Dr. James Babanikos Dr. Wayne Wanta, and Dr. Yong Jae Ko, fo r their mentoring and insight throughout the dissertation process. I would like to thank all my friends across the country for always keeping me positive during the good and bad times Finally, I thank my parents for the years of encouragement, and for al ways motivating me to study by making me realize early on in life that I would never be good enough at sports to make a living doing it.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ 4 LIST OF TABLES ................................ ................................ ................................ ........... 7 ABSTRACT ................................ ................................ ................................ .................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ... 11 Sports Media an d Credibility ................................ ................................ .................. 15 Theoretical Framework ................................ ................................ .......................... 18 2 LITERATURE REVIEW ................................ ................................ ......................... 23 Credibility ................................ ................................ ................................ ............... 23 Perceptions of Online Credibility ................................ ................................ ..... 26 Factors that Influence Online Credibility ................................ .......................... 28 Blog Credibility ................................ ................................ ................................ 30 Social Identity Theory ................................ ................................ ............................ 33 Group Differentiation and Categorization ................................ ........................ 36 Social Identity and the Media ................................ ................................ .......... 38 Social Identity in Sport Spectatorship ................................ .............................. 42 Developing Soci al Identity through Online Social Networks ............................. 44 Fan Identification and Fan Involvement ................................ ................................ 46 Sports Fans and Social Identity ................................ ................................ ....... 47 Effects of Team Identification on Self Esteem ................................ ................. 49 Team Identification in Collegiate Sports ................................ .......................... 52 Sport Media Coverage and Fan Information Seeking Behavior ....................... 53 Fan Involvement ................................ ................................ .............................. 56 The Social Identity Model of Deindividuation Effects (SIDE) Model ....................... 58 3 RESEARCH QUESTIONS AND HYPOTHESES ................................ ................... 63 4 METHODS ................................ ................................ ................................ ............ 70 5 RESULTS ................................ ................................ ................................ .............. 81 Sample Description ................................ ................................ ............................... 81 Research Questions and Hypotheses ................................ ................................ .... 85 Research Question 1 ................................ ................................ ...................... 85 Research Question 2 ................................ ................................ ...................... 86 Research Question 3 ................................ ................................ ...................... 86 Research Question 4 ................................ ................................ ...................... 87
6 Hypothesis 1 ................................ ................................ ................................ ... 87 Hypothesis 2 ................................ ................................ ................................ ... 88 Hypothesis 3 ................................ ................................ ................................ ... 90 Hypothesis 4 ................................ ................................ ................................ ... 92 Hypothesis 5 ................................ ................................ ................................ ... 93 Hypothesis 6 ................................ ................................ ................................ ... 93 Hypothesis 7 ................................ ................................ ................................ ... 94 Hypothesis 8 ................................ ................................ ................................ ... 96 Gender ................................ ................................ ................................ .................. 98 6 DISCUSSION ................................ ................................ ................................ ...... 100 Findings ................................ ................................ ................................ ............... 101 Future Research ................................ ................................ ................................ .. 123 Limitations ................................ ................................ ................................ ........... 124 Conclusions ................................ ................................ ................................ ......... 126 APPENDIX A ORIGINAL ASSOCIATED PRESS ARTICLE ................................ ....................... 130 B REVISED ARTICLE FOR PARTICIPANTS ................................ .......................... 133 C EXPERIMENT QUESTIONNAIRE ................................ ................................ ....... 136 LIST OF REFERENCES ................................ ................................ ............................ 146 BIOGRAPHICAL SKETCH ................................ ................................ ......................... 160
7 LIST OF TABLES Table page 4 1 3 x 4 Experimental Design ................................ ................................ ................. 71 5 1 Profile of Sample Population ................................ ................................ ............. 81 5 2 Offline and Onli ne Sources for Information about Sports ................................ ... 82 5 3 Analysis of Variance for Fans with High and Low Identification on the Fan Identification Scale ................................ ................................ ............................ 84 5 4 Analysis of Variance of Sports Behavioral Involvement for Fans with High and Low Identification ................................ ................................ ........................ 84 5 5 Univariate Analysis of Variance of Credibility Scores for the Online Medium and Wire Service ................................ ................................ ............................... 85 5 6 Univariate Analysis of Variance of Credibility Scores of Medium for Highly Identified Fans ................................ ................................ ................................ ... 86 5 7 Univariate Analysis of Variance of Credibility Scores of Medium for Fans with Low Identification ................................ ................................ .............................. 87 5 8 Analysis of Variance of Credibility Scores for Online Sources ........................... 87 5 9 Univariate Analysis of Variance for Online Source Credibility Based on Identification Level ................................ ................................ ............................. 88 5 10 Univariate Analysis of Variance of Website Ev aluation Score Based on Online Source and Identification Level ................................ .............................. 89 5 11 Means of Credibility Scores for Each Stimuli Group Separated by Identification Level ................................ ................................ ............................. 90 5 12 Univariate Analysis of Variance for Credibility Score Based on Identification Level and Stimuli Group ................................ ................................ .................... 91 5 13 Analysis of Variance of University Identificat ion Score Based on Identification Level ................................ ................................ ................................ ................. 92 5 14 Univariate Analysis of Variance of Credibility Score for Highly Identified Fans Based on User Comment Tone ................................ ................................ ......... 93 5 15 Univariate Analysis of Variance of Credibility Score for Fans with Low Identification Based on User Comment Tone ................................ .................... 94
8 5 16 Univariate Analyses of Variance of U ser Identification Score Based on User Comment Tone ................................ ................................ ................................ 95 5 17 Univariate Analysis of Variance of User Identification Score Based on Fan Identification Level and Stimuli Received ................................ .......................... 96 5 18 Univariate Analysis of Variance of Source Evaluation Score for Fans with Low and High Identification Based on Tone of User Comments ........................ 97 5 19 Univariate Analysis of Variance of Source Evaluation Score Based on Fan Identification Level and Stimuli Received ................................ .......................... 98 5 20 Univariate Analyses of Variance for the Participant Scales Based o n Gender ... 99
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PERCEIVED CREDIBILITY OF SPORTS ARTICLES AND ATTITUDES TOWARD SPORTS SOURCES AND MEDIA: THE ROLE OF SPORT FAN IDENTIFICATION By Sean Sadri December 2012 Chair: Cory L. Armstrong Major: Mass Communication The 24 hour nature of the Internet has altered the news landscape and th e pressure to break stories first has raised questions about the veracity of reported information. The present study examine d whether sports journalism has experienced a diminished level of credibility, looking specifically at how article source, medium, fan identification, and user comment tone can all impact the credib ility of the sports article s a ne ws or sports source. An online experiment was distributed to participants ( N = 376) who were randomly assigne d a sports article in one of twelve stimuli groups The article source was indicated to have appeared on a mainstream sports website (ESPN.com), a sports blog (alligatorarmy.com), a social networ king site ( Facebook ), or a wire service ( Associated Press ). Participants also re ceived the stimuli with either positive user comments, negative user comments, or without comments as well as a pre and post test questionnaire Analysis revealed that fan identification level was an important factor in credibility ratings. T here was min imal difference in cred ibility scores between the wire service and
10 online medium and both were only seen as slightly credible. For the two identification groups highly i dentified fans found the article to be significantly more credible than fans with low identification as a whole and in 11 out of 12 stimuli group s Highly i dentified fans also rated all three websites significantly highe r than low identification fans The disparity in ratings was evident as highly identified fan s rated all the online sou rces favorably and low identification fans rated all the websites slightly negatively except for Facebook Both identification groups rated the blogs as the least favorable online source, and Facebook was the only website that received a positiv e score from each group Additionally, scores on the user identification scale were significantly higher for the positive comments than for the negative comments for both highly identified fan s and fans with low identification User comments did not, however, a ffect credibility ratings or evaluation scores of the sport s sources. The implications of fan identification level on the discrepancies in ratings of perceived credibility and attitude s toward a sports source were explored.
11 CHAPTER 1 INTRODUCTION Few spo rts rivalries compare to a University of Florida and Florida State University f ootball game. Every year, coaches and players spend all season preparing for this rivalry game that they desperately want to win For the players, it means glory on the gridir on and a chance to embarrass their most hated rival. For die hard fans, this rivalry may mean even more. Crowds pack the campus parking lots hours before the ga me to tailgate with friends and show support for their team UF and FSU symbols can be seen o n cars and clothing throughout campus. The attitude between opposing fans can get contentious and even belligerent at times. This is all before the game even starts. From the opening kickoff to the final whistle, the crowd is raucous and loud as fans fr om both teams are screaming and performing cheers. When the game is over, the fans from the winning team go home ecstatic with great memories and a year of bragging rights. For a sports fan, this is completely ordinary and there are millions of fans aro und the world who feel this same passion for sport. Sport has become a staple of our society as millions of fans are influenced by sport every day, and more individuals are becoming interested and active in sports (Wann et al., 2001). Sports have even com e to carry cultural meanings that reflect the cultural ideologies of our wider society (Beyer & Hannah, 2000). The inherent competition within sport can also represent the ultimate struggle of good and evil, where sport and exercise can be common elements of civic engagement, and, ultimately, work to revive civil society (Harris, 1998). The impetus is on the fan, however, to get the most out of the sporting experience through live events and mass media channels.
12 many fans, the connection to a sports team goes beyond the games themselves and successes and failures that it becomes a part of their social identity (Hu & Tang, 2010). social networks will result in an en in a more positive self concept (Hogg & Abrams, 1990). These memberships make up self esteem. The individuals wi thin the group typically develop a system of role relationships, social norms, and values, which regulate their opinions and actions (Turner, 1982). Fan identification, or the personal commitment or emotional involvement a person has with a sports organi zation, has been shown to predict fan consumption behavior through attending live sporting events and sport media usage (Milne & McDonald, 1999). Highly identified fan s are more likely to watch games in person or through media (Laverie & Arnett, 2000), sp end more on team merchandise, pay more for tickets, and stay loyal to a poorly performing team (Fisher & Wakefield, 1998). Additionally, fan identification can positively affect self esteem and eustress (positive levels of stress) (Hu & Tang, 2010). Acco rding to Wann (2006), when someone resides in the environment where the team is found (e.g. a fan of a college football team who lives on campus at the university), that person may gain enduring social connections. However, fans following a more distant t eam do not have the opportunity to make these enduring connections.
13 The connections established through sporting events and interpersonal fan relationships are seemingly amplified through sport media. Wann and Branscombe (1992) found that the emotional re sponses to a sports article were different based on degree of identification with the sports team. Highly identified fans experienced the most positive mood state from a sports article that described a victory for the ingroup, whose author was an admitted loyal fan of the same team. The most negative mood state was evoked when the team lost the competition and the article author was a identification were not significa ntly influenced by the game outcome, group membership When represented through the mass media, the most prevalent elements of sports are action, records, elite performances, aggressions, heroic actions drama, emotions, and sports stars (Schantz & Gilbert, 2001). Newspaper discourses have also been major sporting events (Lee, 2005, p. 194). Lee (2005) describes a Europe an soccer match that was transformed by the media from an insignificant preseason match into a hugely successful spectacle to fans. According to the author, the media can both create an artificial atmosphe re of hysteria for its audience and create a space for constructive discourse in society. The media articulated the grand meaning of the event by placing it in both local and global contexts. Because of the media, sports fans now have a special relationship with the sport spectacle. Sport fandom has al so become an important aspect of social relations, and social scientists have shown interest in understanding fan involvement online (Wann et al.,
14 2003). Online media coverage provides fans with sports experiences that did not exist in the 20th Century. New media has expanded the scope of sport coverage exponentially as the Internet has become a primary source of information for sport consumers (Delpy & Bosetti, 1998). Because of smart phones and iPads, the Internet has become an all day, everyday experi ence for many people as online and offline lives have become interchangeable. Media usage patterns are shifting as more people begin to follow sports online. In a study done before the 2010 Winter Olympics, 18% of sport consumers planned on following the Games online, while 30% planned on following the still using print media, this growing online trend prompted Olympic sponsors, such as Visa and Coke, to put as much as 40 % of their Olympic marketing budget into digital Because of advancements in technology, modern sport may now reflect the general values and understandings of nationalism for cultures around the world (Lee, 2009). Examining t he online discussion of a prominent 2004 Chinese vs. Japan soccer 2). However, this vision was contested by other discussants with opposing viewpoints. Essentially, the online realm of sport provides those with an opinion that goes beyond the games an outlet to express their views, and others a right to refute those vi ews. The global popularity of sport is increasing rapidly as evidenced by the growth of online newspapers and sports dedicated websites, which is now a multi billion dollar
15 industry (Raney & Bryant, 2006). A recent survey found that more Americans now get their news from the Internet (61%) than from newspapers or radio, and over half (52%) said they search for and spend time reading sports news online (Gross, 2010) The gains in online readership over the years may suggest high credibility for news and sp orts websites, but these websites often lack sufficient factual verification, editorial review, and anal ysis of content (Chung, Kim & Kim 2010). Online credibility has been questioned by some scholars (Newhagen & Levy, 1996), and there is a dearth of res earch on the impact of fan identification on online credibility. Sports Media and Credibility For this study, the primary focus will be on the components of online credibility for sports news websites, sports blogs, and social media, and the inherent diff erences from traditional sports media. In the Information Age, perceptions of media credibility have taken a hit in the eyes of citizens as people are now generally skeptical of news from the three major media channels: television, print, and online (Kiou sis, 2001). Others have found that blogs are perceived as more credible than traditional media by Internet users because users often seek out blogs that support their views (Johnson & Kaye, 2004). Assumingly, fans of sports teams seek out blogs focused s pecifically on their favorite teams, but not all of these blog authors are trained journalists. By looking at the role of fan identification and source medium on credibility, sport media executives can better allocate their resources and sports journalist s can better craft their articles to appeal to the most sports fans. Although the subject is sports journalism, the amount of scholarly research on this subject is extremely limited. However, the overall principles of online and traditional media credibi lity are applicable to sports (Oates & Pauly, 2007).
16 of reporting and that journalism should not use sports as an ethical straw man against which to defend the virtue of Much of the academic research on online credibility has focused largely on political news, with online sports journalism receiving seemingly no investigation from academia. Because the literature on sports media (both print an d online) is severely lacking, there is a need for a study on this previously unexplored research area. In turn, this study will focus more on the broader concepts of online credibility, while examining the credibility relationship that remains with print media. Since the turn of the century, the Internet has been blamed by some for the decline in media credibility (Johnson & Kaye, 2000). Because both print and online are being examined, it is useful to explore literature that compares the two. Kiousis ( 2001) looked at the perceptions of news credibility for three media channels: television, newspapers, and online. The results suggest that people are generally skeptical of news from all three channels. However, newspapers were rated as having the highes t credibility, followed by online news, and, finally, television news. The author suggests that the dissemination of new technologies often shifts opinions of older media. In essence, access to the Internet may have increased the public trust in newspape rs, while simultaneously reducing the trust in television. A reliance on online and traditional media has also been shown to be a strong predictor of perceived credibility of online sources (Johnson & Kaye, 2000). Those who use media most frequently typic ally find it to be more credible than others who use it less often. In a study of politically interested web users, Johnson and Kaye (1998) examined the credibility of online publications and traditional print media. The study
17 found that the web users vi ewed online newspapers and online candidate literature as more credible than their traditionally delivered counterparts. However, both traditional and online media were only judged as somewhat credible. Conversely, political opinion polls have been found to be more credible for traditional news m edia than online polls (Kim Weaver & Willnat, 2000). Recent technological advancements have made the job of the journalist much easier, but have also changed how people get news and what they expect from it (Meye r, Marchionni & Thorson, 2010). Meyer, Marchionni, and Thorson (2010) assert organizational credibility. The authors argue that in the modern digital age, perceived exp ertise is predominantly determined by the level of shared meaning between sender (journalist, blogger, etc.) and receiver (audience). Gunter et al. (2009) examined the rise of blogs as news sources of significance. The authors discovered that although so me blogs have become reliable information sources in specific news areas, most do not have the key credibility characteristics of mainstream news and do not drive the public trust. Online newspapers are no longer a simple web based version of a newspaper. They now implement additional content and provide users up to the minute updates of only news site, which can offer users another perspective on national and global news. Chung, Kim and K im (2010) examined the credibility of online newspapers and divided it into three distinct categories: mainstream, independent, and index. The mainstream category is the most common online news form and is basically a redistribution of printed newspapers. The
18 site only. The index category is characterized as online search engines that provide a collection of online news content. The results of the study indicate that the mainstream type of online newspapers, such as USA Today, received the highest scores on most credibility items. Like some readers of online only news sites, web users seek out blogs that are tailored to their specific interests. However, many blog au thors are not trained journalists, and do not feel compelled to be objective with their arguments. Traditional journalists often cite the 2004 presidential election as an example of flawed reporting where some political blogs released incomplete exit poll s, which erroneously predicted a victory for Democratic candidate John Kerry (Carlson, 2007). The author argues that journalists have a role as authoritative providers of political news, but that the current media environment has new forms of complexity a nd competitiveness. Johnson and Kaye (2004) found that blog users judged blogs as highly credible and more credible more depth and more thoughtful analysis than is av Fairness to all political parties is often considered a hallmark of traditional journalism, but the authors suggest that bias is seen as a virtue by blog users. For example, the majority of respondents rated themselves a s conservative, so they actively seek out blogs that support their views. This is similar to sports fans who seek out blogs that are dedicated to their favorite team but may not be objective or critical of that team. Theoretical Framework Social identity theory is inherently based in comparisons, so this theory can be applied to sports fans who often compare themselves to the outgroup (fans of rival
19 may have important bene fits for psychological health involves the strong ties fans often feel for their chosen sport teams (i.e., those fans who are highly identified with their being by increasing temporary and endu ring social connections for the fan (Wann, 2006). Conversely, this beneficial relationship for fans is moderated by threats to social identity, such as a The highs and lows experienced by wins an d losses for fans may even extend past the sports environment itself. Van Leeuwen, Quick, and Daniel (2002) assert that Essentially, students at a university, such as the s ample population of this study, may feel closer to their university through sport than everyday college life. The social identity that accompanies identified sports fans goes beyond on field success and is profoundly affected by interpersonal relationship s. College students list their parents, the talent of the players, geographical reasons (i.e. rooting for the home team), and the influence of following their favorite te am (Wann, Tucker & Schrader, 1996). At certain institutions, university athletic programs also provide a sense of communal involvement within the institution, the local community, and, in some cases, an entire state (Melnick, 1993). Some scholars have fo und a positive relationship between identification for a enjoyment with the university, the extent to which the university met expectations,
20 involvement with the university, and persistence at the university (Wann & Robinson, 2002). This study also examines the impact of user comments on perceived credibility in sports articles. The social identification/deindividuation (SIDE) model of computer mediated communication effects is useful for understanding the influence of visually level of identification with the people posting comments is expected to affect whether the comments influence evaluation s of credibility. Unlike Facebook wall posts, the SIDE model puts primary emphasis on visual anonymity, which is essential for predicting and understanding behavior in new media (Lea, Spears, Watt & Rogers, 2000). Applying the SIDE model to their study, Walther et al. (2010) examined the influence of user comments on perceptions of YouTube anti marijuana public service announcements ( PSAs ) The results showed that supportive or negative comments not affect attitudes toward marijuana. However, the combination of user comments along with the social identification of the participants to the users affected both PSA evaluations and attitudes towards marijuana positively or negatively, depending on th e tone of the comments. This is useful for examining the role of user comments in sports articles where most of the users are completely anonymous. One co mponent of this study examine d how identified sports fans react to the tone of the user comments. H ighly identified sports fan are, perhaps, more likely to find the author and article more credible if the user comments are positi ve about their team than if the comments are negative. Also,
21 participants with low fan identification may not differentiate i n perceived credibility because of user comments. Ultimately, many factors play into the level of credibility attributed to online sports articles. The purpose of this study wa s to examine the impact of the medium and sports news source on the credibil ity of sports articles for sports fans with both high and low identification Looking specifically at differences in online media (sports websites, Facebook Twitter etc.) and comparing it with a more traditional form of media ( wire service ), this study tes t ed the effects of sports fan identification on the perceived credibility of a sports article. Using fan identification and social identity theory as the primary theoretical frameworks, the study also analyze d how attitudes towards a website and a sports article are affected by positive and negative user comments. Building off of Chung, Kim and Kim examined the influence of the website itself on credibility by comparing a mainstream sports news source (ESP N.com), an independent sports news source (a sports blog), and an index sports news source (a Facebook note ) in the experiment. Currently, there is no consensus on the elements that build online credibility, so this study aimed to provide media professiona ls a blueprint of what works for each different medium. Several fac ets of online journalism will be addressed, including new insights on social networking websites. The relevance of this study is also amplified by the sheer lack of research on sports jou rnalism. However, the scope of this study is not limited to sports as all features of perceived credibility can be dictated by some social identity, such as political partisanship. Ultimately, the present study is meant to bring clarity and insight to th e evolving discourse on online credibility.
22 The followi ng chapters will illustrate the different aspects of the study that led to the findings and final conclus ions. Chapter 2 will discuss the relevant literature that has been published about the elements that are being examined in this study: credibility, social identity theory, fan ident ification, and the SIDE model. For c redibility the literature will include studies that have examined medium, source, and online credibility, and the criteria that have been used to measure credibility empirically. The literature on social identity theory will include research on group differentiation ingroup favoritism and outgroup bias and social identity on media and sports. S tudies on fan identification will examine identification with a specific team and how this can influence sport consumerism, information seeking behavior, and connection to a university or community. The final section will discuss the literature published on the social identity model of deindividuation effects (SIDE model) and the potential social influence of anonymous online users Chapter 3 will explain how the resear ch questions and hypotheses were formulated and developed and why they are relevant to the study Chapter 4 will lo ok at the methods used for t his study, focusing on the sampling method, the experimental instrument, and the experimental procedures. Chapter 5 will discuss the results of the experiment and provide all the relevant data and analysis that was conduct ed t o examine the research questions and hypotheses Finally, chapter 6 will discuss the overall findings of this study and interpret why the results either did or did not reflect the previous literature This chapter will also delve into the implications f or future research, the limitations of this study, and the final conclusions that can be drawn from the study.
23 CHAPTER 2 LITERATURE REVIEW Credibility With widespread access to sports information online and a platform in which seemingly everyone with a co mputer has a voice, media credibility has recently come into question (Gunter et al., 2009). Traditional news sources and their online counterparts are subject to professional and ethical pressures to provide unbiased, accurate information, but these same pressures are not expected for many Internet websites (Calabrese & Borchert, 1996). Unregulated Internet media sites, such as blogs, also operate in a 24 hour window, so these sites are more likely to report on rumors that traditional media would take th e time to investigate (Gunter et al., 2009; Bucy, 2003). The current media environment still operates in three distinct channels (print, broadcast, and online), but the credibility of every channel has been questioned by media users (Kiousis, 2001; Pew Re search Center, 2010). In a recent survey by the Pew Research Center (2010), it was found that the public continues to take a skeptical Although this s tudy is focused on the impact of online credibility for sports articles, addressing all aspects of media credibility offers a more accurate perspective on the other fo rms of reporting, and the principles of media credibility are applicable to sports media (Oates & Pauly, 2007). Credibility in the media is often described as having two broad components: medium credibility (Gaziano & McGrath, 1986; Kiousis, 2001; Sundar & Nass, 2001) and source credibility (Hovland & Weiss, 1951; Markham, 1968;
24 Whitehead, 1968). Medium credibility focuses on the differences in perceived credibility based on the format in which the information is presented, such as print, television, radi o, or Internet (Gaziano & McGrath, 1986; Sundar & Nass, 2001). Westley and Severn (1964) conducted one of the first comprehensive studies on medium credibility and noted that demographics, such as age, gender, and education, are variables that can dictate than print news at the time. The authors also found that people did not always feel that the medium they preferred most was the most credible. In a contemporary study on medium credibility, Kiousis (2001) found that people considered newspapers to be the most credible medium, followed by online news, and finally television news. Source credibility, on the other hand, involves the impact of different communicator characteristics and how the characteristics influence the processing of media messages (Addington, 1971; Markham, 1968). In these studies, the communicator is typically defined as a person, group, or organization. Earlier scholars determined that the two key component s to source credibility are source expertise and trustworthiness (Hovland & Weiss, 1951), although later scholars argued that this description of source credibility was too simplistic (Markham, 1968). Competency and objectivity were also added as factors that contribute to source credibility (Whitehead, 1968). Berlo, Lemert, and Mertz (1970) argued that source credibility was multidimensional with three dimensions: safety, qualifications, and dynamism. Although there has been debate over the years as to how to define it, Bucy (2003) classified the dimensions of credibility as believability, fairness, accuracy, informativeness, and depth.
25 Some argue that the legitimacy of news is a matter of branding, where particular news brands have credibility among new s professionals and news consumers (Gunter et al., 2009). News branding is developed through high standards of credibility. Among the most important aspects of credibility are factuality and impartiality (Westerhahl, 1983). Factuality is determined by i ts truthfulness and its relevance, while impartiality is defined by balance in coverage and neutrality. The key defining aspects of truthfulness were defined by McQuail (1992) as accuracy, factualness, and completeness. He states that news reports should be devoid of opinion and should correspond with verifiable versions of reality. McQuail (1992) also asserts that stories should have enough detail for news consumers to have an accurate impression of the issue covered or the event that occurred. Relevan ce is defined by the idea that news only has value to consumers if it deals with matters that currently concern to them (Gunter et al., 2009). Impartiality can occur on several different levels. It can be conceived as bias in the selection of news storie s with more coverage being given to certain topics (Gunter, 1997). Impartiality can also occur within the stories themselves with more attention being given to specific sources or to specific viewpoints than to others. While news media is an important asp ect of American life, only about half of the adult population is properly exposed to it. Using a national sample of nearly 25,000 respondents, Ksiazek, Malthouse, and Webster (2010) found that about half of the American adult population is classified as n ews avoiders and the other half as news seekers who read newspapers, news magazines, and the Internet, as well as watch news on cable and network television. According to the authors, news seekers make up about 50.5% of the population and tend to be olde r, have greater income, and are
26 twice as likely to have a college degree. In a related study, Chan and Leung (2005) potential online news adoption. The authors discov ered that experiencers (people who savor new experiences) read more online international news. In contrast, survivors (people who live narrowly focused lives) seldom read news online. Chan and Leung (2005) also discovered that interactivity is important for satisfying the enjoyment needs and desires for self expression of certain online users. Perceptions of Online Credibility Online journalism is subject to the same credibility standards of other journalistic media, but the round the clock nature of the Internet has changed the way news is reported (Arant & Anderson, 2001). The emergence of the Internet as a news source in the mid 1990s altered the news landscape and introduced a different news reception platform that was still connected in certain ways to traditional news media (Gunter et al., 2009). Early comparisons of offline and online news found few differences between the two (Peng et al., 1999). However, online news outlets became more sophisticated in their presentation of news when the capabil ities of the online format became more advanced (Gunter et al. 2009). The professionalism of online news was questioned in regards to fact checking and accuracy, and new forms of storytelling challenged the established norms of journalism (Deuze, 2003). The credibility judgments of online news were originally linked to both specific news organizations and their reputation (Gunter et al., 2009) and to the perception from news consumers of the Internet as a medium (Choi et al., 2006). Suggesting that onlin e news is more about public subjective perception by audiences and a function of their cognitive processing
27 mechanisms, rather than simply an innate quality of news st ories or sources high standard, but the perceptions of credibility have changed with newer forms of sourcing and storytelling available to online journalists. Online new s has become a staple of American culture, but the vast amount of misleading information on the Internet has called online credibility into question in the past (Newhagen & Levy, 1998). Although the print medium was shown to be most credible medium in an earlier cross medium study (Kiousis, 2001), print news media has recently experienced a fall in public trust (Gunter et al, 2009). A Pew study found a significant drop in the percentage of people who said they believed what they read in newspapers from 84 percent in 1985 to 59 percent in 2006 (Pew Research Center, 2006). However, some scholars have discounted the value of credibility as it relates to media consumerism. Blake and Watt (2002) found that there was no link between media credibility and eithe r support for free expression or purchasing newspapers. than for its impact on such bottom line behaviors as buying newspapers or supporting First Amendment righ According to a recent poll, Americans now spend as much time on the Internet as they do watching television (Brustein, 2010). The poll indicated that people under 30 years old have spent more time with the Internet than television for years, but now even people over 66 are spending about eight hours a week on the Internet. The prevalence of blogs and social networking websites, such as Facebook
28 media environment has changed the way credibility is perceived by the mode rn, tech savvy individual (Gunter et al., 2009). The perception of bias or a lack of accuracy in online news arose from observations that standards of fact checking are less stringent than they are in mainstream news distributed by news broadcasters and ne wspapers (Bucy, 2003). There have been differing accounts from some in the online media industry about the fact checking process. While the Internet circulates an abundance of information on a minute by minute basis, there may be pressures or temptations to take short cuts in fact checking (Allan, 2006). Arant and Anderson (2001) found that this perception was reinforced by some online news editors who claim that less time is spent verifying facts for online posts as compared to news intended for publicat ion in a traditional news medium. However, in the context of blogs, some argue that bloggers check each 2006). Additionally, online journalists claim to place gre ater trust in web based news than more traditional sources for perceptions of source accuracy, believability, fairness, and overall credibility (Cassidy, 2007). Those who work in print journalism typically regard print newspapers as the more credible news source, but do not dismiss online news sources as completely lacking credibility (Cassidy, 2007). Factors that Influence Online Credibility Among online news websites, the reputation of the media source and linking to several other sources can profoundly a ffect credibility. Chung, Kim and Kim (2010) examined the credibility of online newspapers and divided it into three distinct categories: mainstream, independent, and index. The results of the study indicate that the mainstream type of online newspapers such as USA Today received the highest
29 scores on most credibility items, but that the index type was the highest in attractiveness, creativeness, and interesting items. The authors assert that a website like Google News presents users with many choices of content, which empowers users. The Drudge Report an online only news site, had the lowest reported scores of expertise, trustworthiness, and attractiveness, perhaps because it was originally created 0). For citizen journalism websites, Johnson and Wiedenbeck (2009) found that information about the author and hyperlinks to other information enhanced perceived story credibility. The results showed that credibility was enhanced most when both a hyperli nk and writer information were included and, to a lesser extent, when one of the two was present. Choi et al. (2006) investigated the perception of medium credibility of news coverage about the Iraq War. The authors conducted an online survey of respondents who either listed themselves as war supporters, opponents, and neutrals. The results showed that war opponents viewed the Internet as the most credible medium because they perceived the Internet as less aligned with a pro government stance. The opponent group viewed the Internet as more credible than did neutrals or supporters, and also showed a strong negative correlation between perceived pro government alignment an d perceptions of online credibility (Choi et al., 2006). Also, the opponent group cited diversity of information and views on the war as the main reasons for perceptions of high Internet credibility. As a news source, the reputation of the Internet is con tinually evolving. Gunter et al. (2009) asserts that the credibility of news, whether it be online or offline, is mediated
30 partisanship or neutrality of news consumer s on specific issues. The authors view the provided by a number of suppliers. They argue that in terms of credibility, the focus should be on specific news suppliers, i n which Internet users typically trust online news produced by established news providers more than news produced by independent online sources. Additionally, established news brands offline that migrate online generally command more trust from consumers than newer online only brands (Gunter, source was perceived as more trustworthy than its traditional news source. Blog Credibility Blogs have become an onli ne media sou rce in which authors and readers can p. 6). The popularity of blogs continues to rise on the Internet, and many issues published in blogs are recognized as part of the public discourse (Lenhart & Fox, 2006). However, many blogs are maintained by untrained laymen who do not operate with strict ethical standards for journalism. Because blogs can be much more informal than traditional news websites, blog s can offer re aders news as well as a personal interpretation of the news (Lenhart & Fox, 2006). Additionally, b logs present readers with a participatory discourse and interaction that distinguishes it from traditional media (Shannon, 2006). Today, major news organizat ions typically operate across more than one media platform. Print newspapers and broadcast news stations generally have a complimentary online website with additional information. There is also evidence that
31 suggests that the reputation of established ne ws brands carries over from the offline world and int o the online world (Gunter, 2006 ). This principle may also apply to blogs, given their highly opinionated nature and the ability for anyone to create a blog and publish their thoughts on a subject (Gunt er et al., 2009). As blogs become more prominent and influential in our society (particularly blogs affiliated with a news organization), the imperative to know whether they can be trusted is increased (Gunter et al., 2009). Looking specifically at news b logs, Robinson (2006) identified seven different readership forum, 3) a question and answer format, 4) a column or opinion for the web, 5) a rumor mill blog, 6) a rou nd up of news summaries that promote the publication, and 7) a confessional diary written by a reporter about a beat. Robinson (2006) suggests that even with the variety of content offered by blogs, rarely do blogs offer news in a traditional sense. This has brought forth the emergence of a post modern form of journalism, which can be both nonlinear and interactive. Modern news blogging, thus, breaks the boundaries of conventional reporting through elements such as speculation and first person narration mainstream journalism blogs have now become a way of adopting the potential of blogs within traditional journalism. Demographic variables, such as frequency of Internet use and author gender, can also imp act perceived credibility of blogs. One study of regular web users in the United States found that frequent Internet users perceive blogs to be the most reliable information outlet across all media because they offer more depth and thoughtful
32 analysis tha n traditional media (Johnson & Kaye, 2004). Armstrong and McAdams (2009) examined the influence of gender on perceptions of credibility for informational blogs. They found that male authors were deemed more credible than female authors, and that the writ ing style and blog topic were likely to influence the perceived credibility of the post. Also, because the blogs were informational blogs, information seekers perceived the blogs as more credible than non information seekers. Because research on the perce ived credibility of information on social networking sites, such as MySpace and Facebook is extremely limited, it is a research area worth testing empirically. Metzger, Flanagin, and Medders (2010) used focus groups to examine how people view information and source credibility. The authors found that most Internet users rely on others to make credibility assessments, often by using group based tools. The participants cited web based applications and social networking sites to help them assess informatio n or its source. The authors state that social networking, and digital media in general, deviate from a perception of credibility based impenetrable, and singular down concept with people often deferring to experts on a subject. Today, bottom up assessments of information quality are easily constructed through collective efforts enabled by technology ( Metzger, Flanagin & Medders, 2010). In the Technological Age, perceptions of credibility are seemingly changing daily. In the sports world, fans have a bevy of options for online sports information, such as online versions of traditional media outlets (ex ESPN.com and SportsIllustrated.com), sports blogs, and social networking sites (ex. Facebook and Twitter ). Because re
33 tweeting and re posting hyperlinks has become so simple, erroneous information can spread extremely quickly. This current study tests the credibility of these online sports sources to see how sports fans perceive the credibility of the information, and to potentially resolve some of the issues that damage credibility. Social Identity Theory While credibility is judged on several factors dealing with medium and source, it is inherently based on the perceptions of media consumers. These perceptions can be concept which derives from his knowledge of his membership of a social group (or groups) together with the value and asserts that maintaining memberships in important social networks will result in an concept social comparisons to develop and mai esteem. Additionally, an that person belongs (the ingroup). The individuals within the group typically develop a system of role relationships, so cial norms, and values, which regulate their opinions and actions (Turner, 1982). This group structure eventually evolves with mutual interaction and influence. Turner argues that positive self esteem motivates a person to use social comparisons, which d ifferentiate oneself from the ingroup through positively valued outgroup).
34 Modern scholars have suggested that emotional involvement with a group results in a reduction of unce rtainty, which produces a favorable evaluation of ingroup members relative to outgroup members (Hogg & Mullin, 1999). The factors in which ingroups are most positively evaluated become emphasized by the group member, and negative distinguishing characteri stics of the outgroup are favored as well (Jackson et al., 1996). Ellemers, Kortekaas, and Ouwerkerk (1999) investigated three different components of social identity to better understand how specific group characteristics affect reported levels of social identification. Relative ingroup size (majority/minority), relative ingroup status (high/low), and the group formation criterion (self selected/assigned group membership) all affect the extent to which people identify with a group. The authors argue tha t the relative size of the ingroup (whether the group makes up the majority or minority of a population) has the largest impact on whether people self categorize as group members. Conversely, group self esteem is only affected by the relative status of th e group, where members of lower status groups are expected to show less social identification than members of groups with higher status (Ellemers, Kortekaas & Ouwerkerk, 1999). Furthermore, minority group members reported strong self categorization as gro up members and strong personal identification. For example, an African American would be more likely to categorize himself as African American and identify strongly with that group than someone who is part of the majority group, such as a White person. Th e overall opinion one has of his or her ingroup can also have a profound impact on the judgments one makes about others (Mar quez, Yzerbyt & Leyens, 1988). Using social identity t heory as the basis of their study, Marques, Yzerbyt, and Leyens (1988)
35 predic ted that judgments about both likeable and unlikeable ingroup members are more phenomeno n appeared within the study. The results also suggest that ingroup favoritism can emerge in the form of an outgroup bias to preserve the overall positive image of the group. In a related study, Ojala and Nesdale (2004) examined how bullying among childre n could be moderated by ingroup norms and perceived threats to group distinctiveness. The authors discovered that the participants were much more likely to have retained an ingroup member when he behaved in accordance with group norms (i.e. did not bully other ingroup members). The results also suggest that bullying was more acceptable when directed at an outgroup member who was similar and possibly represented a threat to the ingroup. This is analogous to sport where an as fighting, toward the ingroup (teammates) is often seen as more detrimental to team unity than a negative act toward the outgroup (members of the opposing team). Media characterizations may also have a major influence on social comparisons. Mastro (200 3) argues that the comparative dimensions of these social categories may not be based in reality, and television images have the potential to influence comparison between groups. The author determined that negative racial depictions in the media were foun d to be associ ated with social judgments. Mastro (2003) argues that the media can prime certain aspects of social identity, and can impact soc ial judgments and self esteem. In sports media, social identity and social judgments can be i nfluenced by the ra cial depictions of athletes. Bob Cost as, for example, came under scrutiny during
36 the 2012 London Olympics for making a comment about Gabby Douglas becoming the first African around event in gymnastics, despite there being a n imaginary barrier up for gymnasts of her race (Holmes, 2012). Holmes (2012) a rgued that while only 6.61 percent of the participants in American gymnastics programs are Black and 74.46 percent are White, there wa s no need f or Costas to draw attention to her historical achievement ; Douglas just happened to be a minority in a historically White sport. As long as there are perceptions of certain sports being divided along racial lines (ex. swimming as predominantly White) there is the potential for the med ia to prime social judgments from sports consumers toward athletes. Group Differentiation and Categorization Because social identity is so rooted in group comparisons, Tajfel and Wilkes (1963) assert that when certain categories are deemed important, diffe rentiation between groups is intensified. In their article on social identity theory and social mobility, Jackson et al. (1996) examined the social creative strategies people use when they are members of negatively distinctive ingroups. The authors found that when an individual perceives that there is a realistic possibility of changing group membership, he or she will move away from the negatively distinctive ingroup and toward a positively distinctive one. This is either done psychologically (by percei ving themselves as less similar to the ingroup), in reality (by joining another more desirable group), or both. The authors also discovered that an individual may be less motivated to distance himself from the negatively distinctive ingroup if membership is temporary because temporary membership is less threatening to social identity than permanent membership. Group differentiation can also be learned through the media characterizations of ingroup and outgroup members (Mastro, 2003). Anastasio, Rose, and Chapman (1999)
37 and may also create the public opinions it seeks to replicate in the news. According to the authors, even subtle nonverbal cues from television newscasters can influence voting behavior. Their study was designed to mimic the media coverage of the O.J. Simpson case, in which opinions of guilt or innocence were depicte d as being correlated with racial group membership. The experimenters showed a video of a peer tribunal in which a fraternity member was being accused of vandalizing school property, where members of Greek organizations were classified as the ingroup and nonmembers as either portrayed as members of Greek organizations or as nonmembers, revealing their opinions of his guilt or innocence. The authors found that homogeneity of opinion recommended punishment. When opinions of others were perfectly correlated with group membership, Greek subjects sided with the defendant and nonmembers sided against him The authors assert that this type of media exposure can exacerbate the consider the information about a trial closely. Societal roles, such as cultural and gender roles, are also subject to the potentially harmful differentiation that accompanies being a member of a less desirable outgroup. Looking at stereotypical gender roles, Amancio (1989) examined the social roups and how individuals in these groups deal with social stereotypes. The results suggest that members of a dominant
38 group can manage social stereotypes in a way that maintains a positive individual distinctiveness. Conversely, members of a dominated g roup are more likely to strive for intergroup and interindividual differentiation by either identifying more with the outgroup theory, Outten et al. (2009) tested th e capacity for group identification to foster beliefs in predict psychological well being. The participants (all self reported African Americans) who were higher in raci al group identification reported having a more positive well being. The findings suggest that the relationship between minority group identification and well can respond effec tively to a collective disadvantage. Social Identity and the Media Media usage and media messages have also been shown to differ by race, which has been shown to relate to social identity (Fujioka, 2005; Adams & Cleary, 2006). For example, minorities ar e more likely to consider media content as real than Whites (Greenberg & Brand, 1994). They are also more critical when evaluating how the media represents ingroup members (Davis & Gandy, 1999). Some suggest that African Americans have developed strategi es for dealing with biased media images and representations of Blacks to protect themselves from possible negative influence (Davis & Gandy, 1999; Outten et al., 2009). Fujioka (2005) examined the influence of Black images in news and entertainment media action. Basing his study on group threats and coping with threats to social identity, the author found that as Black respondents found Black media images more negative, they judged the images to be less accurate. Because of the media representations, the
39 respondents (all of whom were African American) perceived there to be a lower evaluation of Blacks by other ethnic groups, which, then, encouraged Black respondents to endorse affirmative action (Fujioka, 2005). Conversely, increasing minority staff members at a newspaper may not have an impact on attracting minority readers. In a study on newsroom diversity, Adams and Cleary (2006) found that increased minority staffing on edit orial content at newspapers did no t correlate to increased subscriptions and trust in local newspapers by minority residents. Surprisingly, one finding from that study suggested that minority reader trust may actually be hindered by increased minority staffing levels (Adams & Cleary, 2006 ). A similar issue arises in the sports world when college sports writers are alumni of the university they are covering or when a retired athlete becomes a neutral sports analyst and provides commentary on a former team. This scenario could hinder reade r or viewer trust for identified sports fans who seek unbiased reporting and analysis from members of the sports news organization (like they are more likely to receive at ESPN vs. a fan sports blog). There have also been concerns from women abou t their re presentations in mass media and their roles in the profession. Ferguson (1990) points to the liberal feminist movement, which aims to change the professional pecking order of women in media organizations and their under representation in the media. The a uthor disputes, however, the notion that women who are promoted in media organizations will incorporate a feminist agenda within the media. He argues that there is a common representation and more empowerment from women, but that these assumptions are
40 rarely manifested in actual news content. Thiel (2004) conducted a qualitative study on women online journalists to examine the ways in which the Internet has changed female ro les in newsrooms. She argues that new media has forced women to negotiate their identities as a journalists and technophiles in a continually evolving online medium. The women in the study suggested that online newsrooms offered them an opportunity to in crease cultural capital and become more tech savvy, but that these newsrooms are led by mostly male managers from traditional newsrooms with no technical background. While the subjects believe there is a need to shift their identity to achieve greater suc Despite the overall growth in online newsrooms regarding women and minorities, the sp orts desk remains overwhelmingly white and male (Lapchick et al., 2011). In a 2010 Associated Press Sports Editors (APSE) analysis of more than 320 websites and newspapers, the APSE received a C+ grade for racial hiring practices and an F grade for gender hiring practices. The percentage of sports editors who were women or minorities fell from 11.7 percent in 2008 to 9.4 percent in 2010. The percentage of white males, however, increased by 3 percent for sports editors (Lapchick et al., 2011). The lack of minority and female voices in the sports world may impact the way sports consumers perceive credibility because the coverage may not accurately reflect the sports, the athletes, or the fan demographics Certain professional sports, such as the NFL (67%) and the NBA (77%), are made up of mostly African American athletes (Lapchick et al., 2010a; Lapchick et al., 2010b), and if the people making the decisions about coverage of issues are predominantly white, then the coverage can get skewed to
41 push cer tain a gendas (Paulsen 2011). For example, during the 2011 NBA lockout, it was in the best interest of the media organizations to skew coverage and opinions toward ownership in the labor dispute with billions of dollars in television revenue potentially lost wi th a cancelled NBA season ( Paulsen 2011). However, media organizations, such as ESPN and Time Warner, need a perception of o bjectivity from fans, so they did not blatantly come out against NBA players, although spent much of the coverage on ways to end t he lockout ( Paulsen 2011). Because sport media members are predominantly white, identity conflicts among races can threaten newsroom cohesion and perceptions of impartiality from readers. Female journalists in the sports world also experience similar ide ntity conflicts. In a qualitative study, Hardin and Shain (2006) conducted interviews with 20 women sports journalists. The authors found that female sports journalists must negotiate tensions to their identity, where they are forced to balance their con flicting identities as their consent to be a part of a male dominated domain. The authors even suggest that the interviewees perpetuate myths about women who attain power i n newsrooms as either too much like men (i.e. too bossy) or too much like women (i.e. sexualized, not profession in search of careers that involve less cognitive and emo 334). This history of identity conflicts in media organizations even extends past our own mass media organizations and universities are significa ntly responsible for both
42 formalizing and accrediting informal knowledges of identity argued that professionals within these institutions bridge local, national, and global spheres of social belonging. He references the media org anizations that were formed in Germany after the Berlin Wall was torn down and differing ideologies of East and West German journalists were at odds in newly for med German newsrooms. D espite East and West Germans working together in these newsrooms, Easte rn Germans were (p. 481). The author concluded that these workplace dynamics were manifested in media representations, and identity differences were ingrained in the German culture through the media. Social Identity in Sport Spectatorship Besides the social identity of the journalists, Wann and Grieve (2005) suggest that (p. 533). Research on the social identity of sports fans indicates that fans also display ingroup favoritism a nd outgroup derogation ( Wann et al 2001). Also, these biases are most likely when individuals are highly identified with a team and have been threatened. Wann and Grieve (2005) designed a study to better understand threats to social identity in a field setting. The authors h ypothesized that fans of a home team and fans of a losing team would experience threats to identity, and, subsequently, show high levels of ingroup favoritism. Data gathered at two North American college basketball games confirmed a presumed interaction i n which highly identified fans rooting for a home team that had lost would exhibit the most amount of bias, which were both overly positive biases of ingroup members and unjustly negative biases of outgroup members. An
43 example of this situation is when a supporter of the home team reacts to a questionable call by an official during a game (Wann & Grieve, 2005). If this person reacts by yelling to some and the o utgroup to others. The bias occurs when other ingroup members see the behavior as much less negative, or even justified, than they would if this reaction had been exhibited by an outgroup member. Along the lines of Wann & Grieve (2005) study, Krumm and Co rning (2008) tested the effectiveness of moral credentials at concealing one s prejudice. Participants in the study were most swayed by moral credentials, which are pieces of evidence that can be presented as proof of lack of prejudice, by those who share d an ingroup status with the person displaying the credentials. For example, an African American making a joke about other African Americans would be seen as much more acceptable than a White person telling the same joke. In the context of sport, this th eory asserts that a person acceptable if the person making the remarks is a member of the ingroup (a fan of the same team). The fan group one is a part of can also impact how one Boyle and Magnusson (2007) conducted a study on three distinct fan groups of a collegiate basketball team (current students, alumni, and the general public) and how heightened social identity can build brand equity of a team. According to the authors, social identity formed differently for the fan groups. Team history was shown to be significantly related to social identity for alumni and the general public. Interestingly, current students were most influenced by their sens e of the basketball program being a
44 part of the community. The results provide strong support for the effect of social identity on brand equity of the athletic program overall. Essentially, social identity plays an important role in the way people consum e sports, especially at the collegiate level Developing Social Identity through Online Social Networks In the Technological Age, social network websites have also been shown to aff ect esteem (Barker, 2009). For adolescents, using social media facilitates interpersonal relationships as well as intragroup and intergroup relationships (McKay, Thurlow & Toomey Zimmerman, 2005). Barker (2009) examined the motives for social network site use for older adolescents and the influence of gender, group identity, and collective self esteem. The author found that communication with peer group members was the most important motivation for social networking. Parti cipants high in positive collective self esteem were strongly motivated to communicate with peers using a social network site. Women were more likely than men to report greater overall use, high positive collective self esteem, and high social networking use to communicate with peers. Men were more likely to report negative collective self esteem and social network site use for social identity gratifications. Additionally, negative collective self esteem correlated with social compensation. The author s uggests that adolescents who feel negatively about their social group use social network sites as an alternative to communicating with other ingroup members. Online communities and cyber societies have sprung up over the last two decades, and many have t he potential to work as a communication system which 2007). Some scholars have addressed online identity and acknowledge the anonymity
45 aspect of the Internet that can allow users to mask parts of their identity, such as age, gender, and ethnicity (Turkle, 1995; Markham, 1998). These authors suggest that these communities and this form of online communication can function as expressiveness, bonding, sharing, and identit y construction for those involved. In a qualitative study on accompany these online discussions According to the authors, the discussion boards as welcoming and supportive. Given that the discussion board was not moderated, it created a highly feminized identity f or the users, which some viewed as empowering. The authors conclude that unlike print magazines, the discussion board in an online actualizing interactivity where women can communicate directly in an environment that is seen a s non judgmental and open. The authors also acknowledge that the boards exist in a commercial forum where users are primarily viewed as consumers, so the strength and duration of the bonds formed were too difficult to measure. Social identity can also imp ingroup and outgroup. Knobloch Westerwick and Hastall (2010) tested the selective exposure to positive and negative news articles about people in similar and different age groups. Using software to log r eading times of an online news magazine, the authors discovered that younger individuals (18 30 years old) focused most of their reading attention on same aged individuals, with a preference on positive news about the ingroup. Older individuals (50 65 yea rs old) were more likely to select negative
46 news about the outgroup (young individuals) than positive news and negative news about their ingroup. Additionally, selective exposure to negative news about the es teem. Fan Identification and Fan Involvement Social identity theory suggests that people are motivated by a need to enhance their self esteem, and that this self esteem is established by being members of social groups (Tajfel, 1981). People in these group s continually make social comparisons to enhance self esteem, in which ingroup members are judged more favorably than those in outgroups (Hogg & Mullin, 1999; Hogg & Abrams, 1990). This is also true of sports fans in which highly identified fans are more likely to show favoritism towards fans of their own team and criticize fans of opposing teams (Wann & Branscombe, 1993; Wann & Grieve, 2005). Identified fans of a university sports program have also been shown to be more satisfied and more involved with t heir university than non identified fans (Wann & Robinson 2002) Additionally, t he presence of an athletics program on a university campus has been shown to influence the perceived sense of community on ca mpus (Clopton, 2007) exemplified at a school lik e the University of Florida where athletics are an integral aspect of college life for many students Because g roup identification has been shown to influence evaluations and perceptions of media (Fujioka, 2005; Greenberg & Brand, 1994), sport fan identif ication among university undergraduate students may have the same potential influence on evaluations of a sports article or a sports news source. While sports fans are similar to fans of other interests, sports fans perceive themselves as being in a group even when they are not actively part of an organized group (Reysen & Branscombe, 2010). Sports fans are a unique group of individuals,
47 psychological and behavioral attachment to spo rt, including fan identification (Sutton et al., 1997; Zhu & Won, 2010), team identification (Bizman & Yinon, 2002; Fink et al., 2009; Wann, Tucker & Schrader, 1996 ), and fan involvement (Shank & Beasley, 1998; Wann et al., 2003). Fan identification is mo re firmly rooted in psychological attachment, to their social identity (Fink et al., 2002). Similarly, fan involvement is more behavioral and has to do with the percei ved interest in sports to an individual (Shank & Beasley, attend games, and the level of satisfaction that accompanies a positive game outcome (Wann & Branscombe, 1 993; Madrigal, 1995). These fans inherently see their chosen team as an extension of themselves (Wann et al., 2001). Although most of the research in the current study covers the impact of fan and team identification, fan involvement will also be introdu ced to provide background for the behavioral attachment of sports fans to their favorite teams, sports, and other fans. Sports Fans and Social Identity Using social identity theory as their primary theoretical framework, Dimmock, Grove and Eklund (2005) ex amined the cognitive, affective, and evaluative dimensions of identification with a sports team and relationships between identification and intergroup bias. The findings suggest that cognitive (knowledge of membership to a group) and affective (emotional significance of group membership) identification were stronger for groups in which membership is self selected rather than assigned. Because team identification is a voluntary group membership, the authors determined that a cognitive affective dimension of identification is the best predictor of bias against
4 8 rival fans. Wann and Dolan (1994) also found that highly identified fans will relate a victory to internal factors such as the skill of the team or the coaching. Conversely, a loss is often ascribed to external factors, such as fate or poor refereeing, and not the play of the other team. Essentially, highly identified sports fans undergo a biased attribution process when dealing with a loss (Wann & Dolan, 1994). Sports teams have even been regarded a s hybrid identity organizations (Heere & James, 2007), meaning they have multiple components that would not normally go together. Heere and James (2007) suggest that a team may represent a collection of owners, coaches, and players, but also represent the city, state, or university in which they operate. Heere and James (2007) assert that fans no longer perceive the team and the surrounding community as different entities, but as being linked together. They also view team identity as symbolic of other ty pes of group identities. These include two main types of external group identities: demographic identities, such as geographic, ethnic, and gender identities, and membership identities, such as university based, corporate, religious, and political identit ies. At the collegiate level, many university sports teams are referred to by the state in which they reside. For example, the several other in state universities. In this sense, the external group identity goes from a university based identity (the current students and alumni who identify with the team) and extends to a geographical identity (re sidents of the state). Heere and James (2007) suggest that sports teams should work to identify the external group identities that the
49 identity and their loyalty to that team. the social significance of sport. Gao and Kim (2011) examined the impact of cultural values on spectator sport attitudes and team identification. The authors hypothesized systems that center around the pursuit of knowledge versus cultures that center on individual liberties and life enjoyment. In support of their hyp othesis, the results showed that the relationships between spectator sport attitudes and team identification were higher for American respondents than for Korean and Taiwanese respondents. The rest in sport from an early age and that the importance of sport is then manifested later in life ( Gao & Kim 2011). Effect s of T eam Identification on Self Esteem health t hrough social connections. Wann (2006) developed a Team Identification Social Psychological Health Model, which asserts that team identification facilitates a being by increasing temporary and enduring social connections for the fan, which ca n have a profound impact on social psychological health. Conversely, the author also suggests that this beneficial relationship for fans will be moderated by threats. According to the Team Identif i cation Social Psychological Health Model (Wann, 2006), team identification and social psychological health should be positively related because team identification leads to social connections which, then, facilitate well bei ng. Wann and Weaver (2009) tested this model in their study on identification
50 with a local and a distant team, and the impact this has on well being. Consistent with predictions, identification with the local team was positively related to social well be ing. However, this relationship did not appear for fans of distant teams. The results also showed that identification was a significant predictor of social integration and social coherence. Their findings imply that highly identified fans tend to view t heir social lives as satisfactory and meaningful, which can lead to social capital, generally defined as a network of relationships based on trust, mutual obligation, and cooperation (Putnam, 2000). Team identification can also impact fan perceptions of wi ns and losses. Highly identified fans have been shown to enhance their own well being after v ictories and protect it after defeat s by using bia sed cognitions when recounting the outcome (Wann et al. 2002). Madrigal and Chen (2008) examined the role of te am identification on the summary judgments of a game outcome. The authors found a self serving bias in which highly identified fans were more likely to attribute a team win to causes deemed to be under internal control of that team. They also found that fans with high identification believed that the same game outcome would occur again if the two teams played in a future match. However, highly identified fans did not attribut e losses to external forces, such as injury to key athl etes, referees, etc. (Mad rigal & Chen, 2008). In a related study, Wann and McGeorge (1994) found that highly identified fans report a greater increase in positive emotions after a win and a greater increase in negative emotions after a loss than fans with low identification The authors also found that past team success was an important predictor of fan identification level, but that levels were not affected by game outcome.
51 Team ident ification and game outcome esteem. Bizman and Yinon (2002) examined the effects of distancing tactics on self esteem and esteem and emotional responses of professional basketball fans immediately after their team played an official game. The fans were given the opportunity to increase or decrease their association with the team, and the fans typically associated more with the team after a team win than after a team loss. Additionally, self esteem and positive emotions were higher and nega tive emotions were lower, when measured after the game, instead of before. These effects were more prevalent among fans with high team identification. The authors suggest that there is a distinction between the short term and long term effects of game ou tcome on willingness to associate with a team. In the short term, team association may fluctuate based on team performance, even for fans with high team identification. In the long term, only the fans with high team identification may maintain their alle giance to the team. behavior off of it. Fink et al. (2009) found that unscrupulous acts by athletes off the field Participants were given a scenario field offense and comments from team leaders either supporting the player or admonishing the player. When team leaders came out against the athlete and claimed that the leaders supported the athlete, even highly identified fans experienced a drop in team identification. The authors argue that this was a result
52 where participants in the strong leadership response group could view the act as an anomaly, inconsistent with the group values of the team. Additionally, some evidence suggests that there is a significant correlation between team identification and a general belief in the trustworthiness of others (Wann & Polk, 2007). Team Identification in Collegiate Sports Team identification goes beyond the professional ranks as university sports teams are often the biggest draw in cities w ithout a professional team. In their article, Wann and Robinson (2002) examined the relationship between university sports team identification and perceptions of the university for enrolled students at that institution using two studies. Assessing the le basketball and football teams, undergraduate participants were asked to rank their undergraduate persistence intentions and their perceptions of the university. The results indicated significant positive relationships between identification for the two teams and satisfaction and enjoyment with the university, involvement with the university, the extent to which the university met expectations, and persistence at the university. In the same article, Wann and Robinson (2002) replicated the effects in a while controlling for level of sport fandom. The authors again found that higher levels of identification with the ins school, in general. The findings from both studies imply that university administrators should better promote team identification among the student body, which may, in turn, result in more favorable evaluations of the university by students.
53 the presence of an athletics program on campus and the perceived sense of community their study, Clopton and Finch (2010) ex amined the contribution of team identification to the social capital of college students at 21 NCAA institutions. The results indicated that Also, race and gender were both significant in predicting social capital through fan identification. The results suggest that there is an empirical link between the psychological benefits of fan communities and the contribution to the community as a wh ole, and that this can lead to potential community building activities. Essentially, people in fan communities, such as college students at a Division 1 athletic university, are more likely to volunteer or give back to their community in some way than tho se not involved in fan communities (Clopton & Finch, 2010). In a related social capital study, Perks (2007) tested whether organized sport participation as a youth predicts involvement in community activities as an adult. The findings showed that youth s port participation was positively related to adult involvement in community activities, and that this participation lasted throughout the lifecycle. Essentially, sport participation early in life fosters social capital, which pays off as higher levels of community involvement as an adult. Sport Media Coverage and Fan Information Seeking Behavior The workplace identity in newsrooms can also impact the coverage that fans are exposed to. In a study on the coverage of doping in sport, Sefiha (2010) found that while performance enhancing drug use in sport is considered very newsworthy, there is limited coverage on the issue because of investigative costs, public fatigue, and lack of
54 medical and legal knowledge by journalists. The author asserts that these occu pational and institutional identities exert pressure over what is perceived as possible and desirable by journalists. The coverage a fan receives is, ultimately, determined by the media organ ization but even the most newsworthy issues receive less than s tellar coverage based on the institutional identities of newsrooms (Sefiha, 2010). Because of this, fans may seek out information on an issue like doping through websites that are unaffiliated with major news organizations, but may not be as credible. I n 201 2, former Penn State defensive coordinator Jerry Sandusky was convicted of child molestation, an independent investigat ion determined that school officials covered up the scandal by not reporting the abuse and Penn State was hit with NCAA major sanct ions including a $60 million dollar fine and a four year postseason ban (Chambers, 2012). I n the wake of the convi ction and the lengthy cover up by head football coach Joe Paterno and other university administrators the news coverage of the university an d its football program have been overwhelmingly negative. As the coverage continues, Penn State football fans may seek out local new s organizations or fan websites that are more sympathetic toward the university and local community to alleviate threats to a negatively distinguishable group in which they identify strongly. In this case, when seeking information, group identity and finding a news source that reflects that identity outweighs the p erceived credibility of the source. The identification of a sp orts team or sports in general inherently leads to seeking out information about that team or sport through media use (Randle & Nyland, 2008). In their study on fantasy sports leagues, Randle and Nyland (2008) found that participation in fantasy sports l eagues was significantly related to an increase in
55 traditional television, newspaper, and radio use. The results also showed that participants in fantasy sports frequently visit the websites that host their own league. The authors suggest that mass media organizations should implement interactive fantasy sports leagues via their sports websites to build customer loyalty and increase the use of their traditional media. Additionally, attitude toward the televised sport (American football), perceived ease o f use of the website, perceived knowledge of the intentions toward playing fantasy football (Dae Hee & McDaniel, 2011). t is now time to think less in terms of the longstanding relationship between sport and media, and more about sport as media given the increasing interpenetration of digital media content, sport, and 89). The authors discuss the current role that MyFootballClub (MFC), which is a popular computer game and website as well as the first Internet community to buy and takeover a real world football club in England, plays in the realm of sport media. The appeal of this franchise is due to its ontological value as real life competitions (not digital simulations or fantasy sports) where actual players, a team, and genuine competition are involved. The authors claim the medium to long term, MFC is a case study that ably highlights the changing character and dynamics of the media sports cultural created a longing to construct com munities around fan participation in the ownership and running of a team.
56 The factors that motivate broadcast viewership of sporting events are also affected by fan identification. Hu and Tang (2010) found that entertainment, self esteem, and eustress (po sitive levels of stress) positively affected fan identification, which positively impacted viewing behavior of sports broadcasting. Fan identification as a mediator was important as none of the motivational factors had a significant relationship with view ing behavior. The motivating factors that were most important to viewers were entertainment and excitement, and supporting sports figures from their own country. The authors assert that increasing fan identification clearly helps increase international v iewership and ratings. Team identification may also impact the way people respond to sports media. Wann and Branscombe (1992) found that the emotional responses to a news article were different based on degree of identification with the sports team. High ly identified fans experienced the most positive mood state from an article that described a victory for the ingroup, whose author was an admitted loyal fan of the same team. The most negative mood state was evoked when the team lost the competition and t he article low fan identification were not significantly influenced by the game outcome, group Essentially, fan but other research on the impact of fan identification on credibility is severely lacking. Fan Involvement Scholars have also identified an absence of fan involvement research for sport spectators (Kerstetter & Kovich, 1997; Bee & Havitz, 2010). According to Bee and
57 important in their development of psychologica l commitment. According to Trail et al. (2003), fans and spectators have similar motivations because both groups attend sporting events to socialize with others and escape everyday responsibilities. However, the authors found that fans attend games to ch eer on their favorite teams, while spectators root for a well played game, regardless of which team wins. While many consider males to be more involved fans, some scholars have found different results. Dietz determined that an equal number of males and females considered themselves to be sports fans, although males identified more strongly as fans than females. The results also indicated that males engage in more sport fan behavior than females. Additionall y, females were more likely to report being a sports fan because they watched or attended a sporting event, while males were more likely to report themselves as fans because they played sports and wanted to acquire sports information. Super Bowl broadcast, female viewers had more positive attitudes toward the entertainment elements of the broadcast, such as the national anthem and the halftime show, than men (Clark, Apostolopoulou & Gladden, 2009). Wann et al. (2003) found that althou gh many women consider themselves to be sports fans, men were twice as likely to be self classified as avid, highly involved fans. The study also discovered that more than half of highly involved fans attend multiple home games throughout the season. Int erestingly, the authors also discovered that involved fans admitted to changing their work or school schedules to accommodate their sport viewing plans.
58 responses to the sports page shows that the connection one shares with a team can credibility of those same articles, although previous research on this particular issue is essentially non existent, making the need for this study more vital. Because fan involvement and identification is essential to better understanding fan attitudes toward sport media, scholars have developed instruments to measure these concepts (Wann & Branscombe, 1993; Trail et al., 2003; Shank & Beasley, 1998). The results of these studies suggest that there is a relationship between involvement and sports related behaviors, such as watching sporting events on television, on a computer, and in person. Not surprisingly the previous studies found that more involved fans are more likely to attend a live sporting event or watch one on television than someone with low involvement. The fan involvement scale developed by Shank and Beasley (1998) ical feelings about sport with their behavioral habits. Sport spectating and levels of participation in sport were combined with media viewing habits to create a scale that effectively measures fan involvement and identification as it relates to media. F scale will al so be used as another measurement instrument of fan identification. The Social Identity Model of Deindividuation Effects ( SIDE ) Model While social identity theory is firmly rooted in social comparisons between ingroup and outgroup members, group polarizat ion refers to the finding that after a group discussion, individuals in the group tend to endorse a more extreme position in the
59 direction already favored by the group (Lee, 2007; Hogg, Turner & Davidson, 1990). Essentially, people move beyond the mean po sition of the group to fully differentiate themselves from people holding contrary positions. From this viewpoint, the social identity model of deindividuation effects (SIDE model) asserts that anonymous computer mediation can obscure the interpersonal di fferences that inhibit group identification, which, in turn, heightens group salience and increases adherence to group norms (Lea & Spears, 1991; Spears & Lea, 1992). In other words, anonymous communication lacks individuating cues, which shifts the atten tion from distinctive characteristics of group members, thereby making people more susceptible to group influe nce (Postmes, Spears & Lea, 1998 ). On the Internet, anonymous user comments in discussion boards and online articles are common, and modern interp retations of the SIDE model claim that the lack of individuation information fosters group identification (Lee, 2007). In an experiment on deindividuation, Lee (2007) examined how deindividuation affects group polarization in computer mediated communicati on. Before exchanging opinions about social dilemmas with three partners via a computer, participants either shared personal information (individuated) or did not (deindividuated). Consistent with the SIDE model, deindividuated participants not only exhi bited stronger identification with anonymous partners than did the individuated participants, but they were also more likely to polarize when they identified with th e partners, but this had no significant impact on a post discussion opinion shift. The author argues that the profound effects of deindividuation are highlighted in this study because the mere exchange of minimal personal
60 information created variance in t he perception of ingroup similarity and opinion polarization. In a similar study, Lee (2004) designed two experiments to see how visual representation of interacting participants affects depersonalization and conformity to group norms in anonymous compute r mediated communication. In the first experiment, participants were asked to make a decision about social dilemmas after seeing two and two fictitious, anonymous peers exchanged supporting arguments. On their computer screen, the participant was shown an avatar of him/herself and the other participants. In this study, depersonalization was initiated by the participant seeing uniform responses from the computer mediate d partners, which led to greater group members share similar values and beliefs appears to reflect the perceived nt aimed to test the causal links between depersonalization, group identification, and conformity. Participants were informed that the other avatars were for students at other schools, but the same uniform responses were shown to the participant. The res ults indicated that group identification, rather than interpersonal similarity, is the reason for the impact of depersonalization on conformity. Scholars also argue that an anonymous computer environment gives people the freedom to enact new identities for themselves and liberate themselves from the limitations brought on by identity, reality, expectations, and conventions (Turkle, 1996). Postmes, Spears, and Lea (1998) counter this by asserting that even though a potential
61 for identity replacement exists, people may not always want to free themselves from these social constraints. They claim that cyberspace provides the ideal opportunity to create a new virtual society, but that this new society will quickly resemble the old one, if people actively carry over the constraints of their real world identities. The SIDE model suggests that a deindividuating encounter in a group diverts attention from the individual level of interaction and focuses attention on the social level, which emphasizes the social boun daries of the ingroup and outgroup (Postmes, Spears & Lea, 1998). However, when group members do not identify strongly with their group, they are less likely to respect group boundaries, thus deindividuation should not increase social influence (Spears et al., 1990). Applying the SIDE model to two experiments, Postmes et al. (2001) looked at the effect of priming and anonymity on group behavior in computer mediated communication. In the first study, group members were primed with a certain type of social behavior. Consistent with the model, anonymous groups displayed primed consistent behavior in their task solutions, however identifiable groups did not. The authors suggest that the primed norm of a group happens to a greater extent in anonymous groups. The second study showed that nonprimed group members conformed to the behavior of primed members, but only when the communication was anonymous. This suggests that a primed norm can be socially transmitted from the group to new members (Postmes et al, 20 01). The ability to post comments and its impact on perceived credibility is also examined in this study, so the application of the SIDE model is useful at understanding how fan perceptions can change based on group norms. Thurman (2008) suggests that
62 r eader contributions in online articles are important for increasing circulation, providing a source of stories, and providing content for stories. Applying the SIDE model to their study, Walther et al. (2010) examined the influence of user comments on per ceptions of YouTube anti marijuana PSAs. The results showed that supportive or negative toward marijuana. However, the combination of user comments along with the s ocial identification of the participants to the users affected both PSA evaluations and attitudes towards marijuana positively or negatively, depending on the tone of the comments. Although this has not been empirically tested for sports articles, based on studies concerning the SIDE model (Postmes et al., 2001; Postmes, Spears & Lea, 1998; Lee, 2007), it is possible that sports fans can have similar reactions to the anonymous user comments at the bottom of online sports articles. Deindividuated sports fan s may be more likely to reject or accept the credibility of the articles if anonymous peers speak out in favor of or against the news organization, the author, the article itself, or the team. The dearth of research on this issue and the potential of user for analysis in this study.
63 CHAPTER 3 RESEARCH QUESTIONS AND HYPOTHESES The level of perceived credibility attributed to online sports articles is dependent on many factors, and the present study aims to shed light on the aspects of credibility that are most important to sports fans and media consumers alike. The findings from the present study can be used by those in the media industry to creat e better websites and craft more profound articles that resonate with audiences. In t his study, participants were exposed to the same sports article about a game recap involving the University of randomly ass igned to one of four different sports news sources: three online sources (ESPN.com, AlligatorArmy.com, and Facebook ) and one wire service ( Associated Press ). At the top of the article, participants saw a large banner of the assigned sports news source that the article was said to have originated from. The post test results from the assigned stimuli groups were used to determine the perceived credibility of the article itself the perceived credibility of the medium, the assessments of the four sports news sources and identification with the university The participants were also randomly assigned to one of three scenarios for user comments: positive comments, negative comments, and no comments (control). Unless participants were assigned to the cont rol groups, three positive or negative user comments were shown at the bottom of the article. The post test results from these stimuli groups were used to determine the influence of user comment tone on perceived credibility, sports news source assessment and group identity with the anonymous users. While the main purpose of this study is to examine the impact of the medium and source on credibility, the role of identity has also been shown to have a profound impact
64 on perceptions of media credibility (Ch oi et al., 2006; Wann & Branscombe, 1992; Greenberg & Brand, 1994; Davis & Gandy, 1999) This study builds off o f Chung, Kim, and Kim specific websites on credibility and c omparing a mainstream sports news source (ESPN website) an independent sports news source (sports blog) and a social networking website ( Facebook ) which can be used as an index sports news source. Many in the media industry suggest that traditional news sources and their online counterparts are subject to ethical pressures to provide readers with accurate information, but that these pressures are not expected for many online sites (Calabrese & Borchert, 1996; Newhagen & Levy, 1998; Arant & Anderson, 2001 ). Demographics, such as age, gender, and education, have all been recognized as variables that can (2001) found that people considered newspapers to be the most credible medium, followed by online news, and finally television news. Bucy (2003) found that within age groups, adults rated Internet news to be significantly more credible than television news, while students rated television news as more credible. In Rimmer a study, participants perceived newspapers as having more credibility than broad cast news, but that participants spent more time watching TV news. Because there is still debate as to what the most credible medium is and because there is a dearth of research in the field of sports journalism, the following research questions were developed to examine the perceived le vels of credibility among media and how it may be affected by fan identification:
65 RQ1 : In sports media content, w hich medium i s p erceived as more credible: online or wire service ? RQ2: Do highly identified sports fans perceive the online mediu m to be more credible than a wire service ? RQ3: Do fans with low identification perceive the online medium to be more credible than a wire service ? T he present study also looked at the differences among several online media formats (sports websites, blogs, and socia l networking websites) and tested the effects of sports fan identification and ingroup favoritism on source credibility Accordi ng to some scholars, the credibility of news sources, whether online or offline, is mediated by news consumers themselves on particular issues (Gunter et al., 2009). Established news brands offline that migrate online generally command greater public trust than newer online only brands (Gunter, 2006). Some studies have even found that a major tr aditional news source (Ranie et al 2003). Chung, (2010) found that mainstream online news sources that start as a traditional offline news source and branch out to the Internet, such as USAToday.com and ESPN.com for sports, are perceiv ed as the most credible of all online websites. Distinct from the perceived credibility of a particular ar ticle, the perceptions of each website and the Associated Press as a news source for sports information were also examined. Esse ntially, the present study looked at the concept of established news sources to see if a connection to credibility applies to sports media and an established sport brand, such as ESPN.
66 Along with news brands, fan identification may also impact perceptions of credibility. W ann and Branscombe (1992) found that the emotional responses to a sports article differed based on degree of identification with a sports team. Highly identified fans experienced the most positive mood state from an article that described a victory for th e ingroup and with an author who was an admitted fan of the same team. The most negative mood state was experienced when the team lost the game and the article author was a disloyal fan of the team (Wann & Branscombe, 1992) Additionally, individuals wit h low fan identification were not influenced by the game outcome or the suggests that fans f eel emotionally different about articles after wins and losses, so the potential relationship that may exist between perceived credibility and fan identification is worth exploring. Based on the previous literature, the following research question and fou r hypotheses were developed to test the relationship between perceived credibility of online and offline news source and fan identification, and the relationship between established sports brands and credibility: RQ4: When comparing a mainstream sports web site, a fan website, and a social networking website, which will be perceived as the most credible? H1 : The mainstr eam sports article (ESPN.com) will be viewed as more credible than the other online versions of the article by highly identified sports fans and fans with low identification Additionally, some scholars have found that higher levels of identification with the u to more positive impressions of the school in general
67 (Wann & Robinson, 2002). The sample p opulation o f the present study wa s students this, the present study assumes that fans of the team would inherently feel like part of the ingroup in this case, the university thu s, potentially giving the article and the sources a more favorable rat ing than those in the outgroup ( fans with low identification ). For exa mple, student fans of Gators athletics are assumed to have a higher group identity with the university, and will pr esumably rate an article about a Florida Gator team as more credible than a student who is a fan with low identification This situation occurs because fans have a stronger ingroup attachment to the university and want to perceive themselves as being part of a favorable ingroup (Mar quez, Yzerbyt & Leyens, 1988) H2: Highly identified f ans will evaluate the online sources more fav orably than fans with low identification H3: Highly identified fans will rate the article itself as more cr edible than fans with low identification H4 : Highly identified fans will have a stronger group identity with their univ ersity than fans with low identification Using social identity theory and the SIDE model as theoretical frameworks, the present study also analyzed how att itudes towards a website and a sports article are potentially affected by anonymous user comments. According t o Postmes, Spears, and Lea (1998 ), anonymous communication lacks individuating cues, so the attention shifts away from distinctive characteristic s of group members, thereby making people more susceptible to group influence. In studies using the SIDE model, deindividuated
68 participants exhibit ed stronger identification with anonymous partners than individuated participants, and deindividuated partic ipants were also more likely to polarize their opinions in favor or against an issue (Lee, 2007). Essentially, a deindividuating encounter in a group online diverts the attention from an individual level of interaction to a social level of interaction, wh ich emphasizes the social boundaries of ingroups and outgroups (Postmes, Spears & Lea, 1998). Partic ipants in the present study were exposed to either positive user comments about the team, negative user comments about the team, or no comments at all, whi ch may alter their perceptions of the article and its credibility based on the tone of the comments. Based on the review of literature, the following four hypotheses were developed to test the impact of anonymous user comments on highly identified fan s an d fans with low identification : H5 : Highly identified fans will find the article to be more credible when coupled with the anonymous positive user comments for each source than with negative user comments or without user comments. H6: The perceived credibi lity of the article will be influenced more significantly by the tone of the user comments for highly identified fans than for fans with low identification H7 : Highly identified fans will have a stronger group identity either with or against the anonymous users than fans with low identification based on the tone of the user comments H8: Highly identified fans will have stronger positive or negative feelings about the sports media source itself (website or AP article) than fans with low identification ba sed on the tone of the user comments.
69 Overall, this study aims to provide insight into a marginally explored research area of article credibility. Because there is no clear cut consensus on the elements that build online credibility, th is study provide s media professionals an outline of the elements that enhance perceived credibility, including a better understanding of the role of user ronment. The combination of multiple theoretical frameworks equates to findings that can be viewed from several different perspectives and applied to multipl e disciplines in media
70 CHAPTER 4 METHODS The present study was an online experiment designed to draw conclusions about the perceived credib ility of sports articles and evaluations of sports news sources Experiments can have limitations regarding external v alidity, or the generalizability of the findings across settings and populations The experim ent was looking mainly at online news sources, so to reflect the natural setting of reading an online article, the experiment was administered online. A convenience sample of college students was used based on availability and accessibility to a sample po pulation at the University of Florida, a large public academic institution in the southwestern United States. Although a convenience sam ple is not ideal for experiments because the findings are less generalizable t he use of a convenience sample of colleg e students has been utilized in previous scholarly studies examining sports fans. For example, Wann and Branscombe sports page. Wann (1995) developed a Sport Fan Motivati on Scale using a sample of 272 subjects, 116 of which were college students receiving extra credit. Additionally, Pham (1992) used a sample group of 85 undergraduate students to study the effects of fan involvement and arousal on the recognition of sponso rship stimuli at a sporting event. In the present study, t he students were recruited from seven undergraduate mass communication classes in the College of Journalism and Communications, and extra credit was offered as an incentive for the ir participation in the study. Threats to i nternal validity can also impact the presumed c ausation between variables in an experiment (Shadish, Cook & Campbell, 2002) A randomized groups post test only experimental design was used to counteract many of the potential thre ats
71 to internal validity. Random assignment controls selection as an internal validity threat (Shadish, Cook & Campbell, 200 2). To address any validity issues that may have arose with the testing instrument, a pilot study of about 15 graduate students wa s run before conducting the experiment for data analysis The goal of the pilot study was to fix any problems that could arise during the experiment. This involved incorrect wording for specific questions or having an inappropriate length to the experime nt, which may occur if there are too many or too few questions. Appropriate adjustments were made to the study based on the outcome of the pilot study and recommendations of the participants. Question clarity and the look of the stimuli were t he primary issues that were addressed after the pilot study. Some of the questions in both the pre test and post test were rewritten to avoid participant confusion. Additionally, the article used as the stimulus had formatting and spacing issues that made the artic le difficult to read. Changes to the appearance of the article were made to make it look le ss congested Table 4 1. 3 x 4 Experimental Design ESPN.com Article Sports Blog Article Facebook Note Wire Service Article Positive User Comments (30*) Positive Us er Comments (29) Positive User Comments (33) Positive User Comments (31) Negative User Comments (32) Negative User Comments (32) Negative User Comments (32) Negative User Comments (33) No Comments (30) No Comments (33) No Comments (30) No Comments (31) *Note: Number denotes the number of participants that c ompleted the study in each cell The study was a 3 x 4 experimental design and participants were randomly assigned to one of twelve groups, each receivi ng the same sports article ( Table 4 1). The artic le was taken from a previously published Associated Press article (Pells, 2012) Florida Gators and the Louisville Cardinals ( Appendix A) The article recapped a basketball ga me in which the Florida Gators had lost and were subsequently eliminated
72 from the NCAA Tournament. Some minor changes to the article were made by the author of the present study that did not affect the accuracy or fac tual ity of the article itself ( Appendi x B). During the experiment, one group of participants received the stimuli with a large T he positive comments the article itself. The second group received the stimuli with a large AlligatorA rmy.com banner (which is a popular Florida Gators sports blog) at the top of the article, as well as s team. The third group received the stimuli as a Facebook note with the same fictional author and a large Facebook banner at the top of the article. The note also had three positive comments at the end of the article. The large banners were all the same size and font and were designed to look nearly identical. The user comments were also identical for all three online sources. The fourth, fifth, and sixth groups received the same initial stimuli with the same online sources as the previous three. However, these groups received a set of three State Seminoles, the Kentucky Wildcats, and the Georgia Bulldogs) at the bottom of the article. The negative comments were t argeted solely at basketball team, and were not disparaging of the article itself. The seventh, eighth, and
73 ninth groups were control groups which received the article with one of the three online sources, but without any user comm ent s. The comments and the fictional usernames used were all written by the author of the present study. The user comments were all reviewed in the pilot study and deemed as appropriate for the experiment by the pilot study participants. Because this exp eriment is focused mainly on the potential impact of social identity, the manipulations of positive and negative user comment tone were measure d by how positively or negatively the participants identify with the anonymous users. The m anipulations w ere ana lyzed after the experiment using an anonymous user comment scale, developed by Postmes et al. (2001 ) The user comments with a positive tone ( M = 3.62, SD = 1.62) garnered a higher identification score from participants than the user comments with a negat ive tone ( M = 2.19, SD = 1.35). A univariate ANOVA was conducted, and it was determined that a significant difference existed between the two user comment tones ( F (1, 250) = 57.59, p < .001) thus, validating the manipulations. Furthermore b oth fan ide ntification groups identified significantly higher with the users who posted the positive comments than the users who posted the negative comments ( Table 5 18) which suggest s that the user comments were appropriately constructed for this study. The final three groups received the stimuli as an artic le from a traditional wire service source, the Associated Press The article had an Associated Press banner at only the art icle itself, while the two final groups received either positive or negative reader comments at the bottom of the page.
74 A proposal of the present study and a new protocol submission form were Medical Institution al Review Board. The Institutional Review Board r eviewed and approved the study for testing on human subjects on March 1, 2012. After receiving IRB approval and running a pilot study to fix any potential problems with the study, t he data collection proce ss began April 2, 2012 and ended April 26, 2012. To recruit participants for this study, consenting professors sent an email to the students in their undergraduate classes with a hyperlink to the study. Participants could complete the study from any comp uter with Internet access. Before gaining access, participants were shown an informed consent screen with an option to agree or disagree to the study, and had to agree before proceeding. The experiment began with a pre level of ( Appendix C) Total fan identification was measured for this study by combining scale and su bjects to rate on a 1 7 bipolar scale (Shank & Beasley, 1998) Among tho se items were Boring/Exciting, Uninteresting/Interesting, Valuable/Worthless, Unappealing/Appealing, Useful/Useless, Needed/Not Needed, Irrelevant/Relevant, Unimportant/Important. Three questions were reverse coded to ensure that participants read the que stions carefully, and those that did not were removed from the data set.
75 The team identification scale consisted of seven item s that addressed how participants feel abo ut The questions asked particip ants to rate their feelings on a 1 8 bipolar scale study on sports fans and level of identification with their team Althoug h the present study used primarily 7 point scale s point scale to maintain high reliability for the ir established scale. The questions were also designed to look nearly identical to questions from t he other scales. strongly do your friends see you as a fan of the UF do you follow the UF basketball team via any of the following: a) in person or on television, b) on the radio, c) t Do not dislike/Dislike very Never/Always. The team identification scale was also shown to be very reliable for the For the purposes of this study, t hese two s cales were combined to define participants categorized as highly identified fan s or fan s with low identification The average score of bot h the sports involvement scale and the team identification scale
76 were summed and a split of the summed means determined placement into a fan identification group for the present study. The mean of the combined score was determined ( M = 9.98, SD = 2.93) an d participants with a total above and below that score were placed in the appropriate group. A one way analysis of variance showed that there was a significant difference in mean scores between the two fan identification groups ( F (1, 374) = 763.701, p < 001). Nearly all the participants were at least minimally identified as sports fans, so identification level was defined as eit her high or low in this study. Although not used to determine fan identification, behavioral fan involvement was also assessed u sing four open watching sports week, about how many hours do you spend reading sports open ended responses were used to validate assumptions that highly identified fan s are more invested i n sports and thus attend more sporting events and use more sports media than people who do not identify themselves as fans (Shank & Beasley, 1998). The pre test also included two questions about information seeking behavior of both online and offline sport list of responses was provided to the participant along with a in response box for a participant to list
77 any options that did not appear on the list of possible choices. The pre test concluded with a set of three demographic questions about participant age, gender, and field of study. After participants completed the pre test questionnaire, they were randomly assigned a sports article in one of the tw elve groups listed earlier ( Table 4 1). The online exp eriment was programmed to distribute the stimuli evenly. Before seeing the stimuli, participants received a set of on screen instructions to read everything that appeared on the following screen, and that they will then be asked to answer questions regard ing what they just read. The stimuli had a large banner of the online or traditional news source at the top, followed by the title of the article. Beneath the title of the article again. The sports article, which was the same for every participant, appeared underneath the news participant also saw user comments (either positive or negative) at the botto m of the page. After reading the article, the final aspect of the experiment was the post test questionnaire. Certain aspects of the post test varied based on the stimulus that the participant received. The first scale in the questionnaire measured a par attitude toward the website that the article was said to originate from or the Associated Press (O'Cass & Carlson, 2010). Participants were asked to rate how they feel about ESPN.com, AlligatorArmy.com, Facebook or the Associated Press on a 1 7 Likert
78 with my decision to use the Associated Press te scale was shown to be extremely For the participants that received the stimuli with user comments, the next scale comments (Walther et al., 2010; Postmes et al., 2001). Participants were asked to identify their feelings about f ive statements on a 1 7 bipolar particip ants that were randomly selected for the control groups were not shown this set of statements. This scale was also shown to be very reliable for the present study The next set of post test questions was developed from a scale that m easures the perceived credibility of the article (Bucy, 2003). Participants were asked nine questions about how they feel about the article itself and provided two contradictory items along a 7 point bipolar scale. The credibility questions and respons
79 at all in depth/Very in depth. Additionally, the credibility scale was shown to be highly reliable for this experim Because the entire sample population was made up of University of Florida undergraduate students, the final scale of the experiment was designed to measure the shown ten statements and asked to mark their opinions on a 7 point Likert scale ements used for this scale media criticized UF, I would feel
80 scale was reverse coded before data analysis began because it contradicts the other statements on this scale. This university .896). After answering all the question s on the post test, the participants were shown a post experiment debrief. The debrief provided a brief explanation of the purpose of the experiment, and explained that participants were randomly assigned to one of several conditions. Additionally, the d ebrief explained that while all participants read the same article, the source of that article and the comments following the article were manipulated for the purposes of this study. Finally, the participants were thanked for their time and asked to not t o talk to others about the study while responses were still being collected. The findings and the implications of this study are illuminated in the next two chapters.
81 CHAPTER 5 RESULTS Sample Description The present study had a total of 376 valid partici pants, all of whom were undergraduate students at the University of Florida. The mean age was 20.45 years old and the large majority of participants were female ( N = 255) ( Table 5 1). The ournalism and Communications, so the majors that appeared most frequently in the sample population were majors in the college: Telecommunications ( N = 86), Journalism ( N = 74), and Advertising ( N = 58). The only major outside the college that made up at l east ten percent of the sample population was Business/Finance ( N = 44). Table 5 1. Profile of Sample Population Gender Total( N = 376) Percentage Male Female 121 255 32.2 67.8 Age Average Range Standard Deviation 20.45 17 33 1.757 Field of Study Telecommunications Journalism Advertising Business/Finance Political Science Engineering English Sport Management Natural Science Public Relations Other 86 74 58 44 16 14 11 11 9 8 45 22.9 19.7 15.4 11.7 4.3 3.7 2.9 2.9 2.4 2.1 12.0 Fan Identifi cation Level Low Identification High Identification 174 202 46.3 53.7
82 On the pre test questionnaire, participants were asked to indicate all the sources that they use for seeking information about sports. The participants were shown to be very reliant on the Internet for sports information ( N = 266). Other than online, television is another source that more than half of the participants indicated using with sports TV highlight shows (such as Sportscenter on ESPN) and sports TV talk shows as their prim a ry sources of information ( Table 5 2). Newspapers were also shown to be used by a large portion of the sample population ( N = 135). Table 5 2. Offline and Online Sources for Information about Sports Source Total ( N = 376) Percentage* Seeking Sports Inf ormation Online Sports TV Highlight Shows Newspapers Sports TV Talk Shows Printed Sports Magazines Friends and Family Other None 266 207 135 110 43 18 8 49 70.7 55.1 35.9 23.9 11.4 4.8 2.1 13.0 Seeking Online Sports Information Sports Dedicated Webs ites Facebook Twitter Online Newspaper Websites (such as nytimes.com) Sports Blogs Online Sport Magazine Websites Other None 239 178 137 114 69 61 9 43 63.6 47.3 36.4 30.3 18.4 16.2 2.4 11.4 *Note: Percentage totals exceed 100 because participants were asked t o mark all applicable responses In a separate question participants were asked to specify all the online sources that they s eek for sports information ( Table 5 2). Sports dedicated websites, such as ESPN.com and yahoosports.com, are used by almos t two thirds of the sample population ( N = 239), which was the most common online source by a wide margin.
83 Social media websites Facebook ( N = 178) and Twitter ( N = 137) were the second and third most popular online sources for sports information respectiv ely. Interestingly, online newspaper websites ( N = 114), such as nytimes.com, are more popular among the participants than sports blogs ( N = 69). Although neither source is exceedingly popular among the respondents, online sport magazine websites ( N = 61 ) were used by more people than printed sports magazines ( N = 43). For the purposes of this study, 53.7% of the participants were classified as highly identified sports fans ( N = 202), while 46.3% were classified as fans with low identification ( N = 174). team identification scale were summed to determine a baseline score for how strongly a participant identifies him or her self as a sports fan and as a fan of a University of Florida sports team. a were used to measure fan identification in the present study. The beha study were only used to illustrate differences in involvement between the two identification groups. The mean of the combined score was determined ( M = 9.98, SD = 2.93) and participants with a t otal above and below that score were place d in the appropriate group ( Table 5 3). A one way analysis of variance showed that there was a significant difference between the two fan identification groups between their mean score s on the combined scales ( F ( 1 374) = 763.701 p < .001). Participants were also asked about their behavioral involvement as a sports fan. These questions pertained to the hours per week a participant spends watching sports
84 related programming either on television or on the Internet and the hours per week spent reading sports related periodicals. There was a significant difference for both responses for highly identified fan s and fans with low identification ( Table 5 4). Highly identified fan s ( M = 5.56, SD = 6.79) spent more than f our times the amount of time watching sports related programming than fans with low identification ( M = 1.32, SD = 2.31). Highly identified fan s also spent more hours per week ( M = 2.40, SD = 4.38) reading sports related periodicals than fans with low ide ntification ( M = .76, SD = 3.30). Table 5 3. Analysis of Variance for Fans with High and Low Identification on the Fan Identification Scale Identification Level N Mean SD F d f p value Low Identification 174 7.39 2.00 763.701 374 .000* High Identification 202 12.21 1.36 *p value < .001 Table 5 4. Analysis of Variance of Sports Behavioral Involvement for Fans with High and Low Identification Behavioral Involvement Category Mean SD Std. Error F d f p value Hours Watching Sports Related Programming Low I dentification High Identification 1.32 5.56 2.31 6.79 .176 .478 61.80 374 .000* Hours Reading Sports Related Periodicals Low Identification High Identification .76 2.40 3.30 4.38 .250 .308 16.27 374 .000* Professional Sporting Events Attended Low Identification High Identification 1.78 4.36 2.83 6.58 .214 .463 23.11 3 74 .000* College Sporting Events Attended Low Identification High Identification 5.05 11.46 5.72 10.35 .433 .728 52.88 374 .000* p value < .001
85 Y early game attendance was also measured in this experiment and there was a significant difference for both professional ( F (1, 374) = 23.11, p < .0 0 1) and college sporting events attended ( F (1, 374) = 52.88, p < .0 0 1), based on identification level. Hig hly identified fan s ( M = 4.36, SD = 6.58) attend nearly three times more professional sporting events than fans with low identification ( M = 1.78, SD = 2.83). Because the sample population is solely made up of college students, it was also appropriate to determine the number of college sporting events the two groups attend per year. Highly identified fan s ( M = 11.46, SD = 10.35) attend two times more live college sporting events than fans with low identification ( M = 5.05, SD = 5.72). Ultimately, the sig nificant differences in average scores on the identification scales and on behavioral fan involvement illuminate the stark differences between the two identification groups, and give credence to their use for data analysis purposes in the present study. Re search Questions and Hypotheses Research Question 1 RQ1 sought to determine which medium is p erceived as more credible: wire service or online. Table 5 5 illustrates the mean cre dibility scores for both media In total, the online group had approximately three times more participants assigned to that stimuli, but the cre dibility mean was still slightly lower for the online medium ( M = 4.59, SD = 4.62) than for the wire service ( M = 4.62, SD = .98). A univariate analysis of variance was conducted to dete ct significant differences between means. However, no significant differences were di scovered between the media ( F ( 1, 374) = .091 p = .763). Table 5 5. Univariate Analysis of Variance of Credibility Scores for the Online Medium and Wire Service Medium N M ean SD F d f p value Online 281 4.59 1.02 .09 1 374 .763 Wire Service 95 4.62 .98
86 Research Question 2 RQ2 also examined medium credibility for both the online medium and wire service but looked specifically at highly identified sports fans. This res earch question sough t to determine which medium that highly identified fans found more credible. In contrast to RQ1, the mean scores for the online medium ( M = 4.94, SD = .96) were slightly higher than the wire service ( M = 4.85, SD = .90). Table 5 6 ill ustrates the differences in means for the credibility scale as well as the results of a univariate analysis of variance which did not suggest a significant di fference between the two media ( F ( 1, 200) = .360 p = .549). Table 5 6. Univariate Analysis of Va riance of Credibility Scores of Medium for Highly I d entified F an s Medium N Mean SD F d f p value Online 148 4.94 .96 .36 0 200 .549 Wire Service 54 4.85 .90 Research Question 3 Similar to the previous research question, RQ3 examined whether fans wit h low identification judge the online medium or the wire service to be more credible. Unlike highly identified sports fans, the mean scores for fans with low identification were fo und to be slightly higher for the wire service ( M = 4.32, SD = 1.01) than o nline ( M = 4.19, SD = .94). Table 5 7 demonstrates the differences in credibility scores for both the online medium and wire service as assessed by fans with low identification Additionally, a univariate analysis of variance was run to determine if a di fference in the means exists. However, there was no significant difference between the two media ( F (1, 172) = .583 p = .446).
87 Table 5 7. Univariate Analysis of Variance of Credibility Scores of Medium for Fans with Low I dentification Medium N Mean SD F d f p value Online 133 4.19 .94 .58 3 172 .446 Wire Service 41 4.32 1.01 Research Question 4 RQ4 sought to determine which of the three online sources (ESPN, sports blog, and Facebook ) were perceived to be the most credible. Table 5 8 demonstrates t he differences in means of the online sources and the results of an analysis of variance to determine if there is a significant difference among means ( F (2, 278) = 2.04, p = .131). No significant differences wer e found among the three sources; however ES PN.com was foun d to have the highest mean score on the scale ( M = 4.71, SD = .97). The sports blog, alligatorarmy.co m, had the second highest mean score ( M = 4.63, SD = 1.03), and Facebook was ra ted lowest by the participants ( M = 4.42, SD = 1.04). Table 5 8. Analysis of Variance of Credibility Scores for Online Sources Online Source N Mean SD F d f p value ESPN 92 4.71 .97 2.04 278 .131 Sports Blog 94 4.63 1.03 Facebook 95 4.42 1.04 Hypothesis 1 H1 stated that the mainstream online spor ts arti cle from ESPN.com would be perceived as more credible than the other online sources by both highly identified sports fans and fans with low identification This hypothesis tested the main effects of online source and fan identification level on the credib ility of the article. Table 5 9 illustrates the differences in credibility scores for the three sources based on identification level. Fans with low identification rated the ESPN article ( M = 4.50, SD = .99) as more credible than the sports blog ( M = 4.1 2, SD = .95) and F acebook ( M = 3.99, SD = .83).
88 However, the sports blog ( M = 5.04, SD = .92) had the highest mean score from the highly identified fans followed by ESPN ( M = 4.89, SD = .92) and Facebook ( M = 4.88, SD = 1.06), which were almost even. Tw o univariate ANOVAs w ere conducted to determine if there w as a significant difference among the online sources. As shown in Table 5 9 no significant difference among the online sources was shown for the highly identified fan s ( F ( 2, 145 ) = .44 0 p = 648 ) so for that identification group, H1 was not supported However, a significant difference among the online sources was shown for the fans with low identification ( F (2, 130) = 3.66, p = .028), who rated the ESPN article significantly higher than the oth er two online sources, thus giving partial support to H 1. Table 5 9. Univariate Analysi s of Variance for Online Source Credibility Based on Identification Level Online Source N Mean SD F d f p value Low Identification ESPN 42 4.50 .99 3.66 130 .028 Sp orts Blog 42 4.12 .95 Facebook 49 3.99 .83 High Identification ESPN 50 4.89 .92 .44 0 145 .648 Sports Blog 52 5.04 .92 Facebook 46 4.88 1.06 Hypothesis 2 H2 stated that highly identified sports fans would rate the online website s th emselves more favorably than fans with low identification in the study. This hypothesis was intended to test the main effects of fan identification level on the perce ptions of a website. Table 5 10 demonstrates the differences in website evaluation s cores for the three online sources based on identification level. Even though the sports blog received the highest credibility score from highly identified fans both fan identification groups evaluated the sports blog, alligatorarmy.com, as the least
89 fav orable website. For h ighly identified fan s ESPN ( M = 5.28, SD = .90) had the highest mean score on the source evaluation scale followed closely by Facebook ( M = 5.12, SD = 1.12), while fans with low identification rated Facebook ( M = 4.41, SD = .94) hig her than ESPN ( M = 3.98, SD = 1.17). A univariate analysis of variance was conducted to test the hypothesis and H2 was su pported. Table 5 10 Univariate Analysis of Variance of Website Evaluation Score Based on Online Source and Identification Level Var iable N Mean SD F df p value Identification Level Low Identification ESPN 42 3.98 1.17 52.49 275 .000* Sports Blog 42 3.89 .88 Facebook 49 4.41 .94 High Identification ESPN 50 5.28 .90 Sports Blog 52 4.46 .76 Facebook 46 5.12 1.12 *p value < .001 As shown in Table 5 10 identification level had an impact on evaluations of a website in which every online source was rated higher by the highly identified fan s, and there was a significant difference in mean scores ( F (1, 2 75) = 52.49, p < .001). Essentially, highly identified fan s found the websites ESPN.com, AlligatorArmy.com, and Facebook .com to all be significantly more favorable websites than the fans with low identification Additionally, using a univariate ANOVA, a significant difference for the website scores was found between the online stimuli received by the participants ( F (2, 275) = 9.26, p < .001). Going further, there was also a significant difference in the interaction between the online stimuli received and identification level on the website evaluation scores ( F (2, 275) = 3.53, p = .031). Ultimately, the combination of both
90 independent variables (online source and fan identification level) caused signific ant differences in how participants rated the online sources. Hypothesis 3 H3 stated that highly identified fan s would view the sports article as more credible than fans with low identification T his hypothesis was intended to test the main effects of fan identification level on the perceived credibility o f the article in the twelv e stimuli groups. In Table 5 11 the credibility mean scores for ar ticle credibility are shown. Table 5 11 Means of Credibility Scores for Each Stimuli Group Separated by Identification Level Stimuli Group N Mean SD Low Identif ication ESPN Positive 14 4.79 .88 ESPN Negative 11 4.85 1.09 ESPN Control 17 4.04 .86 Blog Positive 12 4.26 1.20 Blog Negative 15 4.01 .86 Blog Control 15 4.11 .86 Facebook Positive 12 4.00 .93 Facebook Negative 20 3.96 .79 Facebook Control 17 4 .03 .85 AP Positive 13 4.63 .78 AP Negative 17 4.12 1.23 AP Control 11 4.26 .86 Low Identification Total 174 4.22 .95 High Identification ESPN Positive 16 5.26 .80 ESPN Negative 21 4.59 1.00 ESPN Control 13 4.94 .81 Blog Positive 17 5.12 1.0 9 Blog Negative 17 4.97 .94 Blog Control 18 5.04 .77 Facebook Positive 21 5.01 .91 Facebook Negative 12 4.86 .97 Facebook Control 13 4.70 1.39 AP Positive 18 4.96 1.09 AP Negative 16 4.92 .89 AP Control 20 4.70 .73 Low Identification Total 202 4.9 2 .95
91 Highly identified fan s ( M = 4.92, SD = .95) rated the sports article as more credible on average than fans with low identification ( M = 4.22, SD = .95). As shown in Table 5 12 a univariate analysis of variance was conducted to determine if a signi ficant difference in means was present for the two identification levels. T here was a significant difference in credibility scores ( F (1, 352) = 44.89, p < .001), so H3 was supported. Among the stimuli groups, the ESPN. com group with negative comments for fans with low identification ( M = 4.85, SD = 1.0 9) had the highest mean score while the Facebook group with negative comments wa s found to have the lowest mean score for fans with low identification ( M = 3.96, SD = .79). For highly identified fans, the ESPN.com positive grou p received the highest credibility score ( M = 5.26, SD = .80), and, interestingly, the ESPN.com negative g roup received the lowest credibility score from highly identified fans ( M = 4.59, SD = 1.00). Table 5 12 Univariate Analy sis of Variance for Credibility Score Based on Identification Level and Stimuli Group Variable High Mean SD Low Mean SD F d f p value Sports Source ESPN Sports Blog Facebook Associated Press 5.04 4.76 4.64 4.82 .86 1.20 1.03 .98 4.43 4.52 4.30 4.51 94 1.01 .96 1.13 1.68 352 .170 User Comment Tone Positive Negative Control 5.04 4.68 4.62 .86 1.02 .92 4.64 4.30 4.32 1.03 .96 1.15 2.81 352 .062 Identification Level High Identification Low Identification 5.26 4.79 .80 .88 4.59 3.96 1. 00 .79 44.89 352 .000* *p value < .001
92 Aside from fan identification level, a univariate analysis of variance of the credibility scores based o n the stimuli received. This did not reveal significant differences between the twelve assigned stimuli groups themse lves ( F (11, 364) = 1.36, p = .191). Additionally, a difference in credibility scores based on sports source received and user comment tone was also analyzed using a univariate ANOVA, but a significa nt difference was not found in either case (Table 5 12) Essentially, fan identification level was the only factor to have a significant influence on the perceived credibility of the sports article. Hypothesis 4 H4 stated that highly identified fan s would have a stronger group identity with their univers ity than fans with low identification This hypothesis tests the main effects of fan identification level on group identity with the Un iversity of Florida. Table 5 13 illustrates that fans with low identification ( M = 4.65, SD = 1.05) do not identify wit h their university nearly as much as highly identified fan s ( M = 5.60, SD = .77). A one way analysis of variance was conducted to determine if there was a significant difference in the mean scores for the university identification scale. A significant di fference was found between fans with low identification and highly identified fan s on this scale ( F (1, 374) = 104.11, p < .001), thus giving support to H4. Table 5 13 Analysis of Variance of University Identification Score Based on Identification Level Id entification Level N Mean SD F d f p value Low Identification 174 4.65 1.05 104.11 374 .000* High Identification 202 5.60 .77 *p value < .001
93 Hypothesis 5 H5 stated that highly identified fan s would find the article to be more credible when coupled wi th anonymous positive user comments than with negative user comments or without comme nts. This hypothesis tested the main effects of user comment tone on credibility for highly identified fan s. Table 5 14 shows the differences in credibility scores based on the tone of the user comments. Articles with the positive user comments garnere d the highest credibility mean score ( M = 5.08, SD = .97), followed by the control group which received no user comments ( M = 4.84, SD = .91), and the articles with negativ e user comments ( M = 4.82, SD = .94). A univariate analysis of v ariance was run, and found no significant difference between the means for credibility ( F (2, 299) = 1.61, p = .203), so H5 was not supported. Table 5 14 Univariate Analysis of Variance of Cr edibility Score for Highly Identified F an s Based on User Comment Tone User Comment Tone N Mean SD F df p value Positive 72 5.08 .97 1.61 199 .203 Negative 66 4.82 .94 Control 64 4.84 .91 Hypothesis 6 H6 asserted that the perceived credibility of the article would change more significantly by the tone of the user comments for highly identified fan s than for fans with low identification Thus, the hypothesis tested the main effects of user comment tone on perceived credibility an d compared the sig nificant differences between the two identification levels. In H5, the impact of user comment tone on credibility for highly identified fan s is outlined in Table 5 14. Conversely, Table 5 15 illustrates the differences in credibility scores based on the user comment tone for fans with low identification For this identification level, positive user comments also garnered the
94 highest credibility mean score from participants ( M = 4.44, SD = .98). However, articles with negative user comments ( M = 4.17, SD = 1.02) received slightly higher credibility scores than articles without user comments ( M = 4.09, SD = .84). A univariate analysis of variance was conducted to determine if there was a significant difference in means for credibility ratings, but no sign ificant difference was found ( F (2, 171) = 1.95, p = .145). Additionally, t he low identification fan grou p had a higher F statistic than highly identified fan s, which does not give any support to H6. However, credibility was not signi ficantly impacted by user comment tone f or either identification level. Table 5 15 Univariate Analysis of Variance of Credibility Score for Fans with L ow I dentification Based on User Comment Tone User Comment Tone N Mean SD F d f p value Positive 51 4.44 .98 1.95 171 .145 Ne gative 63 4.17 1.02 Control 60 4.09 .84 Hypothesis 7 H7 stated that highly identified fan s would have stronger feelings either in favor or against the users than fans with low identification based on the tone of the comments. The hypothesis test ed the main effects of user comment tone on group ident ity with the users and compared the significant differences between the two id entification levels. Table 5 16 shows the differences in means for both fan identification groups on the user group identi fication scale based on the tone of the comments, as well as the results of two ANOVAs. Fans with low identification ( M = 3.04, SD = 1.49) did not identify as closely with the users who posted positive user comments than the highly identified fan s ( M = 4. 03, SD = 1.60). For the users that posted negative comments, neither highly identified fan s ( M = 2.25, SD = 1.41) nor fans with low identification ( M = 2.14, SD = 1.30) linked strongly with the users. Based on the results of two univariate analyses of
95 va riance, highly identified fan s ( F (1, 136) = 47.94, p < .001) were shown to have a greater difference in means between the positive and negative comments and a highe r F statistic than the low fan identification group ( F (1, 112) = 11.92, p = .001), which giv es support for H7. Table 5 16 Univariate Analyse s of Variance of User Identi fication Score Based on User Comment Tone User Comment Tone** N Mean S D F d f p value Low Identification Positive 51 3.04 1.49 11.92 112 .001 Negative 63 2.14 1.30 High Identification Positive 72 4.03 1.60 47.94 136 .000* Negative 66 2.25 1.41 *p value < .001 **Note: Participants in the control groups were not given this scale in the post test As shown in Table 5 17, another univariate analysis of variance was conducted to examine the potential impact of the three independent variables on group identification scores with the anonymous users. The analysis showed that both fan identification level ( F (1, 236) = 9.07, p = .003) and the tone of the user comments ( F (2, 236) = 49.45, p < .001) had a significant influence on user identification scores. Essentially, there was a significantly higher difference in how highly identified fan s reported group identification with the anonymous users than the fans with low identification The tone of the user comments also significantly impacted participant identification with the anonymous users, with positive comments garnering a significantly higher score on the user identification scale than negative comments. Addition ally, the interaction between identification level and user comment tone also garnered a statistically significant difference in user identification scores ( F (1, 236) = 5.26, p = .023). The combination of these two variables had a significant influence on how participants rated their group
96 identification level with the users. For example, highly identified fans who received the article with positive comments had significantly higher user identification scores than fans with low identification who received the article with negative comments. However, the sports source received did not cause a significant difference in participant identification ratings of the anonymous users ( F (2, 352) = 2.43, p = .090). Table 5 17 Univariate Analysi s of Variance of Us er Identification Score Based on Fan Identification Level and Stimuli Received Variable High Mean SD Low Mean SD F d f p value Identification Level 9.07 236 .003 Low I dentification High Identification 3.29 4.35 1.40 1.69 2.05 2.11 1.31 1.66 Sports S ource .359 236 .783 ESPN Sports Blog Facebook Associated Press 3.58 3.89 3.35 3.68 1.69 1.78 1.49 1.58 2.09 2.13 2.23 2.31 1.53 1.34 1.40 1.19 User Comment Tone Positive Negative 3.89 2.30 1.78 1.19 3.35 2.09 1.49 1.53 49.45 236 .000* Identification Level X User Comment Tone Interaction 3.59 .271** 2.08 .276** 5.26 236 .023 *p value < .001 **Note: Standard Error Hypothesis 8 H8 stated that highly identified fan s would have stronger positive or negative feelings about the media sourc e itself (website or AP article) than fans with low identification based on the tone of the user comments. Ther efore, the hypothesis tested the main effects of user comment tone on source evaluation score and compared the significant differences between t he two id entification levels. Table 5 18 shows the differences in means for both identification groups on the website evaluation scale
97 based on the tone of the comments, and the results of two ANOVAs. For fans with low identification positive user comme nts ( M = 4.24, SD = 1.02) garnered the highest evaluation score of the website, followed by negative user comments ( M = 4.05, SD = 1.34), and no comments trailed both scenarios ( M = 3.93, SD = 1.02). Similarly, highly identified fan s rated the website hig hest when presented with positive user comments ( M = 4.93, SD = .92), which was followed by negative user comments ( M = 4.68, SD = .94) and no comments ( M = 4.63, SD = 1.06). Neither identification level had a significant difference in means on the websit e evaluation scale, which does not support H8. The ANOVA for highly identified fan s ( F (2, 199) = 1.93, p = .148) did result in a higher F statistic than fans with low identification ( F (2, 171) = .95, p = .388); however the results were not significant en ough to fully support the hypothesis. Table 5 18 Univariate Analysis of Variance of Source Evaluation Score for Fans with Low and High Identification Based on Tone of User Comments User Comment Tone N Mean SD F d f p value Low Identification Positive 51 4.24 1.02 .95 171 .388 Negative 63 4.05 1.34 Control 60 3.93 1.02 High Identification Positive 72 4.93 .92 1.93 199 .148 Negative 66 4.68 .94 Control 64 4.63 1.06 Additionally, Table 5 19 outlines a univariate analysis of va riance that was conducted to examine the potential influence of all three i ndependent variables on the evaluation scores of a sports news source. A significant difference in the s ource evaluation scores was shown between the two identification groups ( F (1 352) = 45.25 p < .001) as well as the sports source received ( F (3 352) = 10.95 p < .001). The interaction between identification level and sports source also garnered a statistically
98 significant difference in sports source evaluation scores ( F (1, 352 ) = 3.36, p = .019). Ultimately, the combination of the two factors influenced participant evaluations of the sports sources. For example, highly identified fans who received the ESPN.com or Facebook source, which were the two highest rated sources, had significantly higher evaluations than fans with low identification who received the Associated Press source, which was the lowest rated sports source. User comment tone did not, however, cause a significant difference in source evaluation scores for the p articipants ( F (2, 352) = 2.43, p = .090). Table 5 19 Univariate Analysis of Variance of Source Evaluation Score Based on Fan Identification Level and Stimuli Received Variable High Mean SD Low Mean SD F df p value Identification Level Low Identification High Identification 4.54 5.45 .98 .94 3.51 3.95 1.12 .90 45.25 352 .000* Sports Source 10.95 352 .000* ESPN Sports Blog Facebook Associated Press 4.87 4.39 4.97 4.29 1.13 .79 1.07 .91 4.33 4.03 4.42 3.79 1.19 .91 1.22 .99 User Comm ent Tone Positive Negative Control 4.97 4.85 4.87 1.07 1.28 1.09 4.29 4.03 3.79 .91 .91 .99 2.43 352 .090 Identification Level X Sports Source Interaction 5.30 .144** 3.89 .159** 3.36 352 .019 *p value < .001 **Note: Standard Error Gender A lthough the present study was not focused on gender, the large majority of participants were female (67.8%), so data analyse s that examined potential differences
99 b etween the two genders w ere conducted As outlined in Table 5 20 five univariate ANOVAs wer e conducted to determine if a significant difference existed between male and female participants for any of the variables being studied. The scores on the fan identification scale were not significantly different between the genders ( F (1, 374) = 1.74, p = .190). had a significant difference based on gender ( F (1, 374) = 14.85, p < .001). Males ( M = 4.88, SD = 1.01) rated the article as significantly more credible than females ( M = 4.4 6, SD = .98). However, no significant differences were found between the ratings of the sports sources ( F (1, 374) = 2.39, p = .123), participant identification with the University of Florida ( F (1, 374) = 2.38, p = .124), and group identification with the anonymous users ( F (1, 250) = .82, p = .367). Table 5 20 Univariate Analyses of Variance for the Participant Scales Based on Gender Participant Scale N Mean SD F d f p value Fan Identification 1.72 374 .190 Male 121 10.27 3.10 Female 255 9.84 2.8 5 Credibility 14.85 374 .000* Male 121 4.88 1.01 Female 255 4.46 .98 Source Evaluation 2.39 374 .123 Male 121 4.56 1.09 Female 255 4.37 1.12 University Identification 2.38 374 .124 Male 121 5.04 1.17 Female 255 5.22 .95 User Identification .82 250 .367 Male 82 3.02 1.71 Female 170 2.82 1.62 *p value < .001
100 CHAPTER 6 D ISCUSSION T he global popularity of sport continues to reach new he ights as online newspapers and sports ded icated websites allow fans 24 hour acc ess to their favorite sports and teams making sport media consumerism a multi billion dollar industry (Raney & Bryant, 2006) This instant access to statistics and scores may not, however, have impacted credibility i n a positive way Some websites often lack proper editorial review, analysis of content, and factual verification (Chung, Kim & Kim 2010), and t he pressure to post articles online immediately after a n event may substitute speed for quality (Gunter et al. 2009; Bucy, 2003) Because the landscape of sports journalism is constantly changing, t he present study was designed to examine how medium, article source, fan identification, and user comment tone can all influence the credib ility of the sports article An online experiment was developed, and p articipants ( N = 376) were randomly assigne d a sports article in one of twelve stimuli groups. The article source was indicated to have appeared on a mainstream sports we bsi te, a sports blog a social network ing site, or a wire service Participants also received the stimuli with either positive user comments, negative user comments, or without comments. Prior to receiving the stimuli, participants completed a pre test w ith questions regarding fan and team identification, sports behavioral involvement and demographic information Participants also completed a post test with questions designed to measure article credibility, evaluations of a sports source, group identifi cation with anonymous users, and identification with their university.
101 Overall, fan identification level was shown to be the most important factor in the ratings of perceived credibility for sports articles There was no statistical difference in the cred ibility scores between the wire service and online medium and both were rated as only slightly credible. Highly identified fan s found the sports article to be significantly more credible than the fans with low identification as a whole and nearly every s timuli group. Highly identified fan s also evaluated all three websites significantly more favorably than fans with low identification There was a clear r atings disparity for the online sources as highly identified fan s rated all the websites favorably a nd fans with low identification rated them all slightly negatively, except for Facebook For both fan identification groups Facebook was the only website that received a positive score from both groups while blogs were rated as the least favorable onlin e source by both groups Additionally, highly identified fan s and fans with low identification had a significantly higher group identity with users who posted positive comments than users who posted negative comments. However, u ser comment tone did not i nfluence credibility ratings or evaluation scores of the sports news sources. Each finding will be addressed individually as well as addressed in the larger conte xt of the entire study, sports journalism, and media research. The findings will also be com pared to p revious research on credibility, social identity theory, fan identification, and the SIDE model and juxtaposed with comparisons to the current state of news journalism and sports fan culture. Findings The Internet news cycle has been continually ev olving with advancements in technology and the credibility of news, particularly online news, has recently come into question (Gunter et al., 2009) New (and often unr eliable) websites sprout up
102 frequently and people are, perhaps, more wary of news t han at any time in history (Gunter et al., 2009; Pew Research Center, 2010) Online journalism is often subject to the same credibility standards of traditional media, but the round the clock nature of the Internet has altered the news landscape and the w ay it is reported (Arant & Anderson, 2001). Because news credibility has taken a hit with the public there is reason to believe that this phenomenon has impacted perceptions of media coverage outside of traditional news. This lack of credibility for new s may have also transl ated to the sports world and the present study was designed to examine the factors that may contribute to the perceived credibility of sports article s as well as the factors that can impact attitudes toward a particular news or sport s source. Some scholars argue that online news stories may remain at the same high quality of traditional news, however, the percept ions of credibility have changed with new forms of sourcing and storytelling methods available to online journalists (Choi e t al., 2006). Sports journalism, both online and offline, is built on the sa me principles as traditional journalism and oftentimes can be subject to the same ethical guidelines and standards (Oates & Pauly, 2007) Because a number of factors can impact o sports article, the present study looked at how article source, medium, fan identification, and user comment tone can all impact the credibility of the sports article or a ne ws source itself. As evidenced in earlier studies (Hu & Tang, 2010; Fisher & Wakefield, 1998; consumption habits and his or her perception of sports and media can be highly dependent on fan identification, or the personal commitment or emotional involvement a person has with a sport or sport org anization. Fan identification has been
103 shown to predict fan consumption behavior through attending live sporting events and sport media usage (Milne & McDonald, 1999). Highly identified fans have also been shown to be more likely to watch games in person or through media (Laverie & Arnett, 2000), pay more for sporting event tickets, and stay loyal to a poorly performing team (Fisher & Wakefield, 1998). In the present study, f an identification level was shown to have a significant impact on behavioral fan involvement. Highly identified fan s spent more than four times the amount of time watching sports related programming per week than fans with low identification as well as spent approximately three times as many hours per week reading sports related per iodicals. Additionally, highly identified fan s attend nearly three times more professional sporting events than fans with low identification and attend two times as ma ny college sporting events This discrepancy in sport s consumption helps to illuminate the difference between the two groups and is consistent with the previous research. Highly identified fan s are likely more interested in sports and their favorite teams, so they are willing to devote more time and spend more money on sports. Intere stingl y, this identification distinction not only showed a profound difference in sports behavioral involvement, but also had as large of an impact on the way that participants viewed the credibility of the article. Because article medium has been shown to impac t credibility in the past (Graziano & McGrath, 1986; Sundar & Nass, 2001; Bucy, 2003 ), the present study sought to determine how the medium may impact the credibility of a sports article Looking specifically at wire service and online media there was no significant difference in cred ibility scores between the two medi a as both were seen as slightly credible. This lack of significance contradicts on medium credibility where print
104 was seen as significantly more credible than other type s of news media, including online and television although none of the media were seen as overly credible When looking at credibility in the two identification groups in the present study highly identified fan s were minimally affected by medium whe n judging the credibility of the article. Similarly, there was no significant difference in the credibility scores between the two media for fans with low identification T he results of the present study may signify a distinction between sports and news c overage, and a difference in the modern interests of sports and news media consumers. In this study, t he credibility of the online medium for sports media consumers may now rival traditional media based on the dedication of established news organizations to provide user s with up to the minute sports updates. People have become accustomed to cell phone and online sports access literally at thei r fingertips. Traditional media such as a wire service, may have maintained a level of public trust over the yea rs with fact checking procedures that may not exist for less established websites, but there is an aspect of instant gratification that is severely lacking in traditional media T he speed with which the Internet can provide information may also have been offset by the amount of false information that can spread quickly throughout the Internet. Recently, NBC had come under fire for broadcasting the major competitions at the 2012 London Olympics on tape delay (Levy, 2012) T he absence of live coverage for caused a social media uproar and the Twitter hashtag #NBCFail from people who were getting the competition results ruined for them online (Levy, 2012). Although the televisi on medium was not tested in the present study this illuminates the instant gr atification that people desire from the
105 online medium that has been noticeably absent from all other media. Even NBC, which had the capability to offer viewers live coverage of every Olympic event, chose to broadcast many events in primetime to garner a larger audience and higher advertising revenue. Traditional media is at an even larger disadvantage because there is an even longer delay for sports articles and event results, and this does not resonate with readers who wa nt immediate up dates Essentially, neither medium was seen as exceedingly credible but the seemingly limitless nature of the Internet for broadcasting and reporting sports is an issue for sports journalists and news organizations to consid er when attempt ing to raise the credibility of their brand. Despite research to the contrary (Gunter, 2006 ; Metzger, Flanagin & Medders, 2010; Chung, Kim & Kim 2010) the credibility ratings of the three online sources were not significantly different which reflects th medium credibility Among the three online sources, the article appearing on the ESPN website was seen as slightly more credible than on the sports blog and even more so for Facebook ESPN.com may have received sl ightly higher ratings on the credibility scores because it is an established sports news website and is a mainstay in the culture of sports media (Gunter et al., 2009) T he overall similarity in credibility scores among the three sources may have occurred because the participants in the study all received the same article, which was originally published by the Associated Press Essentially, t he article covered a real life sporting event that may have been fresh in the and was factually accurate, so it seems logical that participants would rate the article as moderately credible.
106 The article also covered a sporting event in which the identified team lost. Wann and McGeorge (1994) found that highly identified fans rep orted a greater increase in positive emotions after a win and a greater increase in negative emotions after a loss than minimally identified fans. Highly identified fans have also been shown to enhance their well being after a win and protect it after a l oss by using biased descriptions about the game outcome (Wann et al. 2002) While the scales used to determine fan identification level appeared before the article in the experiment, the outcome of the game may have impacted participant responses to the c redibility and sports source scales. In the present study, both fan identification groups may have been trying to placate the negative emotions of a team loss by rating the article as less credible than they would a team win, and th is scenario may have ha d a more profound impact on the highly identified fans. Although game outcome and the disparity between wins and losses for fans have emotional state (Wann & McGeorge, 1994; Wann et al., 2002), highly identified fans may be come so accustomed to reading both positive and negative game outcome s throughout a given season that the effect on perceptions of article credibility is minimal. Sports blogs and team dedicated websites may use biased descriptions to soften the impact of a loss in their online articles and this may positively impact the perceived credibility of the article to highly identified fans. The present study also examined the impact that branding can have on credibility assessments for each identification level. Based on the previous literature on news branding (Ranie et al 2003; Gunter et al., 2009) it was hypothesized that both highly identified fan s and fans with low identification would rate the ESPN article as more
107 credible than the same article on the s ports blog or on Facebook because ESPN has an established reputation in sports journalism that the other two do not T here was no significant difference in the credibility scores for highly identified fan s although the credibility means were slightly hig her on the blog than on ESPN or Facebook However, fans with low identification who also reported reading significantly fewer sports periodicals than highly identified fan s, rated the ESPN article as significantly mo re credible than the other two sources providing some validation for the hypothesis Gunter (2006) found that established offline news brands that migrate online command more public trust than more contemporary online only brands. This m ay have occurred because even fans with low identificat ion are at least somewhat familiar with ESPN as a sports news source, but are not as familiar with AlligatorArmy.com as a sports blog or Facebook as more than simply a social networking website for connecting with peers. Highly identified fan s, however, a re seemingly more aware of sports blogs and are more likely to perceive an article on a sports blog as credible. Modern blogging is unlike conventional reporting in that it incorporates elements of speculation and first person narration that is absent in mainstream news (Robinson, 2006). These aspects may be what drive highly identified sports fans to read sports blogs over established sports news brands. Attitudes towards the websites themselves were also examined, and it was hypothesized that highly ide ntified fan s would rate the online websites more favorably than fans with low identification Using scale, highly identified fan s rated all three websites significantly higher than fans with low identificatio n and there was a clear disparity in the ratings as highly identified fan s
108 rated all the online sources favorably and low identification fans rated all the websites slightly negatively except for Facebook Fans with low identification rate d Facebook as t heir favorite website of the three options, but the sports articles appearing on Faceb ook were rated as the least credible Social networking websites serve many p urposes and, as evidenced by these results, are popular among college students. However, th e popularity of Facebook does not seem to translate to sports article credibility, especially for fans with low identification in this study Additionally, t he biggest disparity in the evaluation scores between the two identification groups was their asses sments of the ESPN.com we bsite. Highly identified fan s rated the website 1.3 points higher (out of 7) than fans with low identification and rated it as the most favorable website among the three. These findings are not shocking because highly identified fan s are much more behaviorally involved and are more avid sports consumers, so they should be more likely to rate a sports dedicate d website more favorably than fans with low identification ESPN also established itself as a sports brand on television be fore making an online transition, so it has appeal for sports fans across media. It is somewhat surprising that all three websites were rated more favorably by the highly identified fan s, even though Facebook is generally unaffiliated with sports. This c ould have occurred because the website assessment scale appeared after reading the article, and fans with low identification may have evaluated the website more negatively after reading an article that did not interest them. Although not tested, highly i dentified fan s may also be more likely to have friends who are highly identified fan s. Friends who share this common interest may use Twitter or Facebook to share sports information with each other and links to other
109 websites. Social identity theory asse rts that memberships in important social groups positive self concept for that individual (Tajfel, 1981). The major distinction between social networking websites and ot her websites is that the information is being shared through friends, so there may be heightened level of value to the information and a more favorabl e attitude toward Facebook Highly identified fan s may also be more comfortable using all aspects of onli ne media because sports have become reliant on the Internet. Highly identified sports fans have probably used at least one, if not all, of these websites to look up gam e scores or gain insight about a team. Advancements in technology have also given Inte rnet users more and quicker access to information. Sports fans, particularly, have reaped the benefits of recent technological Olympic G ames in London, NBC created two smartphone apps that allowed for more than 3,500 hours of live streaming Olympic content and in depth details about athletes and the Games (Albanesius, 2012). Attendance at live sporting events is even being threatened by n ew technologies. The commissioner of the National Football League instituted an initiative to make Wi Fi improving quality of view ing football at home with a high definition television o r laptop (Davis, 2012). Because sports fans are becoming more exposed to new media and more familiar with new technology, there is greater potential for m ore positive feelings about all websites. Interestingly the website evaluation scores were also si gnificantly different for the three online sources regardless of identification level Both fans with low identification
110 and highly identified fan s rated the blogs as the least favorable online source, and Facebook was the only website that received a po sitive score from both identification groups. There was also a significant difference in the interaction between online source and identification level, so both factors were shown to contribute to how an individual evaluates a website. T hese findings ul timately, suggest that there is an empirical link between both fan identification and online sources themselves, when one is evaluating satisfaction with a website. Wann (2006) found that team identification and social psychological health are positively related because fan identification leads to social connections which can enhance well being. Wann and Weaver (2009) also found that identification with a local team was positively related to social well being, but not for fans of distant teams. This soci al psychological phenomenon may also be at play in the present study, where participants are all locally connected to the team being studied, whether they are a fan of the team or not. If the highly identified fan s are happier after reading an article abo ut their team than low identification fans then this may have manifested in the way the two groups rated the website. Although the research on the perceived credibility of sports articles is severely lacking, Wann and Branscombe (1992) determined that emo tional responses to a sports article differed based on the degree of identification with a sports team, so the present study hypothesized that a relationship may exist between perceived credibility and fan identification This hypothesis was supported as there was a significant difference in the overall credibility score of the article between the two fan identification groups Highly identified fan s found the article to be more credible as a whole and also in eleven out of twelve stimuli group s ( all exce pt the ESPN negative group). The fans of a
111 particular sport and team find the article about that team significantly more credible and this may be due to a familiarity with the topic. People may be more inclined to judge an article as credible if they ar e familiar with the subject being covered (Henkel & Mattson, 2011) Highly identified fan s i n the present study are presumably much more fa miliar with the topic and the sources providing the information, so it is not surprising that they would rate the article as more credible than fans with low identification Along with topic familiarity, high ly identified fans also have a stronger level of attachment to the topic being covered. While the article was not specifically targeted at fans of either team, highly identified Florida Gator fans may have assumed that the article was tailored more toward their own fan base, and, thus, rated the article as more credible than fans with low identification. Ultimately, different levels of attachment to the team and familiarity with the topic may have caused differing levels of attachment to the article, whic h resulted in a significant disparity of perceived credibility. Th e significant differences in credibility scores may also be attributed to the questions that were used to judge credibility. On the credibility scale (Bucy, 2003), for example, participants were asked question s regarding the fairness and flo w of the article, but also regarding how enjoyable and how interesting it was. A fan with low identification may r ate the article as very fair and having a nice flow, but also may not find the article en joyable or interesting to read. Because all of these factors impact credibility, all of them must be taken into account when assessing it Nonetheless, highly identified fan s found the article to be significantly more credible than fans with low
112 identifi cation which is something for those in the sports media field to consider. Because highly identified fan s already find the articles as moderately credible, sports news sites may try to adapt t heir websites to the needs of fan s with low identification to appeal to a wider audience and gain more readership. T he stimuli groups that garnered the highest credibility scores from each identification group were also noteworthy The stimuli group that was rated highest by low identification fans was the ESPN.com group with negative user comments, and for highly identified fan s it was the ESPN.com grou p with positive user comments. Predictably, t he stimuli group that would presumably score the highest from highly identified fan s was rated highest by that group ESPN is an established news brand, and the comments may strike a chord with those participants in the study. Conversely, the fans with low identification chose the ESPN article with negative user comm ents as the most credible. F ans with low identificatio n were either unaffected by the comments or the lack of connection to the team made the article seem more genuine with comments that did not reflect the common opinion of a Florida Gator sports fan. In fans experienced the most positive mood state from an article that described a victory for their favorite team and whose author was an admitted loyal fan of the same team. Also, individuals with very low fan identification were not significantly influenc ed by the game out come group In a related social identity study, Wann et al (2001) found that ingroup favoritism and outgroup bias are most likely when individuals are highly identified w ith their team and have been threatened. The negative comments can be viewed as a threat to social identity for
113 highly identified fan s, which may have caused them to rate those articles more harshly T findings reflect the notion that h igh fan identification can significantly influence how one views an article and views ingroup/outgroup members, and that low identification has a much milder effect There were also score disparities between the two identification groups for the articles w ith the least credibility. Fans with low identification found the Facebook article with negative comments to be the least credible article Non sports fans in a college environment still utilize Facebook which has over 900 million members worldwide ( Kal las, 2012 ), but probably not for sports information. Because the primary us e of Facebook is not for sports it is somewhat predictable that fans with low identification would rate the article as least credible. For the highly identified fan s, the least c redible article was the ESPN.com group with negative comments. The se findings give mild support to the SIDE model, which asserts that a nonymous computer mediation obscure s the interp ersonal differences in group identification, so people will move toward a more extreme position in the direction already favored by the group (Lea & Spears, 1991; Spears & Lea, 1992). Essentially, anonymous communication lacks individuating cues, so the attention is shifted from distinctive characteristics of group members and people may act more strongly in favor or against the anonymous communication (Postmes, Spears & Lea, 1998). In this situation, highly identified Florida Gator fans react ed more strongly against the comments than they would in a situation that lacks anonymity (e x. hearing the same comments at a liv e football game from an opposing fan) and rated the article as the least credible in this instance A die hard Gator fan would presumably want to hear positive things about his or her favorite
114 team which could potent ially enhance perceived credibility However, negative user comments did not always garner a lower credibi lity rating for highly identified fan s in each stimuli group While anonymous c omments that the potential to cause a negative e motional response in the fan, their influence on credibility may not be as profound sports program lead to more positive impressions of the university itself (Wa nn & Robinson, 2002 ; Clopton, 2007 ) Highly identified fan s in the present study score d nearly a full point higher (out of 7) on the university identifi cation scale than low identification fans This significant dif ference between the two groups sug gests that a nection to the entire university In a previous study, c urrent students at a university were shown to ocial id entity and as being a pa rt of the larger community (Boyle & Magnusson, 2007). T he impact of a sports program is perhaps more profound at a large university where athletics are a large part of the student culture, like the University of Florida Students who are satisfied with and identify w ith the university will presumab ly do better in school and donate more to the university after graduation. This provides validat ion for the importance of sports on college campuses where many lessons can be learned out side the classroom. Additionally, the connection to a universit beyond graduation, so the potential for alumni to contribute as sports consumers is extremely high.
115 Conversely, p rominent university sports programs where fan s g enerally identify highly with the team have a lso been shown to have a potentially dangerous effect on the culture of a university and town, as evidenced by the shocking incidents involving Penn State University. Wolken (2012) blamed the culture of Penn S tate for allowing a sexual predator (Jerry Sandusky) to remain a part of its football program for over ten years after his first molestation incident on campus Speaking about the football program and head coach Joe Paterno, the author write ou combine that unbridled thirst with the money it generates and the power it fuels, anything can happen, even a football coach protecting so meone as sick as Jerry Sandusky (para. 10). Those in power at Penn State knew about several molestation incidents that occurred on campus, and chose to conceal facts from law enforcement what Penn State football was para. 14 ). The fact that a football coach can ascend to such power at a university illuminate s the impact that sports has on a university and how it relates to social identity. Despite the overwhelming evidence to the contrary, there are students on the Penn State campus who believe that Joe Paterno is innocent of all wrongdoing and it may be a social strategy from fans to cope with being members of a negatively distinctive ingroup (Jackson et al., 1996). As evidenced in the present study, highly identified fan s of the fans with l ow identification statements over the statements in an independent report to alleviate the negative impact it has on their own social identity. Although sports may not be as important a s
116 academics, sports can play a prominent and potentially harmful role on university campuse s both in and outside the classroom Additionally, d ata collection for the present study occurred during a time in which the integrity of college athletics was being questioned by some. Along with the Penn State tragedy, the Ohio State University football team was facing sanct ions for trying to cover up student athletes receiving improper benefits (Brooks, 2011) The head coach of the Arkansas Razorbacks football team had also re cently been fired for unfairly hiring his mistress and intentionally misleading his boss about their relationship (Schad, 2012). These incidents and others in the college athletics landscape may have been fresh in the minds of the experiment participants and may have influenced the results of the study to a degree. While the article used as the study stimulus was a game recap and did not address any sports scandals, the current state of collegiate athletics may have negatively impacted how participants ra ted themselves as sports fans and fans of the Real world events may have lowered the overall fan identification levels of the participants, and, in turn, potentially lowered the credibility ratings of the article for bo th identification groups. User comments have become ubiquitous throughout online me dia and continue to evolve with new Internet advancements (Gsell, 2009) Although the impact of anonymous comments on credibility have not been thoroughly researched, anony mous online interactions as developed in the SIDE model, have been shown to polarize a In the present study, t he tone of the user comments was hypothesized to impa ct a highly identified f an perception of credibility. However, t here was no significant difference between the
117 credibility scores based the ton e of the user comments User comments have been a staple of online news and sports articles for several years, and perhaps, online news reader s have become desensitized to user comments over time. Highly identified fan s read more sp orts articles, so it is possible that they would have been more attentive to the user comments. However, the impact of the user comments could have been reduced bec ause highly identified fan s may have grown to block out both positive and negative user comments, especially if they do not post regularly in the comments section. The lack of influence on credibility may also have resulted from the intended message of th e comments. The anonymous user comments were directed at the season. The comments did not, however, praise or disparage the article itself, which may explain why the credibility ratings were not significantly different for either fan identification group. Because highly identified fan s have a connection to the University of Florida basketball team, i t was also hypothesized that the perceived credibility of the ar ticle will change more significantly by the tone of the user comments for highly identified fan s than for low identification fans In both scenarios, credibility was not significantly impacted by the tone of the user comments. For the fans with low identification articles with positive user comments garnered the highest credibility scores from participants. However, unlike the highly identified fan s, articles with negat ive comments were rated as slightly more credible than articles without comments The similarity in credibility scores for both identifi cation groups based on the user comments gives further credence to the assumption that people have come to ignore or block out the user
118 comments. For newspaper websites, Gsell (2009) argued that the initial creation of user comments for online articles was an obvious choice with the open culture of the harassment, and unpaid advertising onto the site, creating for staff th e arduous daily duty of deleting off disrespectful, user comments that can be posted on a single article have changed the way online news consumers view user comments. Political news or c overage of more controversial topics may garner a more profound response from readers who see user comments that greatly support or vehemently oppose their viewpoints, but most sports articles rarely elicit such a response. Even if the presenc e of user co mments has a small impact on news articles, the results of the present study suggest that user comments do not impact the credibility of sports articles especially as off color and irrelevant comments become more prevalent Aside from credibility, the eff ect of user comment to ne on identification with the anonymous users was also examined It was hypothesized that highly identified fan s would have stronger feelings either in favor or against the users based on the positive or negative tone of the comment s The results showed that t he scores on the user identification scale (Walther et al., 2010; Postmes et al., 2001) were significantly higher for the positive comments than for the negative comments for both identification groups, and the differences were more significant for the highly identified fan s. F ans with low identification identified more closely with the positive users than the negative users by almost a full point on the scale (out of 7) and the highly identified fan s identified almost two full points high er with the positive users than the negative users Data a nalysis that
119 combined all independent variables (fan identification level, sports source received, and user comment tone) also showed that two of the variables had a significant inf luence on group identification scores with the anonymous users. Fan identification level and the tone of the user comments were both shown to influence user group identification, and the interaction between identification level and user comment tone was a lso found to significantly influence user identification. In the present study, highly identified fan s who received the article with positive comments identified more strongly with the anonymous users than low identification fans who received the art icle with either positive or negative comments. This provides further evidence for the assumption that user comments can as evidenced earlier, may not have an impact on the perceived credibility of the article. Even though level of fan identification had a significant impact on credibility, the tone of the user comments only i nfluenced findings are noteworthy because they contradict most of the earlier research on anonymous peers and the SIDE model such as in which supportive or negative comments significantly affected the participant evaluations of YouTube PSA s positively or negatively. Sc holars who have used the SIDE model claim that the lack of individuation information in anonymous online communication fosters group identification (Lee, 2007, Postmes et al., 2001) but high levels of identification with the users was noticeably absent in the present study. Surprisingly, both identification groups in the present study identified mostly negatively to t he users. Fans wit h low identification and highly identified fan s both had very low scores for the users who posted negative comments, and highly identified f an s were the only group that
120 even had a moderate connection to the users who posted positive user comments. Al thou gh highly identified fan s identify strongly with the team that is being praised by the users, this did not translate to a strong connection w ith the users who were also on their side. User comments may be a pervasive necessary aspect of modern journalism, but their impact on sports readers and credibility appears to be minimal in this case Finally, the role of user comments in evaluating a n ews source was also examined where i t was hypothesized that highly identified fan s will have stronger positive or negative feelings about the media source itself (website or AP article) than fans with low identification T he evaluation scores of all four news sources were not however, significantly impacted by the tone of the user comments. Conversely data analysis of the influence of all three independent variables on sports source evaluation scores showed a significant influence for two of the variab les Fan identification level and sports source received were separately shown to significantly influence source evaluation scores. The interaction assigned sports source also significantly influenced how favorably that source was rated. Essentially, highly identified fan s rated each sourc e significantly higher than low identification fans who received the same source but this was not impacted by positive or negative user comments For both iden tification groups, positive user comments garnered the highest evaluation scores of the website s or the Associated Press followed by negative user comments, which was trailed by the control group with no comments. These particular findings reflect the other findings of this study in which user comments ha d no significant impact on multiple evalua tion sc ale s Perhaps u ser comments had no effect o n article credibility and the evaluation of a news sour ce
121 because people can now separate the opinions of anonymous users from the facts provided by journalists and news organizations. News readers are pr esumably blocking out the user comments or internally separating opinions from facts, so, ultimately, the impact on the readers is negated. B ecause user comments have become a place where spam and profanity can be as prevalent as the viewpoint of a concer ned citizen (Gsell, 2009), they have turned into an aspect of news that can be glossed over or even completely ignored by readers. Although not initially incorporated into the design of the present study, t he impact of gender on perceived credibility was a lso examined Demographic characteristics, such as age, gender, and education, have been recognized as variables that can dictate Robinson and Kohut (1988) also found that gender is an important facto r in attitudes toward the press, with women consistently being more willing to believe the news media. Conversely, Whitney (1986 ) determined that attitudes towards the media credibility are only weakly correlated with demographic variables. In a contempo rary study in online news credibility, no demographic variables (gender, age, education, race, and political views) were shown to exert a significant influence on perceptions of online credibility (Cassidy, 2007). The sample population shared many of the same demographic characteristics overall particularly age and education ( all of the participants were college students at the same university ) However, participant gender was unbalanced as about two thirds of the sample population was female. The resu lts showed that men and women had similar scores in nearly every scale used. Both genders had statistically similar scores on the fan identification scale, which
122 was the basis for determining whether a participant was classified as a highly identified fan or a fan with low identification Identification scores with the university and with the an onymous users were also close between the two genders. Additionally, m ales and females had similar evaluation ratings of the sports news sources. A gender differ ence did exist, however, for perceived credibility with males rating the article as significantly more credible than females. This correlates with some of the previous research on credibility (Westley & Severn, 1964; Robinson & Kohut, 1988), but the influ ence on all other variables was not significant Although both genders identified fairly equally as fans, m en viewed the sports article as more credible. This may have occurred because be more interested in sports featuring athletes of their own gender. A sports article covering a influenced how both genders rated the article. Although a consensus about the influence of gender on credibility has not been reached, these findings suggest that there may be a disc repancy in how men and women perceive the credibility of sports articles despite having similar levels of fandom. Ultimately, the sample population had twice as many women but it reflected that of a more evenly distributed sample population. Even though article credibility was significantly different, fan identification sc ores were similar between m en and women and other potential factors such as author gender, may have attributed to the differences.
123 Future Research Because the research in this field is severely lacking, future studies in sports journalism can examine wa ys that article credibility in sports is different from article credibility in other fields. Similar to the present study, another study can be designed to measure the factors that contribute most to sports credibility, and juxtapose those results with th e factors that contribute most to the credibility of hard news, entertainment, politics, and business. Online news organizations would benefit from Huffington Post have a mix of all of these genre s on the front page. Websites want to attract new users and maintain those users for more advertising revenue, and understanding the factors that enhance their credibility and reader perceptions could help achieve this goal. To expand on the present study the study can be redesigned so that participants all read the same article but are not all looking at the same background. The stimuli would look more natural if participants see a replica of the actual websites being studied. For example, one article can be photoshopped so that the article appears on a screen of a real ESPN.com background or a Facebook background. By doing this, participants who are not familiar with the website may rate it differently if they judge the website layout more favorably o r negatively The s tudy could also be expanded to recruit from a population in a different demographic that is more reflective of sports fans outside of a college environment. The present study could be also be revisited at different schools, especially schools in different regions and with different athletic cultures. The University of Florida is known for its excellence in athletics and its stellar fan support, so it would be interesting to re create this study at a university with a Division II athlet ics program, where the fans are much less involved and identified with
124 the university teams. In an expanded study, the anonymous user comments could be rewritten to represent positive or negative comments about the article itself not the identified team i n the study to examine how this may impact credibility. The study could game as opposed to t identified t eam could potentially elevate ratings of perceived credibility or evaluations of a sports news source Additionally, a study that looks at specific website genres could also provide insight into the ways that credibility can differ for different audiences Looking at blogs for example, a salacious Tiger Woods article appearing on a gossip blog may be perceived as less credible than the same article on a sports blog. Both blogs attract a different audience, so it wou ld be interesting to see how different readers respond to the articles. Another study that exami nes the credibility of different types of sports articles could also benefit the academic community A study could be done, for example, comparing the perceived credibility of a n article listed as a long feature and an article listed as an opinion edit orial on a website Scholars could also look at how different types of articles affect social identity and fan identification, such as a study comparing the influence of reading a game recap, a player profile, or a column on team identification. Limitations T he largest limitation of the present study is the sample population. Although commonly used in other studies (Wann & Branscombe, 1992; Pham, 1992; Wann, 1995) the use of a convenient sample limit s the generalizability of the findings and hinders external validity All of the participants were undergraduate students at the
125 university of interest, so the findings are less applicable to non students or to undergraduate students at a university witho ut a prominent athletics department. Additionally, most of the participants were students in the College of Journalism and Communication s and are seemingly more adept at answering questions about credibility than the avera ge individual The sample popu lation also had nearly twice as many female participa nts as male participants, so a gender bias may have also shifted the findings. Also, the cell sizes ranged from 29 to 33 participants, so the data analysis may have garnered some artificially high signi ficance levels and perhaps g ave unwarranted support for some hypotheses As a result, the findings should be generalized with caution. The study was also limited by the inherent artificiality of an online experiment. Although participants all received the same article to read as well as the same pre and post test questions, participants were not closely monitored as would have been the case in a controlled experimental setting. P articipants were not allowed to leave the experiment webpage at any time during the experiment, but they were not given a time limit to complete it, so some may have been more judicious than others when reading the stimu li and answering the questions. Additionally, participants who received the w ire service stimuli read the article on a computer instead of a more traditional media format (such as print) which also increased the artificiality of the experiment for the participants assigned to that group. The design of the article page may also have heightened experiment artificiality because part icipants were all shown the same white background when reading the article which does not reflect the actual design of the web sites. Participants not familiar with the design of each website may have rated it
126 differently if they had been shown the articl es on exact webpage replicas. Ultimat e l y, i ssues with the experimental instrument and the sample population may have slightly altered the results. Conclusions Overall, t his study examined the key factors that can impact the credibility of a sports article and percept ions of a news source. Using fan identification as the primary theoretical framework, the effect of the medium and source on the credibility of sports articles for both highly identified and l ow identification sports fans w as tested Additionally, t he study analyze d how attitudes towards a website and a sports article can be affected by positive and negative user comments. Unlike previous findings on credibility (Kiousis, 2001; Gaziano & McGrath, 1986; Sun der & Nass, 2001), medium had little impact on the perceive d credibility of the article in this study The results suggest that c redibility of all media has seemingly evened out in the last decade where all types (print, radio, television, and online) have similar credibility perceptions from news consumers, but none are rated as overly credible. Online news has, perhaps, come the furthest in the last decade to a point where people are comfortable using it, but still do not fully trust the information they receive The onus falls on media organizations to have high ethical standards and fact checking procedures because with so many options available, the medium is becoming a small factor in how the public views credibility. Journalism scholars, both in and out of the sports world, can use these findings as a w ay to educate new journalists on the importance of solid reporting. The news industry is constant ly changing across all media and news professionals must find new and innovative ways to adapt to this new landscape and the modern news consumer
127 While the re was no major difference in the credibility ratings of the online sources overall there was a large disparity in the credibility scores between the two identification groups. O ther studies have found similar reactions from highly identified fan s regard ing their favorite teams although in different research areas (Wann & Branscombe, 1992; Wann & Robinson, 2002; Fink et al., 2009). Highly identified fan s in almost every stimuli group found the article to be more credible than fans with low identificatio n which is beneficial for sports media organizations who are trying to maintain the fans that already visit their website or read their newspaper. However, this large credibility gap can also become a burden for media organizations who are trying to attr act new users or readers. Because highly identified fan s have a generally positive view of sports media, the pressure is now on sports media professionals to deviate from the status quo and experiment with new tactics to draw in the less identified fans, such as utilizing a new webpage layout or new storytelling t echniques with their articles. Fan identification was also shown to have a profound impact on e valuations of the online news sources Highly identified fan s ha d a higher opinion of the websites o verall, while fans with low identification rated all the websites slightly negatively except Facebook Predictably, highly identified fan s rate d the sports w ebsites more favorably t han low identification fans but even Facebook received higher evaluations from the highly identified fan s, which may suggest that highly identified fan s are more exposed to and mo re comfortable using media than low identification fans Sports fans can now watch live games from their computer, cell phone, or tablet, and this level of accessibility may make sports fans more tech nologically dependent than most people. Sports media organizations have opportunities for expansion as technologies develop
128 and these organizations should look for ways to utilize the technology by not simply adopting the new tech craze before a competitor. This involves staying ahead of the competition in all facets of journalism from reporting to presentation in order to increase fan satisfaction and the trustworthiness of their brand Although it has been studied more extensively in traditional journalism (Gunter et al. 2009; Gunter, 2006), sports journalism scholars should examine brand establishment and brand growth in the sports field to better understand the companies in thi s multi billion dollar industry and better educate aspiring journalists in the field. Contradicting much of the earlier research on anonymous computer mediated communication (Postmes, Spears & Lea, 1998; Lee, 2007; Walther et al., 2010), u ser comments were shown to have little impact on credibility or evaluations of a news source The comments only significantly i nfluenced the extent to which the participants identified with the users themselves Altho ugh user comments have become a universal aspect of on line journalism and blogging, these findings suggest that their overal l role in sports journalism has been diminished Whether readers avoid the comments or choose to ignore their message, user comments have lost their power with readers and do not signif icantly impact credibility. In many ways, this is better for news organizations because people can now internally separate the anonymous users from the website and the organizations are not perceived as responsible for the unwelcome comments that some pe ople post. Although it may not affect credibility, social identity sch olars can expand on the findings in this study and design other studies that examine the role of user comments on social identity, ingroup favoritism, and outgroup bias.
129 a environment sports journalists are facing new challenges as they work to appease fans, who now have credibility concerns and unlimited access to coverage of their fav orite teams from a multitude of sources. Sports fans, and news consumers in general, a re looking for information that is both current and credible, and modern sports journalists must look for ways to write up to the minute stories without sacrificing credibility. Ultimately, t suggest that medium, source, and user comme nts matter less to fans when making credibility assessments than fan identification with their favorite team. This puts the onus on sports journalists to create stories that are not generic, exude creativity, and manage to resonate with every fan from th e most causal to the most extreme Not only will the fans benefit, but sports journalism as a whole will benefit.
130 APPENDIX A ORIGINAL ASSOCIATED PRESS ARTICLE Louisville in Final 4 with 72 68 win over Florida BYLINE: By EDDIE PELLS, AP National Writer SECTION: SPORTS NEWS DATELINE: PHOENIX Hated to do that to ya, kid. Rick Pitino nearly came unhinged and his point guard watched the end of the game from the bench. When it was over, though, it was Pitino and Louisville making plans for the Final Four a nd his protege, Billy Donovan, and the Florida Gators wondering what the heck happened. Freshman forward Chane Behanan made the go ahead basket with 1:06 left Saturday and the fourth seeded Cardinals finished the game on a 23 8 run for a 72 68 victory over Donovan's stunned Florida team in the West Regional final. Russ Smith, who finished with 19 points, followed Behanan's bucket with a pair of free throws and then Florida freshman Bradley Beal and teammate Kenny Boynton each missed chances to tie in the fi nal seconds. Louisville made one more free throw to seal the game and reach its ninth Final Four, the second under Pitino, despite playing the final 3:58 without point guard Peyton Siva, who fouled out. "What happens is, you can't lose confidence," Pitino said. "I kept telling the guys, 'We're going to the Final Four. Win the Big East tournament, you're going to the Final Four,' and they did." The Big East tournament champions are now going for the NCAA title, too. They're on an eight game winning streak, w ith a trip to New Orleans on the itinerary and a possible matchup with Pitino's old school, Kentucky, which will have to get by Baylor on Sunday to set up a grudge match to end them all. This game had a much more warm and fuzzy story line: Pitino, the youn g coach who saw something special in Donovan, the undersized guard, and developed a partnership that took Providence on an unexpected trip to the 1987 Final Four. Pitino also gave Donovan his first coaching job and both men conceded theirs was more of a fa ther son relationship than anything else.
131 "I'm so proud of Billy Donovan, the way he coached this team," Pitino said. "He was brilliant. He took us out of the zone. But only one team could play aggressive and come back like this." Seventh seeded Florida (2 6 11) went out in the regional final for the second straight year, with Donovan falling to 0 7 lifetime against the man who hired him as an assistant at Kentucky and felt as proud as a papa when he watched Donovan win his two national titles in 2006 and 20 07. But make no mistake. This was no heartwarmer. Donovan got under Pitino's skin early in the second half during a timeout when he worked over the officials, who promptly called a foul against the Cardinals (30 9) when play resumed. "He called that," Piti no shouted. "Why don't you just give him a whistle?" Pitino couldn't get a break for a while after that and when Siva picked up his fourth foul, the coach stomped onto the court and got hit with a technical. Erving Walker made four straight free throws and the Gators led by 11, setting the stage for what could've been Donovan's fourth trip to the Final Four. But the team that went 8 for 11 from 3 point range in the first half went cold really cold not hitting any of nine attempts from beyond the arc in the second. The Gators missed seven shots and committed one turnover over the last 2:30. They didn't score after Boynton's layup gave them a 68 66 lead with 2:39 left. The game's best freshman? That was Behanan, who was far less heralded than Beal coming o ut of high school, but outplayed him down the stretch when the trip to New Orleans was on the line. The freshman from Cincinnati scored 13 of his 17 points in the second half, including nine over the last 8:02 and Louisville's last two field goals both a fter Siva had fouled out with nine points and eight assists. Beal, meanwhile, matched Erik Murphy with a team high 14 points and controlled this game for the first 37 minutes. But over the last 3, he tried twice to take the ball to the hoop, only to get de nied by 6 foot 10 center Gorgui Dieng. Beal missed the desperation 3 in the waning seconds and also got called for traveling after stealing a wild pass from Smith while Louisville was nursing a one point lead with 18 seconds left. In the first half, Donova n looked like the better coach, though anyone would look good when his team is shooting that way. The Gators went 8 for 11 from 3, 6 for 10 from inside the arc and constantly harassed Louisville en route to a 41 33 lead.
132 But Pitino didn't become the first coach to take three programs to the Final Four for nothing. He scrapped the zone defense, had his players get up in the face of Florida's players and it worked though the Cardinals paid the price in foul trouble. Behanan and Kyle Kuric each finished the game with four fouls. When Siva got his fourth and Pitino got his 'T,' the coach tried to settle things down, stepping up to Siva and saying, "It was a foul. Stop saying it wasn't." Whether he believed it or not, who knows? As for whether he thought a come back was possible, especially playing the final 4 minutes without Siva on the floor? Well, Pitino did say in the lead up to this game that the 1987 trip to the Final Four made him believe anything's possible. This is his first trip back since 2005.
133 APPE NDIX B REVISED ARTICLE FOR PARTICIPANTS Louisville Defeats Florida to Reach Final Four March 25, 2012 Brad Seal ESPN.com Phoenix, Az. Hated to do that to ya, kid. Rick Pitino nearly came unhinged and his point guard watched the end of the game fr om the bench. When it was over, though, it was Pitino and Louisville making plans for the Final Four and his protege, Billy Donovan, and the Florida Gators wondering what ( the heck** ) happened. Louisville freshman forward Chane Behanan made the go ahead basket with 1:06 left and the fourth seeded Cardinals finished the game on a 23 8 run for a 72 68 victory over Donovan's stunned Florida team in the West Regional final. Russ Smith, who finished with 19 points, followed Behanan's bucket with a pair of fr ee throws and then Florida freshman Bradley Beal and teammate Kenny Boynton each missed chances to tie in the final seconds. Louisville made one more free throw to seal the game and reach its ninth Final Four, the second under Pitino, despite playing the final 3:58 without point guard Peyton Siva, who fouled out. "What happens is, you can't lose confidence," Pitino said. "I kept telling the guys, 'We're going to the Final Four. Win the Big East tournament, you're going to the Final Four,' and they did." The Big East tournament champions are now going for the NCAA title, too. They're on an eight game winning streak, with a trip to New Orleans on the itinerary and a possible matchup with Pitino's old school, Kentucky, which will have to get by Baylor on Sunday to set up a grudge match to end them all. This game had a much more warm and fuzzy story line: Pitino, the young coach who saw something special in Donovan, the undersized guard, and developed a partnership that took Providence on an unexpected tr ip to the 1987 Final Four. Pitino also gave Donovan his first coaching job and both men conceded theirs was more of a father son relationship than anything else.
134 "I'm so proud of Billy Donovan, the way he coached this team," Pitino said. "He was brilliant He took us out of the zone. But only one team could play aggressive and come back like this." Seventh seeded Florida (26 11) went out in the regional final for the second straight year, with Donovan falling to 0 7 lifetime against the man who hired him as an assistant at Kentucky and felt as proud as a papa when he watched Donovan win his two national titles in 2006 and 2007. But make no mistake. This was no fun filled family affair Donovan got under Pitino's skin early in the second half during a timeout when he worked over the officials, who promptly called a foul against the Cardinals (30 9) when play resumed. "He called that," Pitino shouted. "Why don't you just give him a whistle?" Pitino couldn't get a break for a while after that and when Siva picked up his fourth foul, the coach stomped onto the court and got hit with a technical. Erving Walker made four straight free throws and the Gators led by 11, setting the stage for what could've been Donovan's fourth trip to the Final Four. But t he team that went 8 for 11 from 3 point range in the first half went cold really cold not hitting any of nine attempts from beyond the arc in the second. The Gators missed seven shots and committed one turnover over the last 2:30. The team didn't sco re after Boynton's layup gave them a 68 66 lead with 2:39 left. The game's best freshman? That was Behanan, who was far less heralded than Beal coming out of high school, but outplayed him down the stretch when the trip to New Orleans was on the line. The freshman from Cincinnati scored 13 of his 17 points in the second half, including nine over the last 8:02 and Louisville's last two field goals both after Siva had fouled out with nine points and eight assists. Beal, meanwhile, matched Erik Murphy with a team high 14 points and controlled the game for the first 37 minutes. But over the last 3, he tried twice to take the ball to the hoop, only to get denied by 6 foot 10 center Gorgui Dieng. Beal missed the desperation 3 in the waning seconds and al so got called for traveling after stealing a wild pass from Smith while Louisville was nursing a one point lead with 18 seconds left.
135 In the first half, Donovan looked like the better coach, though anyone would look good when his team is shooting that wa y. The Gators went 8 for 11 from 3, 6 for 10 from inside the arc and constantly harassed Louisville en route to a 41 33 lead. But Pitino didn't become the first coach to take three programs to the Final Four for nothing. He scrapped the zone defense, ha d his players get up in the face of Florida's players and it worked though the Cardinals paid the price in foul trouble. Behanan and Kyle Kuric each finished the game with four fouls. When Siva got his fourth and Pitino got his 'T,' the coach tried to settle things down, stepping up to Siva and saying, "It was a foul. Stop saying it wasn't." Whether he believed it or not, who knows? As for whether he thought a comeback was possible, especially playing the final 4 minutes without Siva on the floor? W ell, Pitino did say in the lead up to this game that the 1987 trip to the Final Four made him believe anything's possible. This is his first trip back since 2005. *Note: Differences between the two articles are italicized. **Note: The words in parenthe sis were omitted from the article.
136 APPENDIX C EXPERIMENT QUESTIONN AIRE Pre test Questionnaire To me, sports are: 1. Boring 1 2 3 4 5 6 7 Exciting 2. Uninteresting 1 2 3 4 5 6 7 Interesting 3. Valuable 1 2 3 4 5 6 7 Worthless 4. Unappealing 1 2 3 4 5 6 7 Appealing 5. Use ful 1 2 3 4 5 6 7 Use less 6. Needed 1 2 3 4 5 6 7 Not Needed 7. Irrelevant 1 2 3 4 5 6 7 Relevant 8. Unimportant 1 2 3 4 5 6 7 Important 9. In a typical week, how many hours on average do you spend watching sports related pro gramming on television or online? ___________ __ 10. How many professional sporting events did you attend in the past year? _____________ 11. How many college sporting events did you attend in the past year? _____________ 12. In a typical we ek, how many hours on average do you spend reading sports related periodicals? _____________ 13. Among your favorite sports, where do you get your information about sports? (Mark all that apply) 1 Sports TV talk shows 2 Sports TV highlight shows (such as Sportscenter on ESPN)
137 3 Newspapers 4 Printed sports magazines (such as Sports Illustrated ) 5 Online 6 Other ________ ________ 7 None 14. When seeking sports information online, where do you get your information? (Mark all th at apply) 1 Sports blogs 2 Sports dedicated websites (such as ESPN.com or yahoosports.com) 3 Online newspaper websites (such as nytimes.com) 4 Online sport magazine websites (such as sportsillustrated.cnn.com) 5 Twitter 6 Facebook 7 Other ____________________ 8 None For each question, please select the response that best indicates how you feel about the University of Florida men's basketball team. 15. How important is it to you that the UF basketball team wins? Not Importa nt 1 2 3 4 5 6 7 8 Very Important 16. How strongly do you see yourself as a fan of the UF basketball team? Not at all a fan 1 2 3 4 5 6 7 8 Very Much a Fan 17. How strongly do your friends see you as a fan of the UF basketball team? Not at all a f an 1 2 3 4 5 6 7 8 Very Much a Fan
138 18. During the season, how closely do you follow the UF basketball team via any of the following: a) in person or on television, b) on the radio c) television news or a newspaper d) online? Never 1 2 3 4 5 6 7 8 Almost Every Day 19. How important is being a fan of the UF basketball team to you? Not Important 1 2 3 4 5 6 7 8 Very Important 20. How much do you dislike Do Not Dislike 1 2 3 4 5 6 7 8 Dislike Very Much 21. How of of work, where you live, or on your clothing? Never 1 2 3 4 5 6 7 8 Always 22. What is your gender? 1 Male 2 Female 23. What was your age on your last birthday? _____ ___ 24. What is your field of study? 1 Telecommunications 2 Journalism 3 Public Relations 4 Advertising 5 Sport Management 6 Other ________________ Note: Responses to questions 1 8 and 15 21 were combined to make the sport fan iden tification scale of the present study, and the questions were derived from Shank (1993) team identification scale respectively.
139 On the following page, you will be shown a spor ts article. Please read everything you see on the following page. You will then be asked to answer questions regarding what you just read. Please answer those questions openly and honestly. Sample Article Louisville Defeats Florida to Reach Final Four March 25, 2012 Brad Seal ESPN.com Phoenix, Az. Hated to do that to ya, kid. Rick Pitino nearly came unhinged and his point guard watched the end of the game from the bench. When it was over, though, it was Pitino and Louisville making plans for th e Final Four and his protege, Billy Donovan, and th e Florida Gators wondering what happened. Louisville freshman forward Chane Behanan made the go ahead basket with 1:06 left and the fourth seeded Cardinals finished the game on a 23 8 run for a 72 68 vic tory over Donovan's stunned Florida team in the West Regional final. Russ Smith, who finished with 19 points, followed Behanan's bucket with a pair of free throws and then Florida freshman Bradley Beal and teammate Kenny Boynton each missed chances to ti e in the final seconds. Louisville made one more free throw to seal the game and reach its ninth Final Four, the second under Pitino, despite playing the final 3:58 without point guard Peyton Siva, who fouled out. "What happens is, you can't lose confi dence," Pitino said. "I kept telling the guys, 'We're going to the Final Four. Win the Big East tournament, you're going to the Final Four,' and they did." The Big East tournament champions are now going for the NCAA title, too. They're on an eight game winning streak, with a trip to New Orleans on the itinerary and a possible matchup with Pitino's old school, Kentucky, which will have to get by Baylor on Sunday to set up a grudge match to end them all. This game had a much more warm and fuzzy story lin e: Pitino, the young coach who saw something special in Donovan, the undersized guard, and developed a partnership that took Providence on an unexpected trip to the 1987 Final Four. Pitino also gave Donovan his first coaching job and both men conceded thei rs was more of a father son relationship than anything else.
140 "I'm so proud of Billy Donovan, the way he coached this team," Pitino said. "He was brilliant. He took us out of the zone. But only one team could play aggressive and come back like this." Sev enth seeded Florida (26 11) went out in the regional final for the second straight year, with Donovan falling to 0 7 lifetime against the man who hired him as an assistant at Kentucky and felt as proud as a papa when he watched Donovan win his two national titles in 2006 and 2007. But make no mistake. This was no fun filled family affair Donovan got under Pitino's skin early in the second half during a timeout when he worked over the officials, who promptly called a foul against the Cardinals (30 9) wh en play resumed. "He called that," Pitino shouted. "Why don't you just give him a whistle?" Pitino couldn't get a break for a while after that and when Siva picked up his fourth foul, the coach stomped onto the court and got hit with a technical. Ervin g Walker made four straight free throws and the Gators led by 11, setting the stage for what could've been Donovan's fourth trip to the Final Four. But the team that went 8 for 11 from 3 point range in the first half went cold really cold not hitting any of nine attempts from beyond the arc in the second. The Gators missed seven shots and committed one turnover over the last 2:30. The team didn't score after Boynton's layup gave them a 68 66 lead with 2:39 left. The game's best freshman? That was Behanan, who was far less heralded than Beal coming out of high school, but outplayed him down the stretch when the trip to New Orleans was on the line. The freshman from Cincinnati scored 13 of his 17 points in the second half, including nine over the l ast 8:02 and Louisville's last two field goals both after Siva had fouled out with nine points and eight assists. Beal, meanwhile, matched Erik Murphy with a team high 14 points and controlled the game for the first 37 minutes. But over the last 3, h e tried twice to take the ball to the hoop, only to get denied by 6 foot 10 center Gorgui Dieng. Beal missed the desperation 3 in the waning seconds and also got called for traveling after stealing a wild pass from Smith while Louisville was nursing a one point lead with 18 seconds left.
141 In the first half, Donovan looked like the better coach, though anyone would look good when his team is shooting that way. The Gators went 8 for 11 from 3, 6 for 10 from inside the arc and constantly harassed Louisville en route to a 41 33 lead. But Pitino didn't become the first coach to take three programs to the Final Four for nothing. He scrapped the zone defense, had his players get up in the face of Florida's players and it worked though the Cardinals paid the p rice in foul trouble. Behanan and Kyle Kuric each finished the game with four fouls. When Siva got his fourth and Pitino got his 'T,' the coach tried to settle things down, stepping up to Siva and saying, "It was a foul. Stop saying it wasn't." Whether he believed it or not, who knows? As for whether he thought a comeback was possible, especially playing the final 4 minutes without Siva on the floor? Well, Pitino did say in the lead up to this game that the 1987 trip to the Final Four made him believe anything's possible. This is his first trip back since 2005. Comments Positive comments 1. Congrats Gators on a great year! By JoeGator 2. t the win. All the other teams in the country need to watch out for us next year. Go Gators! By BillyD_4_Prez 3. Get ready for big things from this team next season. Final Four here we come! By Gator4Life Negative Comments 1. The Gators choked in t he tournament like they always do. By Noles_fan_87 2. Kentucky still has the best team in the SEC and the best in the country. Go Wildcats! By Kentucky_Blue 3. anythi ng since. By BulldogPride
142 Post Test Questionnaire Questions 1 9 were used as the source evaluation scale and were derived from For each question, please select the respon se that best indicates how you feel about the ESPN.com. (Gatorblog.com, Facebook .com, or the Associated Press ) 1. Associated Press .) Strongly disagree 1 2 3 4 5 6 7 Strongly agree 2. I feel cheer Associated Press .) Strongly disagree 1 2 3 4 5 6 7 Strongly agree Associated Press .) Strongly disagree 1 2 3 4 5 6 7 Strongly agree 4. I am sati Associated Press .) Strongly disagree 1 2 3 4 5 6 7 Strongly agree Associated Press Strongly disagree 1 2 3 4 5 6 7 Strongly agree Associated Press .) Strongly disagree 1 2 3 4 5 6 7 Strongly agree 7. The website does a good job of satisfying my needs. (The Associated Press Strongly disagree 1 2 3 4 5 6 7 St rongly agree 8. I will say positive things about this website to others. (The Associated Press Strongly disagree 1 2 3 4 5 6 7 Strongly agree 9. I will recommend this website to others who seek my advice. (The Associated Press Strongly disagree 1 2 3 4 5 6 7 Strongly agree
143 Questions 10 14 were used as the user comment identification scale for the present study and were derived from Postmes et al. 2001) study. **The following set of questions are all regarding the user comments at the end of t he article. After reading the user comments, 10. I feel a bond with these people. Not at all 1 2 3 4 5 6 7 Very much 11. I see myself as a member of this group. Not at all 1 2 3 4 5 6 7 Very much 12. I regard this group as important. Not at all 1 2 3 4 5 6 7 Very much 13. At this moment, I identify with this group. Not at all 1 2 3 4 5 6 7 Very much 14. The people who posted these comments were personally identifiable to me. Not at all 1 2 3 4 5 6 7 Very much Questions 15 23 were used as t he perceived credibility scale for the present study and were derived from ) study. For each question, please select the response that best indicates how you feel about the article itself. 15 How fair was the author in the article? Not at all fair 1 2 3 4 5 6 7 Very fair 16 How interesting was the article? Not at all interesting 1 2 3 4 5 6 7 Very interesting 17 How clearly written was the article? Not at all clear 1 2 3 4 5 6 7 Very clear 18 How well did the article flow ? Poorly 1 2 3 4 5 6 7 Very well
144 19 How enjoyable was the article to read? Not at all enjoyable 1 2 3 4 5 6 7 Very enjoyable 20 How accurate was the article? Not at all accurate 1 2 3 4 5 6 7 Very accurate 21 How believable was the ar ticle? Not at all believable 1 2 3 4 5 6 7 Very believable 22 How informative was the article? Not at all informative 1 2 3 4 5 6 7 Very informative 23 How in depth was the author on issues? Not at all in depth 1 2 3 4 5 6 7 Very in de pth Questions 24 33 were used as the university identification scale for the present study a nd were derived from Mael and Tetrick ( 1992 ) study. The following questions address your thoughts about the University of Florida. 24 When someone criticize s the University of Florida, it feels like a personal insult. Strongly disagree 1 2 3 4 5 6 7 Strongly agree 25 I am interested in what others think about the University of Florida. Strongly disagree 1 2 3 4 5 6 7 Strongly agree 26 When I talk about Strongly disagree 1 2 3 4 5 6 7 Strongly agree 27 es are my successes. Strongly disagree 1 2 3 4 5 6 7 Strongly agree 28 When someone praises this university, it f eels like a personal compliment. Strongly disagree 1 2 3 4 5 6 7 Strongly agree
145 29 I act like a University of Florida person to a great extent. Strongly disagree 1 2 3 4 5 6 7 Strongly agree 30 If a story in the media criticized UF, I would feel emba rrassed. Strongly disagree 1 2 3 4 5 6 7 Strongly agree 31 Strongly disagree 1 2 3 4 5 6 7 Strongly agree 32 I have a number of qualities typical of UF people. Strongly disagree 1 2 3 4 5 6 7 Strongly agree 33 The limitation s associated with UF people apply to me also. Strongly disagree 1 2 3 4 5 6 7 Strongly agree Note: The listed online source was changed to coincide with the appropriate website or the Associated Press. **Note: Participants in the control group did not receive this set of questions
146 LIST OF REFERENCES Adams, T. & Cleary, J. (2006). The Parity Paradox: Reader Response to Minority Newsroom Staffing. Mass Communication and Society 9 (1), 45 61. Addington, D. W. (1971). The effect o f vocal variations on ratings of source credibility. Speech Monographs, 38, 242 247. Albanesius, C. (2012). NBC, Adobe App to Live Stream Olympic Games. PC Magazine 1. Allan, S. (2006). Online News Open University Press: Maidenhead. Amncio, L. (1989) Social differentiation between 'dominant' and 'dominated' groups: Toward an integration of social stereotypes and social identity. European Journal of Social Psychology 19 (1), 1 10. Anastasio, P.A., Rose, K. C., & Chapman, J. (1999). Can the Media Creat e Public Opinion? A Social Identity Approach. Current Directions in Psychological Science (Wiley Blackwell) 8 (5), 152 155. Arant, M.D. & Anderson, J.Q. (2001). Newspaper online editors support traditional ethics. Newspaper Research Journal 22 (4), 57 69. Armstrong, C. & McAdams, M. (2009). Blogs of information: How gender cues and individual motivations influence perceptions of credibility. Journal of Computer Mediated Communication, 14 435 456. Barker, V. (2009). Older Adolescents' Motivations for S ocial Network Site Use: The Influence of Gender, Group Identity, and Collective Self Esteem. CyberPsychology & Behavior 12 (2), 209 213. Bee, C.C., & Havitz, M. E. (2010). Exploring the relationship between involvement, fan attraction, psychological commit ment and behavioural loyalty in a sports spectator context. International Journal of Sports Marketing & Sponsorship 11 (2), 140 157. Beyer, J.M., & Hannah, D. R. (2000). The Cultural Significance of Athletics in U.S. Higher Education. Journal of Sport Mana gement 14 (2), 105. Bizman, A., & Yinon, Y. (2002). Engaging in Distancing Tactics Among Sport Fans: Effects on Self Esteem and Emotional Responses. Journal Of Social Psychology 142 (3), 381 392. Blake, K.R., & Wyatt, R. O. (2002). Has Newspaper Credibili ty Mattered? A Perspective on Media Credibility Debate. Newspaper Research Journal 23 (1), 73.
147 Boen, F., Vanbeselaere, N., Pandelaere, M., Schutters, K., & Rowe, P. (2008). When Your Team is Not Really Your Team Anymore: Identification with a Merged Bask etball Club. Journal Of Applied Sport Psychology 20 (2), 165 183. Bolt, J., Talbert, J., Lawson Borders, G., Crossman, G., Garmon, M., & Limperos, A (2007). Blogs in communication research: Creating a definition and a content analysis of weblogs. Unpubl ished p aper presented at the 2007 annual conference of the National Communication Association, Chicago, IL Boyer, D. C. (2000). On the Sedimentation and Accreditation of Social Knowledges of Difference: Mass Media, Journalism. Cultural Anthropology 15 (4 ), 459. Boyle, B. A., & Magnusson, P. (2007). Social Identity and Brand Equity Formation: A Comparative Study of Collegiate Sports Fans. Journal of Sport Management 21 (4), 497 520. Brooks, M. (2011, December 20). Ohio State NCAA sanctions: one year post season ban, scholarships reduction, three years probation. WashingtonPost.com Retrieved October 2012, from http://www.washingtonpost.com/blogs/early lead/post/report ohio states ncaa sanctions include 2012 postseason ban announcement coming at 3 pm est/20 11/12/20/gIQAugnR7O_blog.html Brustein, J. (2010, December 13). American Internet Use Catches Up With TV Use. Nytimes.com Retrieved November 2011, from http://bits.blogs.nytimes.com/2010/12/13/american internet use catches up with tv use/ Bucy, E.P. ( 2003). Media Credibility Reconsidered: Synergy Effects Between On Air and Online News Journalism & Mass Communication Quarterly 80 (2), 247 264. Calabrese, A. & Borchert, M. (1996). Prospects for Electronic Democracy in the United States: Rethinking Comm unication and Social Policy. Media, Culture and Society, 18 249 268. Carlson, M. (2007). Blogs and Journalistic Authority Journalism Studies 8 (2), 264 279. Cassidy, W.P. (2007). On line news credibility: An Examination of the Perceptions of N ewspaper Journalists. Journal of Computer Mediated Communication 12 (2), 478 498 Chambers, R. (2012, July 23). Penn State Sanctions: Timeline of Sandusky Scandal That Brought NCAA Penalties. Bleacher Report. Retrieved July 2012, from http://bleacherreport.com/ articles/1268644 penn state sanctions timeline of sandusky scandal that brought ncaa penalties
148 Chan, J.K., & Leung, L. (2005). Lifestyles, reliance on traditional news media and online news adoption. New Media & Society 7 (3), 357 382. Choi, J.H., Watt, J. H. & Lynch, M. (2006). Perceptions of News Credibility about the War in Iraq: Why War Opponents Perceived the Internet as the Most Credible Medium. Journal of Computer Mediated Communication 12 209 229. Chung, C.J., Kim, H., & K im, J.H (2010). An an atomy of the credibility of online newspapers. Online Information Review 34 (5), 669 685. Clark, J. S., Ap ostolopoulou, A., & Gladden, J. M. (2009). Real Women Watch Football: Gender Differences in the Consumption of the NFL Super Bowl Broadcast. Journal of Promotion Management 15 (1/2), 165 183. Clopton, A.W. (2007). Predicting a sense of community amongst students from the presence of intercollegiate athletics: What roles do gender and BCS affiliation play in the relationship? The Sport Management and Rela ted Topics (SMART) Journal, 4 (1) 95 110. Clopton, A. W., & Finch, B. L. (2010). Are College Students 'Bowling Alone?' Examining the Contribution of Team Identification to the Social Capital of College Students. Journal Of Sport Behavior 33 (4), 377 402. D ae Hee, K., & McDaniel, S. R. (2011). Using an extended Technology Acceptance Model in exploring antecedents to adopting fantasy sports league websites. International Journal Of Sports Marketing & Sponsorship 12 (3), 240 253. Davis, J.L., & Gandy, O. H. (1 999). Racial identity and media orientation. Journal of Black Studies 29 367 397. Davis, N. (2012, May 24 ). Wi Fi key to packing NFL stadiums, Goodell says. USA Today Delpy, L. & Bosetti, H.A. (1998). Sport management and marketing via the World Wide Web. Sport Marketing Quarterly, 7 (1), 21 27. Deuze, M. (2003). The web and its journalisms: considering the consequences of different types of newsmedia online. New Media and Society 5 (2), 203 230. Dietz Uhler, B., Harrick, E. A., End, C., & Jacquemott e, L. (2000). Sex Differences in Sport Fan Behavior and Reasons for Being a Sport Fan. Journal of Sport Behavior 23 (3), 219 231. Dimmock, J. A., Gr ove, J. R. & Eklund, R. (2005). Reconceptualizing Team Identification: New Dimensions and Their Relationship to Intergroup Bias. Group Dynamics: Theory, Research, and Practice 9 (2), 75 86.
149 Duke, V., & Renson, R. (2003). From factions to fusions? The rise and fall of two club rivalries in Belgian football. International Review for the Sociology of Sport, 38 6 1 77. Ellemers, N., Kortekaas, P., & Ouwerkerk, J. W. (1999). Self categorisation, commitment to the group and group self esteem as related but distinct aspects of social identity. European Journal of Social Psychology 29 (2/3), 371 389. Fink, J.S., Park er, H. M., Brett, M., & Higgins, J. (2009). Off Field Behavior of Athletes and Team Identification: Using Social Identity Theory and Balance Theory to Explain Fan Reactions. Journal Of Sport Management 23 (2), 142 155. Fink, J.S., Trail, G.T., & Anderson, D. F. (2002). An examination of team identification: which motives are most salient to its existence? International Sports Journal 6 195 207. Fisher, R.J ., & Wakefield, K. (1998). Factors leading to group identification: A field study of winners and lose rs. Psychology and Marketing 15 (1), 23 40. Fujioka, Y. (2005). Black Media Images as a Perceived Threat to African American Ethnic Identity: Coping Responses, Perceived Public Perception, and Attitudes Towards Affirmative Action. Journal of Broadcasting & Electronic Media 49 (4), 450 467. Ferguson, M. (1990). Images of power and the feminist fallacy. Critical Studies in Mass Communication 7 215 230. Attitudes and Team Identification: An East West Perspective. Social Behavior & Personality: An International Journal 39 (5), 587 596. Honors. Gatorzone.com Retrieved December 2011, from http://www. gatorzone.com/story.php?id=21973 Gaziano, C., & McGrath, K. (1986). Measuring the concept of credibility. Journalism Quarterly 63 (3), 451 462. Greenberg, B.S., & Brand, J. E. (1994). Minorities and the mass media: 1970s to 1990s. In J. Bryant & D. Zillma nn (Eds.), Media effects: Advances in theory and research (pp. 273 314). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Gross, D. (2010, March 1) Survey: More Americans get news from Internet than newspapers or radio. CNN Tech Retrieved September 20 12, from
150 http://articles.cnn.com/2010 03 01/tech/social.network.news_1_social networking sites social media social experience?_s=PM:TECH Gsell, L. (2009). Comments Anonymous. American Journalism Review 31 (1), 16 17. Gunter, B. (1997). Measuring Bias on Television University of Luton Press: Luton. Gunter, B. (2006). Who do online news consumers trust?. Library and Information Update 5 (9), 40 41. Gunter, B., Campbell, V., Touri, M., & Gibson, R. (2009). Blogs, news and credibility. Aslib Proceedings 61 (2), 185 204. Hardin, M., & Shain, S. (2006). "Feeling Much Smaller than You Know You Are": The Fragmented Professional Identity of Female Sports Journalists. Critical Studies in Media Communication 23 (4), 322 338. Harris, J. C. (1998). Civil Societ y, Physical Activity, and the Involvement of Sport Sociologists in the Preparation of Physical Activity Professionals. Sociology of Sport Journal 15 (2), 138 153. Heere, B., & James, J. D. (2007). Sports Teams and Their Communities: Examining the Influence of External Group Identities on Team Identity. Journal Of Sport Management 21 (3), 319 337. Henkel, L.A., & Mattson, M. E. (2011). Reading is believing: The truth effect and source credibility. Consciousness & Cognition 20 (4), 1705 1721. Hogg, M. A., & A brams, D. (1990). Social motivation, self esteem, and social identity. In D. Abrams & M. Hogg (Eds.), Social identity theory: Constructive and critical advances (pp. 28 47). New York: Springer Verlag. Hogg, M.A., & Mullin, B. A. (1999). Joining groups to r educe uncertainty: Subjective uncertainty reduction and group identification. In D. Abrams & M. A. Hogg (Eds.), Social identity and social cognition (pp. 249 279). Oxford, England: Blackwell. Hogg, M.A., Turner, J. C., & Davidson, B. (1990). Polarized nor ms and social frames of reference: A test of the self categorization theory of group polarization. Basic and Applied Social Psychology 11 77 100. into Olympic history mea ns. Yahoo! News Retrieved August 2012, from http://news.yahoo.com/the white world of sports -what gabby douglas%E2%80%99 vault into olympic history means.html
151 Hovland, C. & Weiss, W. (1951). The influence of source credibility on communication effectiv eness. The Public Opinion Quarterly, 15 (4), 635 650. Hu, A. W., & Tang, L. (2010). Factors Motivating Sports Broadcast Viewership with Fan Identification as a Mediator Social Behavior & Personality: An International Journal 38 (5), 681 689. Hutchins, B. Rowe, D., & Ruddock, A. (2009). "It's Fantasy Football Made Real": Networked Media Sport, the Internet, and the Hybrid Reality of MyFootballClub. Sociology Of Sport Journal 26 (1), 89 106. Jackson, L., Sullivan, L., Harnish, R., & Hodge, C. (1996). Achi eving positive social identity: Social mobility, social creativity, and permeability of group boundaries. Journal of Personality and Social Psychology 70 241 254. Johnson, K. A., & Wiedenbeck, S. (2009). Enhancing Perceived Credibility of Citizen Journal ism Web Sites Journalism & Mass Communication Quarterly 86 (2), 332 348. Johnson, T.J., & Kaye, B. K. (1998). Cruising is Believing?: Comparing Internet and Traditional Sources of Media Credibility Measures Journalism & Mass Communication Quarterly 75 (2 ), 325 340. Johnson, T.J., & Kaye, B. K. (2000). Using is Believing: The Influence of Reliance on the Credibility of Online Political Information Among Politically Interested Internet Users Journalism & Mass Communication Quarterly 77 (4), 865 879. Johns on, T.J., & Kaye, B. K. (2004). Wag the Blog: How Reliance on Traditional Media and the Internet Influence Credibility Perceptions of Weblogs Among Blog Users Journalism & Mass Communication Quarterly 81 (3), 622 642. Kallas, P ( 2012, May 10). Facebook S tatistics 2012: Taking Over the World. Dreamgrow.com Retrieved July 2012, from http://www.dreamgrow.com/ Face book statistics 2012 taking over the world/ K erstetter, D.L., & Koovich, G. M. (1997). An Involvement Profile of Division I Women's Basketball Spec tators. Journal of Sport Management 11 (3), 234. Kim, J., & Park, H. (2011). The effect of uniform virtual appearance on conformity intention: Social identity model of deindividuation effects and optimal distinctiveness theory. Computers in Human Behavior 27 (3), 1223 1230. Kim, S.T. Weaver, D., & Willnat, L. (2000). Media Reporting and Perceived Credibility of Online Polls Journalism & Mass Communication Quarterly 77 (4), 846 864.
152 Kiousis, S. (2001). Public trust or mistrust? Perceptions of media cre dibility in the information age. Mass Communication and Society 4 (4), 381 403. Knobloc h Westerwick, S., & Hastall, M. R. (2010). Please Your Self: Social Identity Effects on Selective Exposure to News About In and Out Groups. Journal of Communication 60 (3), 515 535. Krumm, A. J., & Corni ng, A. F. (2008). Who Believes Us When We Try to Conceal Our Prejudices? The Effectiveness of Moral Credentials With In Groups Versus Out Groups. Journal of Social Psychology 148 (6), 689 710. Ksiazek, T. B., Malthouse, E. C., & Webster, J. G. (2010). News seekers and Avoiders: Exploring Patterns of Total News Consumption Across Media and the Relationship to Civic Participation. Journal of Broadcasting & Electronic Media 54 (4), 551 568. Lapchick, R., Kaiser, C., Russell, C. & Welch, N. (2010, June 9). The 2010 Racial and Gender Report Card: National Basketball Association. Retrieved April 2012, from http://tidesport.org/RGRC/2010/2010_NBA_RGRC.pdf Lapchick, R., Kitnurse, J. & Moss, A. (2010, September 29). The 2010 Racial and Gender Report Card: National Football League Retrieved April 2012, from http://www.tidesport.org/RGRC/2010/2010%20NFL%20Racial%20and%20Gend er%20Report%20Card.pdf Lapchick, R., Moss, A., Russell, C. & Scearce, R. (2011, April 27). The 2010 11 Associate Press Sports Editors Racial and Gender Report Card Retrieved April 2012, from http://www.tidesport.org/RGRC/2011/2011_APSE_RGRC_FINAL.pdf Laverie, D.A., & Arnett, D. B. (2000). Factors affecting fan attendance: The influence of iden tity salienc e and satisfaction. Journal of Leisure Behavior 32 (2), 225 246. Lea, M., & Spears, R. (1991). Computer mediated communication, deindividuation, and group decision making. International Journal of Man Machine Studies 34 283 301. Lea, M., Spears, R., Wa tt, S. E. & Rogers, P. (2000). The InSIDE story: Social psychological processes affecting on line groups. In T. Postmes, M. Lea, R. Spears, & S. Reicher (Eds.), SIDE issues centre stage: Recent developments in studies of de individuation in groups (pp. 47 6 2) Amsterdam: KNAW Lee, E.J. (2004). Effects of visual representation on social influence in computer mediated communication. Human Communication Research 30 (2), 234 259.
153 Lee, E.J. (2007). Deindividuation effects on group polarization in computer medi ated communication: The role of group identification, public self awareness, and perceived argument quality. Journal of Communication 57 385 403. Lee, F. F. (2005). Spectacle and Fandom: Media Discourse in Two Soccer Events in Hong Kong. Sociology of Spo rt Journal 22 (2), 194. Lee, F. F. (2009). Negotiating sporting nationalism: debating fan behaviour in 'China vs. Japan' in the 2004 Asian Cup Final in Hong Kong. Soccer & Society 10 (2), 192 209. Lenhart, A. & Fox, S. (2006 July 19 ). Bloggers: A portrai storytellers Washington: PEW Internet. Pewtrusts.org. Retrieved November 2011, from http://www.pewtrusts.org/uploadedFiles/wwwpewtrustsorg/Reports/Society_and_ the_Internet/PIP_Bloggers_071906.pdf Levy, D. (2012, July 31). 2012 Oly mpics: NBC Finding New and Unique Ways to Fail American Viewers. Bleacher Report Retrieved August 2012, from http://bleacherreport.com/articles/1279506 2012 olympics nbc keeps finding new and unique ways to fail american viewers Madrigal, R. (1995). Cog nitive and affective determinants of fan satisfaction with sporting event attendance. Journal of Leisure Research, 27, 205 227. Madrigal, R., & Chen, J. (2008). Moderating and Mediating Effects of Team Identification in Regard to Causal Attributions and S ummary Judgments Following a Game Outcome. Journal Of Sport Management 22 (6), 717 733. Mael, F. & Ashforth, B. (1992). Alumni and Their Alma Mater: A Partial Test of the Reformulated Model of Organizational Identification. Journal of Organizational Beha vior 13 (2), 103 123. Mael, F. & Tetrick, L. (1992). Identifying Organizational Identification. Educational and Psychological Measurement, 52 813 824. Markham, A. (1998) Life Online: Researching Real Experiences in Virtual Space. London: Sage. Markha m, D. (1968). The dimensions of source credibility of television newscasters. Journal of Communication, 18, 57 64. Marques, J.M., Yzerbyt, V. Y., & Leyens, J. (1988). The 'Black Sheep Effect': Extremity of judgments towards ingroup members as a function of group identification. European Journal of Social Psychology 18 (1), 1 16.
154 Mastro, D. E. (2003). A social identity approach to understanding the impact of television messages. Communication Monographs 70 (2), 98 113. McKay, S., Thurlow, C. & Toomey Zimme rman, H. (2005). Wired whizzes or techno slaves? Young people and their emergent communication technologies. In Williams, A. & Thurlow, C. (Eds.) Talking adolescence: perspectives on communication in the teenage years New York: Peter Lang, 185 206. McQua il, D. (1992). Media Performance: Mass Communication and the Public Interest Sage: London. Melnick, M. J. (1993). Searching for sociability in the stands: A theory of sports spectating. Journal of Sport Management 7 44 60. Metzger, M.J., Flanagin, A.J. & Medders, R. B. (2010). Social and Heuristic Approaches to Credibility Evaluation Online. Journal Of Communication 60 (3), 413 439. Meyer, H. K., Marchionni, D., & Thorson, E. (2010). The Journalist Behind the News: Credibility of Straight, Collaborative Opinionated, and Blogged "News ". American Behavioral Scientist 54 (2), 100 119. Milne, G.R., & McDonald, M. A. (1999). Sport marketing: Managing the exchange process. Sudbury, MA: Jones and Bartlett. Newhagen, J.E. & Levy, M.R. (1998). The future of jo urnalism in a distributed The Electronic Grapevine: Rumor, Reputation, and Reporting in the New On Line Environment (pp. 9 21). Mahwah, NJ: Lawrence Erlb aum Associates O'Cass, A., & Carls on, J. (2010). Examining the effects of website induced flow in professional sporting team websites. Internet Research 20 (2), 115 134. O'Connor, B., & MacKeogh, C. (2007). New media communities: Performing identity in an online women's magazine. Irish Jo urnal of Sociology 16 (2), 97 116. Almost As Many Follow Olympics Online As in Print. Mediapost.com Retrieved June 2012, from http://www.mediapost.com/publications/?fa=Article s.showArticle&art_aid=123022 Oates, T. P., & Pauly, J. (2007). Sports Journalism as Moral and Ethical Discourse. Journal of Mass Media Ethics 22 (4), 332 347. Ojala, K., & Nesdale, D. (2004). Bullying and social identity: The effects of group norms and di stinctiveness threat on attitudes towards bullying. British Journal of Developmental Psychology 22 (1), 19 35.
155 Outten, H., Schmitt, M.T., Garcia, D.M., & Branscombe, N. R. (2009). Coping Options: Missing Links between Minority Group Identification and Psyc hological Well Being. Applied Psychology: An International Review 58 (1), 146 170. Paulsen. (2011, November 17). Plenty of Conflicts of Interest For the Media in the NBA Lockout. Sport s Media Watch Retrieved November 2011, from http://www.sportsmedia watch.com/2011/11/plenty of conflicts of interest for the media in the nba lockout/ Pells, E. (2012, March 25). Louisville in Final 4 with 72 68 Win over Florida. The Associated Press. Retrieved March 2012, from http://www.lexisnexis.com. Peng, E.Y., T ham, N.I. & Xiaoming, H. (1999). Trends in online newspapers: a look at the US web. Newspaper Research Journal 20 52 63. Perks, T. (2007). Does Sport Foster Social Capital? The Contribution of Sport to a Lifestyle of Community Participation. Sociology O f Sport Journal 24 (4), 378 401. Pew Research Center. (2006, February 9). Bush a Drag on Republican Midterm Prospects. Retrieved October 2011, from http://www.people press.org/2006/02/09/bush a drag on republican midterm prospects/ Pew Research Center. ( 2010, September 12). Americans Spending More Time Following the News. Retrieved October 2011, from http://www.people press.org/2010/09/12/section 5 news media credibility/ Pham, M. (1992). Effects of involvement, arousal and pleasure on recognition of sp onsorship stimuli. Advances in Consumer Research, 19 (1), 85 93. Postmes, T., Spears, R., & Lea, M. (1998). Breaching or building social boundaries: SIDE effects of computer mediated communication. Communication Research 25( 6), 689 715. Postmes, T., Spe ars, R., Sakhel, K., & de Groot, D. (2001). Social influence in computer mediated communication: The effects of anonymity on group behavior. Personality and Social Psychology Bulletin 27 (10), 1243 1254. Putnam, R.D. (2000). Bowling alone: The collapse an d revival of American community. New York, NY: Simon & Schuster Randle, Q., & Nyland, R. (2008). Participation in Internet Fantasy Sports Leagues and Mass Media Use. Journal of Website Promotion 3 (3/4), 143 152. Raney, A. A., & Bryant, J. (2006). Handb ook of sports and media Mahwah, N.J: L. Erlbaum Associates.
156 Reicher, S. D., Spears, R., & Postmes, T. (1995). A social identity model of deindividuation phenomena. European Review of Social Psychology 6 161 198. Reysen, S., & Branscombe, N. R. (2010). F anship and Fandom: Comparisons Between Sport and Non Sport Fans. Journal Of Sport Behavior 33 (2), 176 193. Rimmer, T., & Weaver, D. (1987). Different question s, different answers? Media use and media credibility. Journalism Quarterly, 64, 28 36 Robins on, M. J., & Kohut, A. (1988). Believability and the Press Public Opinion Quarterly 52 (2), 174. Robinson, S. (2006). The mission of the j blog: recapturing journalistic authority online. Journalism 7 (1), 65 83. Schad, J. (2012, August 2012). Bobby Pe trino emotional, regretful. ESPN.com Retrieved October 2012, from http://espn.go.com/college football/story/_/id/8251166/bobby petrino fired arkansas razorbacks coach apologizes interview espn Schantz, O. J., & Gilbert, K. (2001). An Ideal Misconstrued: N ewspaper Coverage of the Atlanta Paralympic Games in France and Germany. Sociology of Sport Journal 18 (1), 69 94. Sefiha, O. (2010). Now's When We Throw Him Under the Bus: Institutional and Occupational Identities and the Coverage of Doping in Sport. Soc iology Of Sport Journal 27 (2), 200 218. Shadish, W. R., Cook T.D., & Campbell, D. T. (200 2 ). Experimental and quasi experimental designs for generalized causal inference Boston: Houghton Mifflin. Shank, M.D., & Beasley, F. M. (1998). Fan or fanatic: Refi ning a measure of sports involvement. Journal of Sport Behavior 21 (4), 435. Shannon, B. L. (2006). Building blogs: A multi dimensional analysis of the distribution of frames on the 2004 presidential candidate Web sites. Journalism & Mass Communication Qua rterly, 83 (2), 329 345. computer mediated communication. In M. Lea (Ed). Contexts of computer mediated communication (pp. 30 65). Hemel Hempstead, UK: Harvester Wheatshe af.
157 Spears, R., Lea, M., & Lee, S. (1990). De individuation and group polarization in computer mediated communication. British Journal of Social Psychology 29 121 134. Sundar, S. S., & Nass, C. (2001). Conceptualizing sources in online news. Journal of Communication 51 (1), 52 72. Sutton, W.A., McDonald, M.A., Milne, G.R. & Cimperman, J. (1997). Creating and fostering fan identification in professional sports. Sport Marketing Quarterly 6 (1), 15 22. Tajfel, H. (1978). Social categorization, social iden tity and social comparison. In H. Tajfel (Ed.), Differentiation between social groups: Studies in the social psychology of intergroup relations (pp. 61 76). London: Academic Press. Tajfel, H. (1981). Human groups and social categories. Cambridge, UK: Camb ridge University Press. Tajfel, H., & Wilkes, A. (1963). Classification and quantitative judgements. British Journal of Psychology 54 101 114. Thiel, S. (2004). Shifting identities, creating new paradigms. Feminist Media Studies 4 (1), 21 36. Trail, G ., Robinson, M., Dick, R. & Gillentine, A. (2003). Motives and points of attachment: Fans versus spectators in intercollegiate athletics, Sport Marketing Quarterly, 12 (4), 217 227. Turkle, S. (1995). Life on the Screen: Identity in the Age of the Interne t. New York: Simon and Schuster. Turkle, S. (1996). Parallel lives: Working on identity in virtual space. In D. Grodin & T. R. Lindlof (Eds.), Constructing the self in a mediated world (pp. 156 175). Thousand Oaks, CA: Sage. Turner, J. C. (1982). Towards a cognitive redefinition of the social group. In H. Tajfel (ed.), Social identity and intergroup relations Great Britain: Cambridge University Press. Van Leeuwen, L., Quick, S., & Daniel, K. (2002). The sport spectator satisfaction model: A conceptual framework for understanding the satisfaction of spectators. Sport Management Review, 5, 99 128. Walther, J. B., DeAnd rea, D., Kim, J., & Anthony, J. C. (2010). The Influence of Online Comments on Perceptions of Antimarijuana Public Service Announcements on YouTube. Human Communication Research 36 (4), 469 492.
158 Wann, D. L. (1995). Preliminary Validation of the Sport Fan Motivation Scale Journal of Sport & Social Issues 19 (4), 377 396. Wann, D. L. (2006). Understanding the Positive Social Psychological Bene fits of Sport Team Identification: The Team Identification Social Psychological Health Model. Group Dynamics: Theory, Research, and Practice 10 (4), 272 296. Wann, D.L., & Branscombe, N. R. (1992). Emotional Responses to the Sports Page. Journal of Sport & Social Issues 16 (1), 49 64. Wann, D.L., & Branscombe, N. R. (1993). Sports Fans: Measuring Degree of Identification with Their Team. International Journal of Sport Psychology 24 1 17. Wann, D.L., Dimmock, J., & Grove, J. R. (2003). Generalizing th e Team Identification Psychological Health Model to a Different Sport and Culture: The Case of Australian Rules Football. Group Dynamics: Theory, Research, and Practice 7 (4), 289 296. Wann, D.L., & Dolan, T.J. (1994). Attributions of highly identified s ports spectators. The Journal of Social Psychology, 134, 783 792. Wa nn, D. L., Friedman, K., & McHale, M. (2003). The Norelco Sport Fanatics Survey: Examining Behaviors of Sport Fans Psychological Reports 92 (3), 930. Wann, D.L., & Greive, F. G. (2005). Biased Evaluations of In Group and Out Group Spectator Behavior at Sporting Events: The Importance of Team Identification and Threats to Social Identity. Journal of Social Psychology 145 (5), 531 545. Wann, D. L., & McGeorge, K. K. (1994). Relationships B etween Spectator Identification and Spectators' Perceptions of Influence, Spectators' Emotions, and Competition Outcome. Journal Of Sport & Exercise Psychology 16 (4), 347 364. Wann, D.L., Melnick, M.J., Russell, G.W. & Pease, D. G. (2001) Sport fans: The psychology and social impact of spectators. New York: Routledge. Wann, D. L., Morriss Shirkey, P. A., Peters, E., & Suggs, W. L. (2002). Highly Identified Sport Fans and Their Conflict between Expression of Sport Knowledge and Biased Assessments of Team P erformance. International Sports Journal 6 (1), 153. Wann, D. L., & Polk, J. (2007). The Positive Relationship between Sport Team Identification and Belief in the Trustworthiness of Others. North American Journal Of Psychology 9 (2), 251 256.
159 Wann, D.L., & Robinson, I. N. (2002). The Relationship Between Sport Team Identification and Integration into and Perceptions of a University. International Sports Journal 6 (1), 36. Wann, D.L., Tucker, K.B. & Schrader, M. P. (1996). An exploratory examination of the factors influencing the origination, continuation, and cessation of identification with sports teams. Perceptual and Motor Skills 82 995 1101. Wann, D. L., & Weaver, S. (2009). Understanding the Relationship between Sport Team Identification and Dimensi ons of Social Well being. North American Journal Of Psychology 11 (2), 219 230. Westerhahl, J. (1983). Objective news reporting. Communication Research 10 403 24. Westley, B. H.,& Severin, W. J. (1964). Some correlates of media credibility Journalism Qu arterly, 41, 325 335. Whitehead, J. L. (1968). Factors of source credibility. Quarterly Journal of Speech, 54 59 63. Whitney, D. C. (1986). The media and the people: Americans' experience with the news media : a fifty year review New York: Gannett Center for Media Studies, Columbia University. FoxSports.com Retrieved August 2012, from http://msn.foxsports.com/collegefootball/story/penn state football culture to blam e in scandal 071312 Zhu, Z., & Won, D. (2010). Buyer or browser? An analysis of sports fan behaviour online. International Journal of Sports Marketing & Sponsorship 11 (2), 124 139.
160 BIOGRAPHICAL SKETCH Sean Sadri is a n orthern California native who is extremely passionate about teaching and research in the field of journalism, and has been fortunate enough to gain valuable experience doing both at the Univers ity of Florida. He completed a Doctor of ollege of Journalism and and Communication at the University of California, Davis. His passi on for journalism began in high school where he started writing sports articles for his high school newspaper. This enthusiasm for journalism carried over to the UC Davis, where he started covering campus news, and eventually the arts, for the s student newspaper. Additionally, while pursuing his M.S., he worked in the sports department of two local television stations in Upstate New York. These experiences aided in his development and implementation of an undergraduate Sports Reporting course at the University of Florida. The course, which he taught for two semesters, was designed to develop skills in reading, writing, gathering information, and generating ideas in sports journalism.