Intergroup discrimination is well established in the social psychological literature, where in-group members often hold pervasive biases favoring their in-group over outgroups (Balliet, Wu & De Dreu, 2014). Sport fans are no different and have been known to categorize themselves and others as in-groups and outgroups (Voci, 2006). Research suggests that sport fans often foster positive attitudes for their favorite teams and less favorable attitudes for opponents (Wenger & Brown, 2014; Wann & Grieve, 2005). This disfavor for opponents, and rivals more specifically, is based on both emotions and cognitions, and is accompanied by both conscious and unconscious processes (LeDoux, 1996).
Explicit attitudes involve intention and conscious awareness, whereas implicit attitudes can be activated automatically and guide behavior unconsciously (Dovidio, Kawakami, Johnson, Johnson, & Howard, 1997; Greenwald & Banaji, 1995). Recently, Wilson, Lindsey, and Schooler (2000) proposed a model of dual attitudes, defined as different evaluations of the same attitude, one implicit and the other explicit. This model allows for the coexistence of independent implicit and explicit attitudes toward the same attitude object (Perugini, 2005) and highlights the potential conflict between implicit and explicit systems (e.g., a negative implicit attitude is incongruent with a positive explicit attitude). Furthermore, this model considers that the attitude people endorse at any point in time depends on whether they have the cognitive capacity to retrieve the explicit attitude and whether the explicit attitude overrides the implicit one (Wilson, Lindsey, & Schooler, 2000). Specifically, motivated overriding allows individuals to be aware of both their positive and negative attitudes and to override the unwanted attitude with the wanted/acceptable one (Wilson, Lindsey, & Schooler, 2000).
This view is consistent with Greenwald and Banaji’s (1995) assumption that the evaluative content of an implicit attitude toward B may disagree with results from a direct measure of attitude toward B. The degree to which fans identify with their in-group influences both attitudes and behaviors. In a study by Hong, McDonald, Yoon, and Fujimoto (2005), the relationships between identification with the players and the teams, revealed that team identification is positively influenced by fans’ identification with the players.
Every year, players are traded, but what happens when a player from a rival outgroup team is traded to your favorite in-group team? Our negative attitudes towards the previously outgroup players may be held implicitly but, given team loyalty and cognitive resources, the more favorable explicit attitude toward this player may override our implicit bias. For example, highly identified fans, such as season ticket holders, view rival teams and players not only as an outgroup, but also as a threat to their identity due to the ongoing history of competition (Tyler & Cobbs, 2015). Therefore, there is good reason to believe that fans will demonstrate outgroup derogation, even after a player from a rival outgroup team is traded to your favorite in-group team. We plan to investigate this by targeting a well-known rivalry between the New York Rangers and the Metropolitan Division teams in the NHL. The purpose of this study is to consider both implicit and explicit attitudes of fans in the context of outgroup (rival) player trade acquisitions.
Specifically, we aim to find out whether fans form more positive implicit attitudes toward rival players turned in-group players over time. It is important to note that explicit attitudes change relatively easily, whereas our implicit attitudes take more time to change (Wilson, Lindsey, & Schooler, 2000). For that reason, a focus is placed on the effects of time and familiarity on fans’ attitude change both implicitly and explicitly after the short-term effects of the trade have dissipated.
We hypothesize: 1) participants’ responses will demonstrate an implicit favorable attitude for their favorite team and implicit negative attitude for a rival team; 2) there will be an inconsistency between the attitudes participants manifest implicitly and those they report explicitly as fans of the newly acquired rival players for the New York Rangers; 3) explicit attitudes (e.g., team loyalty) will override and take priority over any implicit attitudes of dislike, discrimination or rejection for the outgroup turned in-group players; 4) implicit attitudes will become more positive as fans become familiar with the player and accept them as part of the in-group team over time. It is important to note that these effects are expected only among highly identified fans (e.g.
, season ticket holders/fans with a strong psychological attachment to the team). We expect to find results that confirm these hypotheses. Moreover, this study will provide experimental evidence regarding how implicit and explicit attitudes change over time after teams acquire a new outgroup player and contribute to a gap in the literature on such fan experiences.MethodParticipants A total of 125 season ticket holders for the New York Rangers NHL hockey team will be recruited to participate in the study. Of these participants, we expect 100 to be eligible for the study and agree to participate in, and complete all procedures. The prerequisite for participation is that season ticket holders had to have attended at least half of the New York Rangers home games the previous season (n=20). Participants will be recruited each time the New York Rangers acquire a rival player from a Metropolitan division opponent (e.g.
, Columbus, Washington, New Jersey, New York, Pittsburgh, Carolina and/or Philadelphia) at home games during the regular season (October 4th) until the trade deadline (February 26th). After providing researchers with consent and an email address for contact, participants will be sent study measures through email at each one of the identified time points. Design and Procedures The study will take place entirely online and follow a longitudinal design where fans will be assessed at several time points including: immediately after the outgroup/rival player is acquired by the in-group team (New York Rangers) and 6, 12, 18, and 24 months post trade/signing. All measures will be completed online at the participants convenience through an email address the participant provides. Fans should expect each assessment point to take about 20 minutes to complete. Participants will be asked to respond to items about their fan behavior during the New York Rangers season and their attitudes toward the rival trade/signing under consideration. Implicit measure (Table 1). Participants will be asked to complete the Implicit Association Test (IAT) programmed for the purpose of the study (e.
g., related to their favorite and rival teams) to assess their attitudes for these teams and key players. Using an IAT, participants will respond as quickly as possible to concepts related to two rival hockey teams (New York Rangers and a Metropolitan division opponent) and concepts that were positive or negative. Participants will respond to concepts from four categories: good, bad, New York Rangers, and Metropolitan Rival.
The “good” words are: loyal, family, happy, nice, and trust. The “bad” words are: stink, dishonest, bully, cheat, and mean. The New York Ranger words are: blueshirts, Henrik, bleed-blue, Rangerstown, Matteau. The Metropolitan Rival words are: Columbus, Devils, Penguins, Broad-street-bullies, Caps. Participants will complete five blocks of trials. Table 1 shows the sequence of the blocks and trials in the IAT. Three blocks are practice blocks and two blocks are critical blocks for the analyses (blocks 3 and 5). In the third block, the category labels are ‘New York Rangers/good versus ‘Metropolitan division team/bad’ (congruent condition); in the fifth block, the labels are ‘Metropolitan division team/good and ‘New York Rangers/bad’ (incongruent condition).
Participants will be informed that stimuli that are typically “good” and “bad” will appear on the screen. They will then be instructed to correctly classify the stimuli that appear in the middle of the screen into distinct categories by pressing one of two keys on the left (“A” key) or right (“L” key) side of the keyboard. Explicit measure. After completing the IAT, participants will respond to a survey to evaluate their explicit attitudes toward veteran New York Rangers players and newly acquired players from rival teams.
These evaluations will take about 5 minutes. First, participants will be asked to rate “How committed are you to liking and supporting the New York Rangers?” and “How committed are you to liking and supporting a Metropolitan division rival?” Ratings are given on 5-point scales (1: No commitment, 3: Medium commitment, 5: Extreme commitment). Then, participants will complete a “feelings thermometer” to indicate how favorably or unfavorably they feel toward the veteran New York Rangers players and the newly acquired/previous rival team member. The thermometer will be scaled from 1–100 degrees with 1 labeled extremely cold or unfavorable/hated, 50 labeled neutral or indifferent, and 100 labeled extremely warm or favorable/likeable. These ratings will be particularly useful because they will provide a contrast for the explicit favorability or negativity participants have for the two categories of teams/players being evaluated.Expected Results We expect to find participants’ responses to demonstrate an implicit favorable attitude for their favorite team and implicit negative attitude for a rival team. An implicit belief that the New York Rangers are more closely associated with “good” words (and Metropolitan division rivals less so) will be reflected in faster responses when stimuli are categorized under conditions where the classification task block is congruent with the implicit attitude as opposed to the incongruent condition. The response speed will indicate the strength of the associations between the categories.
To analyze this data, we will calculate the difference score (D score) used by Greenwald, Nosek, and Banaji (2003) to calculate the difference in reaction times for the two critical blocks. To do this, the congruent condition – incongruent condition is divided by the pooled standard deviation of the response latencies (Greenwald, Nosek, & Banaji, 2003). The D score is then interpreted as follows: a positive (negative) score means that the participants responded faster (slower) when asked to group the New York Rangers with the “good” words relating to fan attitudes than for the Metropolitan division team rivals (Koenigstorfer & Groeppel-Klein, 2012), and we expect to see positive D scores for participants. Thresholds for ‘slight’ (.15), ‘moderate’ (.
35) and ‘strong’ (.65) will be selected according to psychological conventions for effect size and we expect to see a strong strength of implicit preference. Therefore, the results of the analysis of the reaction times will reveal that the 100 participants have a stronger implicit association between the New York Rangers and “good” words (congruent condition) compared to Metropolitan division teams (incongruent condition). Additionally, the response latencies will be significantly lower when the stimuli are presented in the congruent conditions than when presented in the incongruent conditions, which supports hypothesis 1. However, by the 24-month post trade assessment, we expect to see more negative scores (slower sorting) and implicit attitudes becoming more positive toward Metropolitan division teams as fans become familiar with the player(s) from those rival teams and accept them as part of the in-group team over time, which supports hypothesis 4. We also expect there to be an inconsistency initially between the attitudes participants manifest implicitly and those they report explicitly toward the newly acquired rival players for the New York Rangers. These inconsistencies will decrease and eventually dissipate over time. Explicit attitudes such as team loyalty will motivationally override any implicit attitudes of dislike, discrimination or rejection for the outgroup turned in-group members.
We will calculate a feelings thermometer difference score by subtracting the two thermometer items. The Likert items and the thermometer difference score will then be combined into a total explicit measure score by standardizing each and averaging the two resulting scores. Additionally, two analyses will be run. First, a correlation will be conducted for all participants using their total explicit measure score as the explicit measure and their congruent and incongruent reaction time scores from the IAT using an alpha level of 0.
05 as statistically significant. Significant interactions will be observed between both the reaction time scores and explicit total score. Second, an ANOVA will be run for participants who reported a clear commitment to either the New York Rangers or a Metropolitan division team. When rating their commitment to liking and supporting the New York Rangers or Metropolitan division team, participants who indicate scores between 1 and 3 points are categorized as low commitment and between 4 and 5 points as high commitment.
This will be a 2 (favorite team) × 2 (level of commitment) between-participants design, where the dependent measure is the implicit New York Rangers- “good” words preference score (congruent condition). Here, the interaction between favorite team and level of commitment will prove to be statistically significant, such that the difference in implicit preference scores is smaller for fans with higher explicit commitment. These results support hypotheses 2 and 3 and demonstrate that fans’ implicit attitudes are inconsistent with their explicit attitudes, which are more favorable.Discussion We expect to find participants’ responses to display an implicit favorable attitude for their favorite team and an implicit negative attitude for a rival team as demonstrated by the IAT. More specifically, a positive D score and strong effect size will be reported for the congruent condition.
By the final assessment, implicit attitudes will become more positive toward rival teams as fans are exposed to and become more familiar with the previously outgroup, turned in-group players. Further, there will be an inconsistency between the attitudes participants manifest implicitly and those they report explicitly toward the newly acquired rival players. This inconsistency is reflected in the significant interaction between favorite team and level of commitment in the ANOVA. Finally, a correlation demonstrates that explicit attitudes override implicit attitudes. To date, fan evaluations of an outgroup (rival) player trade to an in-group favorite team have not be investigated. Almost all of the previous research has evaluated team and player commitment using explicit, self-report measures and neglected implicit measures.
More specifically, previous experiments have not acknowledged the dual attitudes that sports fans may hold and how motivational overriding can account for the favorable explicit attitudes portrayed and the inconsistencies between implicit and explicit attitudes more generally. For example, participants almost always report explicitly that they hold favorable attitudes toward members of their favorite team, even when some members are newly acquired from a rival team. However,when they are evaluated implicitly, their attitudes toward these team members tend to be less favorable. The focus of the current experiment would help explain the role that outgroups play in in-group discrimination, and more specifically, how our implicit and explicit attitudes are effected by such player acquisitions. Using sport fans (e.g., season ticket holders) elucidates how in-group favoritism interacts with outgroup derogation to create attitudinal biases that are reflected in both implicit and explicit measures.
Moreover, this study will add to the intergroup discrimination literature and contribute to our understanding of attitudinal change in sports.