Abstract
This study examined the Penn State sex abuse scandal by applying traditional crisis communication strategies, usually invoked by an organization, to the online communication of the university’s active stakeholders, sports fans. Previous findings suggest sports fans will act on behalf of an organization during a crisis. Yet, the tweets of Penn State fans showed that they turned on the university and placed their loyalty with Coach Joe Paterno. Furthermore, this study discovered that fans engaged in the ingratiation, reminder, and scapegoat strategies most befitting the typology offered by Coombs, empowering the active stakeholders in the process. Results of this case provide a warning for organizations that online fan-based crisis response may not always be enacted in their best interests.
At some point, all sports teams encounter crises; player/coach crime scandals, alleged National Collegiate Athletic Association (NCAA) violations, and troublesome contract negotiations pepper sports headlines each week. Yet, in 2011—a year with many high-profile college football scandals at The Ohio State University and University of Miami (FL)—no team faced a scandal with as many ramifications as Penn State University (PSU). On November 5, 2011, Jerry Sandusky, a former Penn State assistant football coach, was charged with sexually assaulting eight boys over a 15-year period. In the following weeks, two more boys alleged that Sandusky had also abused them. Penn State graduate assistant coach Mike McQueary corroborated these allegations, reportedly seeing Sandusky assault a young boy in the showers of the Penn State locker room. In June 2012, Sandusky was found guilty of 45 of the 48 charges filed against him for assaulting 10 boys. Following the guilty verdict, Sandusky was sentenced to at least 30 years in prison, which the judge stated was intended to put him in prison for the rest of his life (Smith, Loyd, & Curry, 2012).
This scandal called into question the Penn State athletic department and university administration, as the public sought to ascertain how much university officials knew about Sandusky’s actions while he was still employed at the university. In relatively short order, the allegations led to the firing of Penn State President Graham Spanier and Penn State head football coach, Joe Paterno, on November 9, 2011 (CNN Wire Staff, 2011). The media firestorm raged for months, as one of college football’s traditional powers and one of college football’s legendary coaches was at the epicenter of the largest scandal in the history of college sports (Wetzel, 2011).
In July 2012, the Freeh Report (Freeh, 2012), the result of an independent investigation launched by the Penn State Board of Trustees, indicated that individuals affiliated with PSU attempted to conceal facts about Sandusky’s actions “in order to avoid the consequences of bad publicity” (Freeh, 2012, p. 16). Thus, NCAA President Mark Emmert, in an unprecedented move, levied severe sanctions against PSU, including a 4-year bowl ban, loss of scholarships, and a US$60 million fine (Prisbell, 2012). The NCAA also vacated 111 of Joe Paterno’s wins, stripping him of his title as college football’s “winningest coach” (Prisbell, 2012). The unique characteristics of the Penn State scandal caused it to become one of the largest scandals in U.S. history, making the response to it—particularly in the age of social media—worthy of investigation.
A crisis, at its core, completely disrupts an organization (Coombs, 2012), which was certainly true for Penn State. The scandal involved the sexual abuse of children, leaving Penn State with a transgression that was universally held as a deep act of immorality. Other acts of immorality can be debated in society; child molestation is not typically viewed as one of those topics. Additionally, at the time, Penn State was one of four NCAA Division I programs from the major Bowl Championship Series conferences to have never been found guilty of a major NCAA rules violation. The Penn State football program’s reputation of being a “clean” program was irreparably tarnished, narrowly avoiding the NCAA’s harshest punishment, the death penalty (Prisbell, 2012). Joe Paterno was considered to be above reproach (ESPN.com news services, 2012), achieving the all-time NCAA victories record (before it was vacated in July 2012), while doing much good for the State College community and beyond; yet, the information surrounding his involvement and relative lack of action made him a key player in the scandal.
Past studies have examined how different communication strategies employed by organizations have excelled or faltered during times of crises (Benoit, 1997; Benson, 1988, Coombs, 2007b). Several studies have analyzed crisis communication from an organizational perspective in sports, yet only one has interrogated how fans are now included in the strategy (Brown & Billings, 2013) as social media outlets allow fans to potentially adopt and implement these same strategies in hopes of repairing the reputation of their team and school.
Fan involvement was featured at every phase of the response to news of the scandal. They held candlelight vigils supporting the victims and rioted in the streets of State College after receiving the news of Joe Paterno’s firing. Therefore, fans also utilized a medium widely available to them—Twitter—to vocalize their support of their university. This study examines the social media–based online response, illustrating how fans effectively become an unsolicited and unsanctioned—yet powerful—arm of a university’s crisis response strategy. A content analysis of Penn State fans’ Twitter postings identifies how fans employ traditional crisis communication strategies and also react to their university’s crisis response. In doing so, findings illuminate how growing use and impact of social media among sports fans lead to the inclusion of a school’s fan base as an outlet for crisis communication response.
Literature Review
Crisis Communication Theories
The large impact crises can have on organizational reputations has caused crisis communication strategies to be the subject of many previous studies. Some of these interrogations evaluate the effectiveness of an organization’s crisis response plan and its subsequent impact (Jeong, 2009; Olaniran, Scholl, Williams, & Boyer, 2011; Williams & Treadaway, 1992). Other scholars have addressed how an organization’s crisis response has directly impacted organizational reputation (Coombs & Holladay, 1996; Lyon & Cameron, 2004). Yet, despite the many different approaches scholars have adopted to study crisis communication, no universal definition of crisis exists. For the purposes of this study, crisis is defined as the “perception of an unpredictable event that threatens important expectancies of stakeholders and can seriously impact an organization’s performance and generate negative outcomes” (Coombs, 2012, p. 2).
Communication occurring after a crisis can aid in repairing or reducing the damage to an organization’s reputation, as it is imperative for organizations to influence perceptions of their stakeholders, defined as “individuals, groups, organizations and communities who can both affect and be affected by the incident” (Pearson, Clair, Misra, & Mitroff, 1997, p. 58). Since stakeholders are integral to an organization, when they perceive that an organization is encountering a crisis, a crisis indeed exists. Thus, strategies must be employed to address stakeholder concerns and resolve the crisis in a manner that will reduce the damage caused by the crisis to repair the organization’s reputation.
A key component of the crisis communication process is the crisis response: What an organization says and does during the crisis situation (Coombs, 2007a). An organization can invoke a number of communication response strategies to manage not only the outcome of the crisis but also the impact on the organization’s key stakeholders and the public’s perception of the crisis (Coombs, 2012). Crisis response strategies must be used to protect the public’s opinion of the organization’s reputation, minimizing the damage caused by the crisis. There are two prominent typologies of crisis response strategies organizations can utilize that have been outlined by scholars, Benoit’s (1995, 1997) image repair typology and Coombs’s (2007a) reputation repair strategies.
First, regarding Benoit’s (1995) typology, two key assumptions guide his image repair theory (IRT): (1) communication is a goal-directed activity and (2) maintaining a favorable image is one of those goals. Based on a synthesis of previous literature rooted in apologia (Ware & Linkugel, 1973), self-defense rhetoric and rhetorical response to guilt, Benoit’s (1995) typology stated five strategies that accused parties use when their reputation is under attack: (a) denial, (b) evading of responsibility, (c) reducing the offensiveness of the event, (d) corrective action, and (e) mortification, with additional strategies further extrapolated from the first three strategies. Table 1 provides a complete list of these image repair strategies.
Benoit’s (1995, 1997) Response Strategies Based on Image Repair Theory.
In contrast, Coombs’s (2007a) reputation repair typology stems from the development of his situational crisis communication theory (SCCT). According to SCCT, in order to choose the correct response to a crisis situation, a crisis manager must first assess the initial degree of crisis responsibility placed on the organization. Then, based on the state of the organization’s crisis history and relationship with its stakeholders, the degree of responsibility may increase. After the crisis and relationship history is determined, the crisis manager chooses the response based on the refined level of responsibility (Coombs, 2007b). SCCT reputation repair strategy typology was created through a synthesis of previous response typologies, including Benoit’s (1995) IRT typology. Coombs (2007a) nine reputation repair strategies are clustered into four groups of similar strategies, later dubbed postures. The denial posture strives to remove any connections an organization has with a crisis. These include attacking the accuser, simple denial, and scapegoating. The diminishment posture attempts to reduce attributions of organizational control and reduce negative effects of the crisis. These include excusing and justification. The rebuilding posture attempts to improve the organization’s reputation. These include compensation and apology. The bolstering posture seeks to build a positive connection between an organization and its publics. These include reminding and ingratiation. It is recommended that the bolstering strategies be used only in conjunction with other strategies to avoid a superficial perception placed on the organization because of the strategies’ focus on the organization (Coombs, 2007a). Table 2 shows the crisis response typology for SCCT.
SCCT Crisis Response Typology, by Posture (Coombs, 2007b).
There are several differences between IRT and SCCT. First, IRT has been used primarily for individual-actor crises. A content analysis revealed that out of 11 articles focusing on individual-actor crises, 10 employed IRT and the other employed a combination of IRT and SCCT (Avery, Lariscy, Kim, & Hocke, 2010). SCCT, however, was created to examine organizational-actor crises, primarily in corporate settings (Coombs, 2010; Fussell-Sisco, Collins, & Zoch, 2010).
Additionally, SCCT uses attribution theory (Heider, 1958) as a theoretical link to match crisis types to crisis responses, whereas IRT does not employ a matching strategy (Sheldon, 2006). Benoit (1995) did not provide a theoretical link for matching crisis types with crisis responses because the theory focuses on analysis of crises from a source-oriented perspective. Previous research in crisis communication has exhaustively studied the source-oriented point of view, examining how the organization facing a crisis attempts to restore its image (Burns & Bruner, 2000). Few studies have used IRT to examine crises from the audience-oriented point of view, defined as measuring how the audience response to an organization’s attempt of crisis response can influence overall crisis management (Brown, Dickhaus, & Long, 2012; Coombs & Schmidt, 2000). These types of audience-oriented, experimental studies are the foundation of the research that uses and evaluates SCCT.
The audience-oriented perspective of crisis response research is an important focus for future studies, because additional power has been given to audiences through the rise of social media. Thus, it is timely to study a new type of audience-enacted crisis response, where fans engage in unsolicited crisis response strategies on behalf of an organization with which they identify. Previously, research only examined response strategies provided by the organization during times of crisis via social media (i.e., Shultz, Utz, & Goritz, 2011; Utz, Schultz, & Glocka, 2012). Given the presence of two bodies of crisis response strategies, it is important to determine which is better suited to explore this new phenomenon. Therefore, the first research question is posed:
Online Crisis Communication
The growth and evolution of communication technology should cause organizations to refine and rethink reputation repair strategies. The increased presence of social media and other online communication cannot be ignored in crisis response, as it provides a venue in which a crisis can be influenced by anyone at any time in any manner (Brown & Billings, 2013; Jin & Fisher Liu, 2010). While Row (2005) asserts that “blogs don’t change the fundamental nature of a crisis,” he did note that they change the “basics of how we can respond to them” (p. 1). Coombs (2012) noted that, for a variety of reasons, the Internet increased the need for organizations to engage in effective crisis management, as the Internet and social media led audiences to demand information be delivered virtually instantaneously, increasing the pressure on crisis managers (Malone & Coombs, 2009). In addition, the Internet allows online communities to form and become groups that can either hinder organizational attempts at reputation repair or become a part of the reputation repair strategy.
It is certainly true that many organizations have a social media presence, yet most of a person’s social media connections are individuals. Social media makes the community a part of a crisis communication response via citizen journalism as audiences are no longer passive recipients of information that traditional media, like television, radio, and newspapers wish to disseminate (Veil, Buehner, & Palenchar, 2011). Instead, the Internet has transformed people into active information seekers who desire instant information (Stephens & Malone, 2009). Jin and Fisher Liu (2010) explored the role of blogs in crisis communication, affirming that stakeholders attend to influential blogs for three main reasons: issue relevance, information seeking and sharing, and emotional venting and support. Yet, Jin and Fisher Liu note that blogs have the ability to indirectly impact large groups of people through offline word of mouth and potential media coverage.
The same trend can be observed in the microblogging website, Twitter, as traditional media outlets such as CNN and ESPN often integrate tweets into their broadcast, showcasing viewers’ instant reactions and opinions regarding the day’s news stories (Twitter, 2012). This media practice amplifies social media’s power (Jin & Fisher Liu, 2010) and allows active users a unique venue to broadcast their talking points immediately when crises occur and when they continue to develop. In fact, Shultz, Utz, and Goritz (2011) stressed that it is important to address Twitter users during times of crisis because they are more likely to quickly share messages and links to news and information.
Previous studies have identified information seeking as one of the primary motivations for social media use (Lee & Ma, 2012). In an era of online communication, stakeholders have an increased desire to obtain information as soon as it is available (Coombs, 2012). A 2011 Pew Research study showed that 65% of online adults use social networking websites such as Facebook or Twitter (Madden & Zickuhr, 2011); 2 years later, that number is likely substantially higher. Thus, if stakeholders are seeking information through online social media websites, it is important to determine the types of crisis response strategies they are more likely to read online and how the desire for immediate information impacts the time frame during which such strategies are used. It is also critical to determine whether the messages posted online by the stakeholders of an organization can act on behalf of the company or organization itself—in this case a university with which they heavily identify. Coombs (2007a) noted that the Internet has led to increased activity among stakeholders and a desire for the immediate release of information. Coombs suggested these trends could lead to the development of more crises for organizations; however, he did not consider the degree in which this increased activation could benefit an organization. Given the social media users’ desire for instant information and the noted responsiveness of social media use, a second research question is proffered:
Sport as a Venue for Crisis Communication
Crises can result from a variety of situations in sport, including during in-game competition, off-field situations, or broader reflections about general issues in sport such as race and gender (Billings, Butterworth, & Turman, 2011). When these situations occur, reputation repair strategies must be utilized. This area has received increased interest in recent scholarship largely because of the prominence of sports issues in the media and the “increased activism of sports fans” (Brown et al., 2012, p. 155). Several studies employed Benoit’s (1995) approach to image repair within sport (Benoit & Hanczor, 1994; Fortunato, 2008; Len-Rios, 2010; Sanderson, 2008). Benoit and Hanczor (1994) conducted one of the initial sports image repair studies as they examined Tonya Harding’s defense of her image on Eye-to-Eye with Connie Chung after her involvement in an attack on her rival, Nancy Kerrigan. Brown, Dickhaus, and Long (2012) examined the negative backlash LeBron James received from the media and fans after announcing that he would leave the Cleveland Cavaliers and sign with the Miami Heat on his televised special, The Decision. Glantz (2010) used a similar method to evaluate 2006 Tour de France winner Floyd Landis’s attempt at image restoration following charges of illegal substance use. Sanderson (2008) analyzed Curt Schilling’s online crisis response through his blog, 38pitches.com. This study showed the growing importance of online communication in crisis response.
The scandal occurring at Penn State is so anomalous that reputation repair techniques should be (re)evaluated.[Per style word constructs such as (re)evaluated are not allowed, Please check and edit as necessary.] While this study does not seek to evaluate the organizational response to the allegations against Jerry Sandusky or Penn State’s involvement with the former coach, it does seek to establish this situation as a crisis worthy of study, greatly impacting the university’s fans and alumni. Coombs (2012) outlined four characteristics of a crisis, noting that a crisis is simultaneously: (a) perceptual, (b) unpredictable, (b) violating stakeholders’ expectations of how an organization should act, and (d) holding the potential for a serious, negative impact. The Penn State situation fits each of these characteristics of a crisis. When news of the allegations against Jerry Sandusky was released, it caught the public largely by surprise (Snyder, 2011) as the 48 counts of child abuse in which he was eventually charged occurred over a 15-year period. Reports showed that Penn State officials were notified of incidents that occurred between Sandusky and a child in the athletic facilities. Yet, officials failed to immediately and permanently remove Sandusky from his position in the university, causing public outrage. Given Penn State’s reputation as one of the cleanest programs in the country, this situation violated stakeholders’ expectations of how PSU should act. Also, this situation did not simply hold a potential for a serious, negative impact; rather, the negative consequences began almost immediately with constant media attention surrounding and criticizing Penn State’s response.
Previous studies have analyzed different crisis communication responses in the context of sports (Brown et al., 2012; Glantz, 2010); however, fan impact must be added as a central component to the equation. Brown and Billings (2013) was the first study to explore online audience enacted crisis response, analyzing University of Miami fans’ responses to NCAA allegations against their football program. The study found that sports fans acted on behalf of the university with which they identified as their messages employed the traditional crisis response strategies of ingratiation, reminder, and attack the accuser. Yet, this study focused on a very specific type of crisis, NCAA allegations, and the authors called for future studies to apply these strategies to different types of crises.
When the news broke regarding the Penn State scandal, fans sought different outlets to express their thoughts. When an event receives the level of media attention of the Penn State scandal, it should be expected that all associated with the university would seek to express themselves. Sanderson (2011) notes that “social media is a viable mechanism for people to display their fandom” (p. 72), attributing this social nature of sports as a reason why many fans gravitate to social media. Sports fans have made their presence on Twitter known, as a majority of Twitter’s record traffic/tweets occur during live sporting events (Gregory, 2009). For example, Olivarez-Giles (2012) reports that during Super Bowl XLVI between the New England Patriots and New York Giants, Twitter users set a record of 12,233 tweets per second as they offered their reactions to the game.
Twitter has become an easily accessible outlet for sports fans to express the connection they feel to the team with which they identify (see Wann, 2006). Thus, heavily identified fans become stakeholders of their respective team with an inherent interest in protecting the team’s reputation. For the purposes of this study, it is crucial to define who constitutes an “active stakeholder.” Active stakeholders are clearly highly identified with their respective sporting organization. Consistent with previous fan identification literature (Wann, 2006), highly identified fans will often display the connection they feel with their team for others to witness. Thus, an online active stakeholder will visibly identify with their respective team. Finally, in order to be classified as an online active stakeholder, a person must be willing to speak out publicly about the crisis facing their organization, thereby, acting on the connection felt with their team.
The very unique nature of the Penn State scandal makes evaluating fan responses intriguing. This situation was unlike any other crisis that has been seen in collegiate athletics as it took place away from the field of play, impacting the lives of innocent children while creating a strong form of cognitive dissonance for fans. Fans feel a connection to their team and university and coach Joe Paterno; however, one could not deny that the allegations against Jerry Sandusky were horrific. Fans had to reconcile the connection they felt with their historic university and coach with the revulsion they experienced as each new allegation surfaced. The Penn State crisis involved more than rules violations; rather, it conjured an issue of morality. In fact, the Pennsylvania police commissioner charged that Joe Paterno did not perform his “moral responsibility” by failing to report Jerry Sandusky’s actions to the police (Grimm, 2011, p. 1). This cognitive dissonance makes fan reactions to the Penn State scandal compelling in determining how fans utilize traditional reputation repair strategies online when the crisis involves more than the team itself. Thus, the final research question deals with a compare and contrast between the one prior examination (Brown & Billings, 2013) dealing with NCAA violations and the current case involving a much different set of circumstances:
Method
Universe of Investigation and Sampling Technique
The data used in this study were gathered using a content analysis of tweets commenting on the child abuse scandal involving former PSU assistant coach Jerry Sandusky. Tweets were collected from November 5, 2011 (when the story broke) until January 21, 2012 (the day before Joe Paterno died), a timeline ensuring that tweets were reacting to the scandal rather than Joe Paterno’s death, which drastically impacted the storyline while offering a new narrative not within the current purview of this study. Tweets posted by PSU fans were the unit of analysis used in this study to determine how fans that strongly identify with PSU reacted to the news of the scandal involving Jerry Sandusky. To find these highly identified fans, the Twitter search engine Twazzup was used. Twazzup identifies key Twitter users who actively tweet about a given key word, denoting them as “influencers” on that specific topic. The key words “Penn State,” “PSU,” and “Nittany Lions” were used to find Twitter users whose tweets often included one of these key words. Once a list was generated, only Twitter users who self-identified as being Penn State football fans in either their “biography” section or through a Penn State athletic logo on their Twitter avatar were used in this study. The authors used this process to ensure that all tweets analyzed in this study were from self-proclaimed Penn State stakeholders who tweeted about Penn State athletics, thus, classifying them as Penn State fans. A total of 150 Penn State fans were selected for the analysis, and these fans had a combined total of 68,896 people who “followed” their tweets. All tweets these users posted from November 5, 2011, to January 20, 2012, were collected, totaling over 41,000 tweets.
First, tweets that did not include any mention of Penn State were immediately eliminated from the sample. This left each fan posting an average of 76 tweets about Penn State during the time span from November 5, 2011, to January 20, 2012, lending more evidence to their identity as a highly identified Penn State fan. The largest number of tweets posted by an individual fan was 253, and the fewest number of tweets posted was 10. After the tweets were selected based on relevance, the sample yielded nearly 13,000 tweets. Thus, a systematic sample was conducted by selecting every sixth tweet. As the researcher systematically selected tweets to be included in the sample, some additional tweets were eliminated if they were discussing certain unrelated aspects to Penn State that did not fit the nature of this study (e.g., course praises and complaints by students, restaurant and business location descriptions, etc.). This process resulted in a final sample that included 2,000 tweets.
Coding Instrument and Procedure
A coding instrument was used to analyze each tweet in the sample. To accurately measure which typology would be best applied to social media fan-enacted crisis communication, 20% of the sample was coded using both Benoit and Coombs’s strategies, with the strategy that encompassed the most number of tweets becoming the de facto primary coding measure for the remainder of the sample. Since Coombs (2012) recognized that stakeholders perceive a crisis as being caused by either internal or external factors, some strategies of reputation repair were coded in a secondary manner for further analysis. Specifically, for tweets invoking the “scapegoat” strategy, the following categories were used to determine exactly who was being used as a scapegoat: PSU administration, Jerry Sandusky, Mike McQueary, sports media, or others.
Consistent with Brown and Billings (2013), tweets were coded for using general hashtags (e.g., #PennState, #PSU, #WeAre) and hashtags that were specifically related to the scandal (e.g., #psucharges and #WeStillAre). Tweets containing links, suggesting the occurrence of information sharing, were also coded. To grasp how the crisis response strategy changed over time, tweets were coded by the week in which they were posted, with a total of 11 weeks in the sample.
Two coders were trained to recognize each reputation repair strategy. To ensure intercoder reliability, both researchers selected and coded 20% of the tweets in the sample. The intercoder reliability was found using Cohen’s κ and was found to be “outstanding” reliability (Landis & Koch, 1977) (κ = .82; p < .001). After all tweets were coded into 1 of the 10 reputation repair strategies, the data were analyzed using the Statistical Package for the Social Sciences. Descriptive statistics were used to determine which crisis communication strategy was employed most often by Penn State fans.
Results
Research Question 1 sought to determine whether Benoit’s IRT (1995) or Coombs’s (2007a) reputation repair strategies was better suited as the theoretical lens with which to study the crisis techniques found in sports fans’ Twitter feeds. Through the coding process, it was determined that Benoit’s IRT categories of (a) denial, (b) evading responsibility, (c) reducing offensiveness, (d) corrective action, and (e) mortification were not sufficient for this application, as 29% of analyzed tweets could not be placed within Benoit’s list of image repair strategies, whereas just 1.6% of the analyzed tweets failed to fit Coombs’s list of nine reputation repair strategies. For example, one Twitter user posted, “I will NEVER be ashamed to be a #NittanyLion. WE ARE bigger than the actions of a few. WE ARE the #NittanyNation. WE ARE #PENNSTATE.” Another fan posted, “I believe this dark time is gunna bring everyone together even more. Adversity shows character and it’s time to show ours. #PSUNation.” Both of these tweets fell nicely within Coombs’s ingratiation strategy, as they were meant to rally the fan base together and unite them during a difficult time. However, these tweets were not classified through Benoit’s strategies. This tweet aligned most closely with Benoit’s “bolstering” strategy; however, this tweet could not be categorized as bolstering since it was not reminding others of positive events that occurred in the past. Thus, the Coombs’s (2007a) list of reputation repair strategies appeared most inclusive and flexible enough to be applied online, answering Research Question 1. Benoit’s list did not include a strategy similar to Coombs’s ingratiation strategy, meaning that there was no way to code tweets that were used to rally and unite the Penn State fan base.
Research Question 2 questioned how crisis response changed over time. Figure 1 shows how the number of tweets changed over the course of 11 weeks.
Number of tweets over time.
Without question, the immediacy and newsworthiness of the story are observed in the number of tweets posted during Week 1 as nearly 60% of all tweets about the Penn State scandal were posted in the first week after the allegations were reported. The numbers then steadily dropped, with a spike in Week 9, likely related to the hiring of Bill O’Brien to serve as Penn State’s new head football coach.
Figure 2 shows how the top three strategies employed by Penn State fans changed over time. In the initial weeks following the allegations, scapegoating was the primary strategy used. However, subsequent week-by-week posting analysis found the reminder strategy was incorporated more frequently, likely because of the THON events being held. THON is a great source of pride for Penn State fans, as it is the largest student-led philanthropy established to fight pediatric cancer. In Week 9, the ingratiation strategy was employed often, likely to unite the fan base following the hire of new head coach Bill O’Brien. Since 60% of tweets were posted in the first week following the scandal, the answer to Research Question 2 lends support to Benoit’s (2007) postulation that technology, like social media, increases the speed with which information is spread, thus shortening the time for crisis response.
Change in strategies over time.
Research Question 3 queried how fan reaction to the Jerry Sandusky scandal differed from Miami fans’ reactions to NCAA allegations as described in the Brown and Billings’s (2013) study. Miami fans primarily responded to allegations of NCAA wrongdoing by employing three primary strategies: ingratiation, reminder, and attacking the accuser. In a similar vein, Penn State fans also utilized the ingratiation (27.1%, n = 541) and reminder strategies (22.3%, n = 445), but in a secondary function to a new primary strategy of scapegoating (39.2%, n = 783), evidenced in Table 1. Combined, these three strategies alone accounted for nearly 89% of the sample, diminishing the role of attack the accuser that was primarily used in the Miami scandal.
The largest strategy employed by Penn State fans in response to the scandal involved scapegoating. Given the breadth and nature of the incident, it would be nearly impossible for fans to use the “attack the accuser strategy” as Miami fans did. Rather, they needed another source with which to place their blame. In this case, four different subjects were offered as scapegoats: the Penn State administration (20.5%), sports media (8.2%), Jerry Sandusky (7.1%), and Mike McQueary (3.3%). Essentially, Penn State fans attacked two other sources more often than the alleged perpetrator, Jerry Sandusky, including their own institution. Many of the disgruntled tweets directed at Penn State officials related to the firing of head coach Joe Paterno, with 31.6% of total tweets mentioning his name. One fan called for the firing of Penn State President Graham Spanier by tweeting, “I am sorry to see a legacy tarnished. I believe Joe left to save the school’s face; Graham is staying to save his own. #grahammustgo.” Another fan noted that “Penn State had a 95-page report on the 1998 incident and the person overseeing the campus police didn’t even know that it existed.” Penn State fans also blasted their own administration for a lack of transparency during the crisis response but also blamed the sports media for their coverage of the Penn State scandal, especially the “riots” that broke out on campus following Joe Paterno’s firing. One fan tweeted, “The coverage of the riot and PSU’s demise was incredible. The coverage of tonight’s vigil and $200K raised for @RAINN01 was non-existent.” This person was just one of many arguing the media’s portrayal of Penn State following the allegations focused more on the negative events and fan reactions rather than the positive ones. Fans also felt that the sports media was demonized PSU as well as all those associated with the school. One fan tweeted, “How long will it be before SI and all news outlets put every PSU student and alumni on their front page? As if every1 is guilty. #PennState.”
Penn State fans also used the ingratiation strategy (30.2%) to garner support from the Penn State fan base. These tweets are meant to celebrate what it means to be a true Penn State fan. For example, one fan tweeted, “#PennStatePride #PennStateForever #pennstateproud #Westillare this simply is the best place in the country!” Fans also used this strategy to celebrate the positive manner in which many reacted to the scandal, as was the case when thousands held a candlelight vigil for the victims. One fan tweeted, “#PennState proved they are more than a few bad people … 10K at the vigil tonight, what a scene! #WeAre #PennStatePride #PennStateForever.” Penn State fans also replicated a strategy observed in Miami fans’ tweets by using special hashtags to unite the fan base. Nearly 32% of tweets contained some type of Penn State hashtag, and approximately 16% of tweets contained a specialized hashtag that was directly tied to the scandal such as “#WeStillAre” or “#psucharges.” Many fans also mentioned the words found in the Penn State alma mater, “may no act of ours bring shame” to rally the fan base. These words seemed especially poignant given the details of the scandal. For instance, one person tweeted, “May no act of ours bring shame … PSU, let’s stand by our own words. #WEstillARE and if we want to stand by JoePa, let’s do it the right way.”
The reminder strategy (25.1%, n = 445) was employed to reiterate to people Penn State’s good works and their storied football history. One of the activities Penn State fans touted was the student-led philanthropy THON. One fan tweeted, “Fair questions? What’s the most profitable activity at #PennState? Football. What’s the most important activity at #PennState? #THON.” Fans also bragged about the typical positive reputation boosted by Penn State. For example, one fan tweeted, “For your information, #PennState students are the number 1 employed grads in the nation. Fight on State!” This strategy is meant to refocus attention from the scandal and place it on facts that shed Penn State in a more positive light.
Given the nature of the scandal, the remaining strategies employed by Penn State fans were largely different than the ones employed by Miami fans. For instance, the allegations against Jerry Sandusky left little room for the denial strategy. By the time the news was released, a strong case has already been built against him. Instead, fans engaged in other strategies. In Miami’s case, there was no real way for fans to engage in compensation. However, Penn State fans banded together and raised over US$500,000 for victims of sexual abuse, leading to 32 tweets celebrating the money fans raised. One fan tweeted, “WE DID IT! #ProudPSUforRAINN exceeded our $550K goal. Our work isn’t done. We need to cont helping prevent abuse and supporting victims.” Tweets such as this show one way that Penn State fans worked together to give back to those who were hurt most by Jerry Sandusky’s actions. In addition to the compensation strategy, fans apologized to the victims through their tweets (n = 32). The use of these two strategies shows that there are cases in which fans will engage in Coombs’s most accommodative strategies.
The fourth most prominently employed strategy used by Penn State fans was victimage (n = 133). In this strategy, Penn State fans noted that the college was also a victim of Sandusky’s actions. The Sandusky scandal tarnished Penn State’s previously bright reputation. Fans noted that this scandal was caused by the actions of one man, yet the thousands of other people affiliated with PSU had done nothing wrong. For instance, one fan tweeted, “Sad and unfair. But also true: 409 wins wiped out by one pervert. #Paterno scandal.” The fans lamented that Sandusky’s actions harmed, not only the school’s reputation but also, Joe Paterno’s.
Perhaps the largest difference observed in the responses of Penn State and Miami fans was the connection Penn State fans felt with head coach Joe Paterno. His firing altered not only the storyline for these fans but it altered their allegiance. Paterno’s firing seemingly forced fans to choose a side between Paterno and Penn State, with most siding with their longtime coach. Some fans’ connections to Paterno were so strong they mentioned feeling like he was their “grandfather.” Nearly one third of all tweets in the sample mentioned Joe Paterno. For instance, one fan tweeted, “Joe Pa was the definition of Penn State. He was the one who REALLY bled blue and white.” Another fan tweeted, “With Joe Pa gone, what does Penn State stand for now? #sad #weloveJoePa.” These tweets show that Joe Paterno was so strongly tied to the brand of Penn State football that the fans could not decide how to separate the two. The idea of Penn State football without Joe Paterno at the helm frightened many fans. One fan tweeted, “Look at Bama after Bear left, ND after Lou left, OU after Barry left. All those teams struggled to win. The same will happen to #PennState.” Overall, the final research question is answered in that Penn State fans differed from Miami fans by using the scapegoat strategy instead of attacking the accuser, supporting their head coach (even at the university’s expense), and using repair strategies intended to be more accommodating toward the victims.
Discussion
This study provides many useful conclusions at the theoretical and applied crisis communication levels. First, the repair strategies outlined by Coombs (2007a) clearly fit the Penn State scandal more than the Benoit theory of image repair, at least in terms of the proportion of stakeholder tweets that could be placed into each strategy. However, questions persist as to why the Coombs model was the most appropriate match for this, as well as the similar sport institution crisis studied in the University of Miami case (Brown & Billings, 2013). One could surmise that Coombs’s work may more closely match both new media and social media models, yet another explanation could be that both of these sport-based Twitter studies dealt with institutions in crisis (as opposed to single individuals facing crises). Brown et al. (2012) argue that Benoit’s image repair strategies most befit individual athlete crisis responses, so perhaps the most appropriate division to interrogate in the future involves whether such divisions persist in social media where Benoit’s strategy would best be used for individual crises and Coombs’s strategy would best be used for larger organizational crises.
Moreover, the definition and role of a stakeholder is crucial to comprehending the contributions of this work. The case of Penn State offered a new approach to defining an organization’s stakeholders (one who can affect or is affected by a crisis, see Pearson et al., 1997) by introducing a new category of active stakeholders (in this case, people not only affected but willing to publicly comment about the crisis). Traditionally, crisis literature has not provided delineation between those who are affected by a crisis and those who wish to make an impact during crisis response. The social media–based shaping of the Penn State story was influenced predominantly by this smaller latter category of active stakeholders: people not a part of an organizational structure, who turn to social media outlets to help shape a story since mainstream media options are not available to them. While this study focused on the actions of active stakeholders, future research should explore differences between these two stakeholder groups.
Thus, when combining insights regarding the utility of both Coombs’s and Benoit’s typologies with this enhanced notion of the social media–defined active stakeholder, alteration of the original intent of each typology appears warranted. For instance, one major component to active stakeholder responses involved the ingratiation strategy—a component present in the Coombs typology but not offered in the Benoit typology. This single difference made Coombs’s strategy more befitting for both this study and the prior Brown and Billings (2013) analysis, representing a significant opportunity to expand/advance Benoit’s work.
Relatedly, Coombs’s work can be advanced as well in terms of primary and secondary levels and functions. It appears social media stakeholders are most likely to employ three strategies significantly more than all others: ingratiation, reminder, and scapegoating. All three of these macro-level categories could be usefully subdivided into ancillary categories, with the scapegoating category being an example of useful second-level classification in the present study. Penn State fans “attacked” the administration, media, and Jerry Sandusky by offering them up as scapegoats whose actions caused the situation, with the magnitudes of each of these scapegoating targets providing unique insight into the overall strategies Penn State stakeholders were using to change the direction of the story. Future work could benefit from the specific deconstruction of the major categories most likely to be employed by active stakeholders. Wann (2006) notes that when a team encounters a loss, highly identified fans are likely to showcase their knowledge by reminding others of previous team success, which is similar to Coombs’s (2007a) reminder strategy. In addition, heavily identified fans are likely to rally together as a fan base in support of their team and one another (Wann, 2006), which is similar to Coombs’s (2007) ingratiation strategy. This study shows that fans will also engage in this type of behavior when their team faces a crisis situation. Thus, this observed connection between crisis communication and fan identification literature must be further explored.
Moreover, one must ascertain whether online crisis communication responses are hybrids of past models or entirely new phenomenon for communication scholars to interpret. Certainly, there is evidence of the latter within the current study. For instance, previous conceptions of organizational crisis response involved advanced planning and systematic execution; neither is possible within the realm of a medium like Twitter. As shown in Figure 2, the moment a story breaks is also the moment in which active stakeholders begin to shape the message in the most abundant and immediate manner. Furthermore, while the Penn State case undoubtedly was going to be a page one story immediately after it became public, future research should examine not only whether social media can shape crisis responses but also pinpoint the extent to which it can cause a crisis in the first place. Current conceptions of social media (and the Internet more broadly) place outlets such as Twitter in a reactionary mode—a response to a mainstream media report or a response to the organization’s crisis response. However, subsequent analysis should explore the notion that social media can cause mainstream media to focus on a story, creating a crisis where it previously may not have existed.
Additionally, Coombs’s (2007a) SCCT is rooted in attribution theory, stating that a crisis leads to stakeholder judgments regarding where responsibility for an offending action lies. Coombs (2007a) also states that the more severe a crisis is, the more accommodating the repair strategies used by the organization must be. Without question, the case at hand was quite severe, and Penn State active stakeholders used the accommodating strategies of “compensation” and “apology,” something that was not present in social media responses to crises of lesser magnitudes. Thus, it is important to note that attribution is a key function of social media-based active stakeholders, making it a crucial area for additional exploration based on the severity of the transgression.
Moving to the level of applied contributions, highly identified fans directly associate their team’s on-field performance with their self-worth (Wann, 2006). One way fans express the connection they feel with their team is through posting team-related comments on social media websites. Thus, when child abuse allegations against former Penn State assistant coach Jerry Sandusky dominated the headlines in November, fans filled their Twitter feeds with comments on the situation. Previous research has shown that fans can act as an unsanctioned arm of an organization during crisis response. However, this was not the case with the Jerry Sandusky scandal. The Penn State scandal created a new wrinkle in the realm of audience-enacted crisis communication. While at times there is synergy between the organizational crisis response and the response offered by an organization’s active stakeholders, this situation showcased the potential for stakeholders to rebel against the organization itself when the interests of another entity or individual with which they identify are directly at odds with the organization’s interests. During times of crisis, fans pay close attention to their university’s crisis response, showing a willingness to respond online as each new development occurs. Thus, organizations must be aware that their official crisis response can have very powerful effects in an online environment.
Penn State fans offered the Penn State administration as a scapegoat, citing their lack of transparency and failure to turn in evidence against Sandusky as evidence. One interpretation could be that since fans generally tend to act in the best interest of the team with which they strongly identify, Penn State fans felt more connected to Joe Paterno than they did with Penn State as a whole. Perhaps such findings give credence to NCAA President Mark Emmert’s claim of a “warped athletic culture” and “hero worship” being present at Penn State (Prisbell, 2012). Another interpretation would be that fans immediately understood the gravity of the situation: The Sandusky crisis was so big that the football program was a lost cause, the institutional reputation would inevitably be tarnished, and, thus, fans opted to salvage the one component of the crisis they deemed retrievable: Joe Paterno’s legacy.
However, the Penn State administration still felt they had little choice but to terminate Paterno’s career at the university several days later in an attempt to separate itself from anyone with any prior knowledge of the allegations against Sandusky, enraging many Penn State’s active stakeholders in the process. Penn State’s administration could no longer hold a “football-first” attitude, but would, instead be forced to focus on “transforming the culture that permitted Sandusky’s behavior [to preserve athletic success and avoid bad publicity for the program]” (Freeh, 2012, p. 18). Joe Paterno possessed a god-like status among the Penn State faithful (Grollmus, 2011), the danger of which was revealed during the Penn State scandal as fans were forced to observe his fall as the NCAA eliminated his record as the all-time winningest college football coach (Prisbell, 2012). Paterno’s firing transformed the “Jerry Sandusky scandal” into the “Penn State scandal.” Despite the fact that Sandusky’s job had been terminated at Penn State many years prior to the reports, Penn State found itself at the center of the story, and unlike Miami fans during their school’s NCAA investigation, in this case, the fan base did not help the organization. Rather, the fan base hurled insults toward Penn State officials through their social media accounts. It was made clear that a line was drawn separating Penn State and Joe Paterno, and fans were siding with Paterno, drastically changing Penn State’s institutional crisis response. Thus, the concept of fan-enacted crisis communication has been broadened to represent more than fans simply heralding a cause for their university. Rather, the Penn State reactions illustrate a much more complex concept in which fan-enacted crisis response can become a detriment for the institutional crisis response.
The different levels of scapegoating provided interesting results in this study. Jerry Sandusky, the alleged perpetrator from which this scandal began, was the third most cited “scapegoat” with Penn State fans criticizing both the Penn State administration and the sports journalists more frequently than Sandusky. This is not to say active stakeholders were excusing Sandusky’s actions, but rather that they had recognized the relative certainty that Sandusky would be found guilty for his alleged crimes and that the debate now resided in the establishment of who else would be held responsible beyond the obvious target of Sandusky. Thus, they scapegoated the Penn State administration, passing judgment on many decisions that were made during the organization’s crisis response, especially Paterno’s firing. These results lend more support to the idea that stakeholders will attend to and cast their opinions regarding an organization’s official response during crises.
This study also provides many useful avenues for crisis communication research, as it is a useful heuristic for understanding the role of social media in overall image repair strategies. One fundamental academic need involves the necessary division of social media components, as Twitter, Facebook, and Pinterest, for instance, are all considered social media and yet function in very divergent manners. While Twitter seems to allow for the perfect marriage between journalistic and user-generated content, the popularity and potential of other social media websites cannot be ignored. Moreover, future studies should implore an experimental methodology to explore how these Twitter feeds impact social media users’ opinions of a university during a time of crisis. These studies should also take into account the joint lenses of traditional, new, social, and user-generated media to ascertain relationships between the manner in which people receive information regarding a crisis and attitude formation. Online fan responses should also be studied to evaluate the quality of organizational crisis responses. For instance, Penn State fans complained about the lack of transparency as Penn State responded to the allegations and the slow speed with which university officials would offer comments on the situation. Websites such as Twitter and Facebook could serve as an important evaluation tool for gauging organizational crisis response effectiveness.
Implications and Conclusion
Whether fan-enacted crisis communication supports or opposes the organization, it is critical for practitioners to monitor social media outlets during times of crisis to observe the reaction from an obviously active stakeholder group, whether or not the organization involved is a sports organization. It is apparent that these stakeholders will respond immediately to any crisis response disseminated by the organization, evident by the tremendous amount of tweets during the first week of news coverage of the Penn State scandal. This quick influx of stakeholder reaction could warrant a quick response from practitioners if the tone of social media responses opposes the organization because the responses of active stakeholders could influence other stakeholders to form an opinion. On the other hand, practitioners can use supportive social media postings to their advantage by using the statements from active stakeholders to supplement the organization’s response to the crisis situation.
This study applied traditional crisis communication strategies, often invoked by the organization themselves, to stakeholders of a university athletic department who possessed an active connection to the organization through their sports fandom. Unlike other findings suggesting sports fans will act on behalf of an organization during a crisis, Penn State fans turned on the university and placed their loyalty with their legendary coach, Joe Paterno. Furthermore, this study discovered that fans engaged in the ingratiation, reminder, and scapegoat strategies most befitting the typology offered by Coombs (2007a), empowering the active stakeholders in the process. However, in the case of Penn State, university officials were more frequently scapegoated than the actual alleged perpetrator of the crimes, Jerry Sandusky. Thus, this case provides a warning for organizations that online fan-based crisis response may not always be enacted in their best interests.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.