A case of multiple (brand) personalities: expanding the methods of brand personality measurement in sport team contexts.

Greenhalgh, Greg ; Dwyer, Brendan ; LeCrom, Carrie 等

Introduction

Like all products and services, sport organizations have BPs which shape the way consumers (fans) interact with the organization, attend events, purchase merchandise, and generally view different sport organizations (Braunstein & Ross, 2010; Heere, 2010; Ross, 2006). Within spectator sport, the BP of a team can impact the way fans of that team view their favorite team, as well as the way fans of other teams view the organization. Brand personality differences provide product differentiation opportunities for teams in a number of ways. For example, in a crowded college football marketplace, teams like the University of Oregon and Baylor University have used flashy uniforms and up-tempo offenses to create a "cutting-edge" brand for their respective programs (Staples, 2015). Likewise, teams such as Penn State and Wisconsin have utilized recruiting and playing style to create a brand that suggests "toughness" and a "blue-collar" mentality (Staples, 2015). The same can be said for professional sports. Over the course of several decades, the Dallas Cowboys and Oakland Raiders have created very distinct BPs for their organizations. The Cowboys, in particular, embraced the title of "America's Team" in the late 1970s and utilized Monday Night Football and Thanksgiving Day to reinforce a boastful, marquee, yet traditional BP (Carlson, Donovan, & Cumiskey, 2009; Mosely, Seifert, Walker, & Graham, 2009). Meanwhile, the Raiders utilized free agency, on-field bravado, and recklessness to create an "outlaw" image for their organization (Graham, 2014). Taken together, BP represents an opportunity for organizations to differentiate through a number of avenues such as uniforms, logos, media and marketing communication, playing style, or even scheduling. Sport managers now view their organization/team as a brand in need of management (Kang, Bennett, & Welty Peachey, 2016; Schade, Piehler, & Burmann, 2014).

The importance of measuring spectator sport BP has been reiterated by numerous researchers (c.f., Braunstein & Ross, 2010; Carlson et al., 2009; Heere, 2010; Ross, 2006, 2008) as sport organizations need to be aware of their perceived BP so they can more accurately manage their BP. However, current forms of measurement have faced significant criticism. BP measurement within spectator sport has generally taken two forms. A number of researchers (c.f., Braunstein & Ross, 2010; Carlson & Donovan, 2013; Carlson et al., 2009; Kim, Magnusen, & Kim, 2012; Ross, 2008; Smith, Graetz, & Westerbeek, 2006) have tried to apply (or adapt) scales developed to generically measure BP within a variety of industries (e.g., Aaker, 1997) to spectator sport. Consistently, these attempts have been met with reliability and validity issues (Braunstein & Ross, 2010). Other researchers have attempted to create their own spectator sport BP scale by going straight to the source: sport managers to create the BP adjectives subsequently sent to fans to assess the adjectives' applicability with fans' perceptions of the team's BP (Heere, 2010; Tsiotsou, 2012; Walsh et al., 2013). However, these attempts may lack generalizability as the scale created is only applicable to the sport organization under investigation. Another notable limitation of this strategy is the fact BP adjectives originated only from the administrators within the sport organization under investigation. One could argue that these administrators are likely to give adjectives symbolic of the BP they aspire to have or an ideal BP rather than a true evaluation of the current BP of their organization. In short, it is unlikely many sport managers/marketers would be willing to provide negative or even neutral adjectives to describe their brand (see Heere, 2010 and Walsh et al., 2013).

Finally, a significant gap exists within the current literature. To date, researchers have only looked at the BP perceptions of fans or spectators of a particular organization. Yet, to remain sustainable, sport organizations need to be able to attract spectators who may not consider themselves fans. For any brand two main segments exist: current consumers and non-consumers (Romaniuk, 2008). Spectators may select a small number of away games they will travel to and support their favorite team play, so the BP of the opposition may assist in deciding which away games they attend. Many fans are displaced and attend games of teams who are geographically convenient and the BP of the optional organizations could impact these decisions. Finally, non-mainstream or niche sports, such as the National Lacrosse League, aim to expand the popularity of their sport in general and are looking to engage with, or create fans of the sport in addition to fans of a particular team. However, to date, there is a paucity of research specifically examining the BP assessment of non-fans. The purpose of this study was to assess the efficacy of adjective collection from three different sources--administrators, fans, and non-fans--as well as evaluate the differences in perceived BP of an organization between fan and non-fan respondents.

Review of Literature

Brand Personality

A brand is a "mechanism for achieving a competitive advantage for firms, through differentiation. The adjectives that differentiate a brand provide the customer with satisfaction and benefits for which they are willing to pay" (Wood, 2000, p. 666). People naturally attach human characteristics to inanimate objects; this association of human characteristics to non-living items is termed anthropomorphism and has been found to exist with brands (Freling & Forbes, 2005; Kang et al., 2016). Consumers associate human characteristics such as kind, cool, interesting, strong, or dependable to companies and/or their brands (Aaker, 1997; Anandkumar & George, 2011; Avis, 2012; Carlson, Donavan, & Cumiskey, 2009; Freling & Forbes, 2005; Heere, 2010; Stinnett, Hardy, & Waters, 2012). Collectively, these characteristics make up what is termed the brand's "personality." Azoulay and Kapferer (2003) defined BP as the set of human personality states that are both applicable to and relevant for brands. In short, BP consists of adjectives used to describe brands (Carlson & Donovan, 2013).

Not only do consumers associate human personalities with brands, but consumers prefer brands with strong, favorable personalities (Freling & Forbes, 2005). Furthermore, consumers are able to express themselves by the products with which they associate via that brand's specific personality (Anandkumar & George, 2011). BP has been found to impact the ways in which consumers interact with the brand, specifically, how consumers seek information about the brand (Esch, Langer, Schmitt, & Geus, 2006), consumers' attitudes towards products (Aggarwal, 2004), and consumers' purchase intention (Aggarwal & McGill, 2007). Also, BP can be utilized as a way to differentiate a brand from its competitors as companies have found that differentiating from competitors based solely on functional aspects or price has become nearly impossible (Anandkumar & George, 2011) and BP is a controllable way to differentiate one's brand.

Importance of Brand Personality Measurement

A favorable and unique BP can create positive brand equity which in turn can lead to a competitive advantage and subsequent revenue generation via ticket sales and sponsorship (Ross, 2006). Enhancing the need for spectator sport organizations to create and shape a positive BP is the challenge of core product uncertainty, as sport marketers have no control over the on-field/on-court product that is intangible, subjective, and unpredictable (Ross, 2008). Many sport marketers are increasingly viewing their organization as a brand in need of management in an effort to build a positive brand image, which can positively impact the behavior of fans and other stakeholders, independent of their team's success on the court/field (Gladden & Funk, 2002; Schade et al., 2014). However, without empirical analyses of an organization's BP it is nearly impossible for the team to know where their brand sits within the minds of their consumers. Without the knowledge of their BP, marketers cannot effectively shape or change their BP.

The study of BP within the marketing domain dates back to the 1950s (Anandkumar & George, 2011) but the bulk of work came after Aaker's (1997) seminal work and her creation of the brand personality scale. Based on the Big Five personality dimensions, Aaker's 42-item scale was designed to assess brands in all industries. With nearly 6,000 citations, according to Google Scholar at the time of this writing, Aaker's work has been well utilized in the marketing literature base. The brand personality scale has been used to assess organizations in the tourism industry (Hosany, Ekinci, & Uysal, 2006), restaurant industry (Austin, Siguaw, & Mattila, 2003), automotive industry (Pappu, Quester, & Cooksey, 2005), financial services industry (Mizik & Jacobson, 2008), and the medical industry (Yoon, Gutchess, Feinber, & Polk, 2006) to name a few.

However, the brand personality scale created by Aaker (1997) has faced significant criticism. Azoulay and Kapferer (2003) argued that Aaker's BP scale is fundamentally flawed. While the term 'BP' is defined by Aaker (1997) as "the set of human characteristics associated with a brand" (p. 347). Azoulay and Kapferer (2003) noted the use of the term personality has a very specific psychological meaning, yet has been used as an "all-encompassing pot pourri"(p. 150) within the BP literature. The broad use of the term BP has caused researchers to include concepts beyond those of personality and brand them as BP, by using this loose definition may allow researchers to include almost everything related to a human being and apply it to brands (Azoulay & Kapferer, 2003). For this reason, Azoulay and Kapferer (2003) proposed that BP be defined as "the set of human personality traits that are both applicable and relevant for brand" (p. 151). These criticisms have been echoed within the sport marketing literature as well.

Challenges of Brand Personality Measurement in Spectator Sport

Narrowing the focus of brand personality assessment to the sport industry, Aaker's (1997) scale has been tested several times with limited success. In an attempt to assess the utility of Aaker's BP scale in an Australian sport setting, Smith et al. (2006), noted their findings "yielded a different factor structure to Aaker's original formulation" (p. 261). In 2008, Ross noted that his findings on the BP of a collegiate basketball team indicated that Aaker's (1997) brand personality scale was both invalid and not generalizable. Carlson et al. (2009) noted that Aaker's BP dimensions were developed to assess traditional, tangible brands, however sport organizations are intangible quasi-brand meaning the brand personality scale may not directly apply. Braunstein and Ross (2010), in the professional sport setting, reported "the reliability and validity of the [brand personality scale], in its current state, do not provide a sufficiently sound instrument" (p. 13). Heere (2010) also noted that Aaker's model has severe limitations. One of Heere's (2010) major contentions with the brand personality scale is that he believes, much like Azoulay and Kapferer (2003), that Aaker's perspective of BP is conceptually flawed as a brand cannot inherently have a personality, but rather a brand is given a set of traits by the marketing arm within the given brand. Within the sport setting, one could argue that fans of the organization may also shape the BP of their favorite team: think the outlaw nature of the Oakland Raiders.

Alternative Measures of Brand Personality

As an alternative to Aaker's approach of having a one scale fits all, Heere (2010) suggested managers of the organization under investigation create the list of BP adjectives to be measured. "This list of personality associations lies arguably at the foundation of the marketer's attempts to create and manipulate a BP for their consumers, and could be used to assess the effectiveness of the marketers to shape this image within the minds of their consumers" (Heere, 2010, p. 19). In his assessment of the BP of five netball teams in New Zealand, Heere (2010) determined the following adjectives were suitable for describing these teams: competitive, exciting, professional, dynamic, passionate, proud, accessible, warm, cool, and attractive. To get to this final list of appropriate BP adjectives, Heere first collected the BP adjectives sport managers of five netball teams felt best described the BP of their team. After filtering and aggregating the list of adjectives, Heere (2010) sent the list back to the managers of each team and they agreed this collective list sufficiently described the BP of their team. Spectators of each of the five teams then assessed the representativeness of each BP item in describing the team's game they were attending when they completed the survey. All 10 items were reported to be representative of the brands.

Walsh et al. (2013) followed the methodology suggested by Heere (2010) as they collected the adjectives from five event managers of a major NCAA Championship event. The adjectives determined to be representative of the event included: exciting, passionate, entertaining, intense, competitive, fan-friendly, skilled, elite, and fast-paced. Much like Heere (2010), Walsh et al. (2013) found the adjectives provided by the managers to be highly representative of the BP perceptions of fans when fans were asked to rate the representativeness of each BP item. As noted by Heere (2010) this method is likely more useful to sport managers as it provides a more accurate picture of their brand.

Theoretical Development

Given the theoretical and methodological suggestions of Braunstein and Ross (2010) and Heere (2010), the current study chose to adopt a stakeholder perspective of brand personality. A similar approach was taken by Scott and Lane (2000) to explain organizational identity, another fundamental attribute of an organization. Akin to BP, organizational identity is a collectively held frame from which employees, participants, leaders, partners, and customers derive meaning (Dutton, Dukerich, & Harquail, 1994; Weick, 1995). Scott and Lane (2000) chose to apply a stakeholder perspective, as managers do not solely set an organizational identity. According to the authors, managers play a central role in establishing an identity, but often times the identity lives in the mind of stakeholders; thus, the most accurate account of organizational identity is where managers and stakeholders meet.

Brand personality is no different. Brand personality in sport resides in the mind of both managers and fans (both affiliated and non-affiliated with the team). Braunstein and Ross (2010) found that Aaker's (1997) comprehensive brand personality instrument did not meet the needs of sport teams, and Heere (2010) suggested seeking administrator input in the development of brand personality items. The current authors propose a framework where administrators and stakeholders (in the form of fans and non-fans) each create brand personality items for a specific team to be tested on a target population of stakeholders.

Stakeholder theory suggests individuals identify and connect with organizations where there is an overlap with self-identification and the cognitive image of an organization (Freeman, 1984; Laplume, Sonpar, & Litz, 2008). The process of sport team attachment and identification is slightly different that in traditional goods and services, as it has been found that fans connect with several different points including, but not limited to the region/community, players, coaches, institution, style of play, or the sport itself (Robinson & Trail, 2005). In addition, the avidity and strength for which sport fans connect with teams has been found to exceed the levels for which traditional product and service consumers connect with an organization (Dwyer, Mudrick, Greenhalgh, LeCrom, & Drayer, 2015). This is despite the volatility within front office/administration, coaching, and player positions where organization/team philosophies and identities, including playing styles and success, change often.

Team BP represents a brand management function that is an opportunity for both administrators and stakeholders to potentially gain control and make sense of the attributes associated with the team. Administrators will inherently position the brand only in a positive light, where stakeholders will often be more objective. That is not to say stakeholders, especially highly identified fans and highly identified nonfans/fans of rival teams are not subjective, because they are. It is rather that a collective approach to item development will provide a more objective list for which to test on a larger sample of stakeholders.

While Heere's method in securing BP adjectives is likely more accurate and applicable than Aaker's approach, in the current study we aim to investigate if sport managers alone are the best source of adjective attainment. Are there other more appropriate methods and/or sources of attaining BP adjectives for sport organizations? Sport organizations need to be cognizant of the BP they exude beyond their own fan base, as non-fans are also consumers when they purchase tickets to their favorite team's away games or are simply looking for entertainment options. Fans must decide which of their favorite team's away game(s) they will attend and the BP of the team they are looking to visit may impact their decision. Therefore, if a sport organization conveys a BP of uninteresting, unsafe, or inferior they may risk the loss of potential ticket sales beyond their core fan group. Similar implications can be inferred for media rights and sponsorship funding. Insights into the BP perceptions of non-consumers can help to increase the customer base and replace consumer attrition (Romaniuk, 2008).

The following research questions were created to guide the study:

[RQ.sub.1]: Is there a difference in the BP perception of fans of a program versus non-fans of the same program?

[RQ.sub.2]: Is there a difference between the adjectives deemed most descriptive of the team and the adjectives deemed least descriptive of the team by fans?

[RQ.sub.2]: Is there a difference between the adjectives deemed most descriptive of the team and the adjectives deemed least descriptive of the team by non-fans?

[RQ.sub.4]: Is there a difference in the BP perception of fans of a program and non-fans of the same program based on the source of the adjectives (administrator, fans, or non-fans)?

Methods

The current study was set within a mid-Atlantic region of the USA. BP adjectives were collected from several administrators within one Division I university. Five administrators provided the adjectives they felt best described their men's basketball program.

Additionally, three fans of the team and seven fans of three separate opposing teams within the same State provided adjectives they deemed most fitting for the men's basketball program under investigation. This convenience sample of administrators, fans, and non-fans were selected based on their knowledge of the subject matter. Further collection of adjectives was terminated as we reached a point of repetitiveness. Adjectives provided by these three sources were reviewed by the research team and filtered to ensure all adjectives met Azoulay and Kapferer's (2003) definition of BP (the set of human personality states that are both applicable and relevant for brands), as well as to eliminate redundancy and alleviate confusion.

Instrument

In total, 42 unique adjectives (see Table 1) provided by the three different sources passed the screening of applicability, clarity, and redundancy. Adjectives were set to be randomized within the online survey instrument to help minimize the effects of participant fatigue. Respondents were asked to report their favorite men's basketball program. This response was used to identify fans versus non-fans for further analyses. Next, irrespective of the team reported to be their favorite, all respondents were then asked to assess how well each of the 42 adjectives described the college men's basketball program under investigation ranging from 1 strongly disagree to 7 strongly agree. Respondents' demographic information was also collected.

Sample

A link to the online survey, hosted by Qualtrics, was distributed via the Facebook pages and Twitter accounts of three different Division I collegiate athletic departments in the Mid-Atlantic region of the United States. This sampling method provided an opportunity to get responses from both fans and non-fans as well as assess the appropriateness of the adjectives provided from each of the three sources (administrators, fans, and non-fans). Those respondents who identified themselves as a fan of the program under investigation were classified as 'fans' and those respondents who identified any other program as their favorite college basketball program were coded as 'non-fan'.

Social media such as Facebook and Twitter have been utilized by numerous social scientists as an effective sampling strategy (McCormick, Lee, Cesare, Shojaie, & Spiro, 2015). This methodology has also been used in sport marketing research (e.g., Larkin, 2015; Ruihley, Billings, & Rae, 2014). The greatest limitation to this technique is the inability to procure a response rate as it is impossible to know how many individuals are exposed to the Facebook post or tweet containing the link to the survey. However, these sources allow for the collection of data from a heterogeneous, demographically diverse, geographically dispersed sample (Bhutta, 2012; McCormick et al., 2015). The benefits of social media sampling far outweigh the limitations (Bhutta, 2012; McCormick et al., 2015).

A total of 311 individuals from three mid-Atlantic college basketball fan bases evaluated how well they felt each of the 42 adjectives described the men's basketball program under investigation in the current study. The sample was primarily male (65.4%), Caucasian (89.3%), and highly educated with 76.5% attaining at least a bachelor's degree. The average age of the sample was 40 and the majority (57.8%) had an annual household income between $50,000 and $149,999 per year. As it relates to team preference, just over one-third (36.3%) of the sample considered themselves fans of the team under examination.

Procedures and Analyses

To gain a better understanding of the overall sentiment of the adjectives provided, the research team coded adjectives as being positive, negative, or neutral in connotation. Thirty adjectives were deemed positive, four were neutral, and eight were negative in sentiment. Administrators provided nine adjectives: all positive. Fans were responsible for providing 15 positive adjectives, three neutral adjectives, and four negative adjectives. Non-fans provided nine positive, two neutral, and five negative attributes. See Table 1 for a complete list of adjectives, their source, and sentiment. The current study further supported the notion presented by Azoulay and Kapferer (2003) and advanced by Heere that brands do not necessarily possess a personality, but rather a personality is given to a brand. Adjectives such as 'frustrating' provided by fans may be a glaring example of this phenomenon.

All analyses were conducted using SPSS 23. Research question one was assessed via a multivariate analysis of variance (MANOVA). To address [RQ.sub.1], respondents' status (fan or non-fan) was the independent variable and each of the 42 BP adjectives acted as the dependent variables.

Research questions two and three were addressed by identifying the adjectives with the 10 highest means, those which best described the program's BP and the 10 adjectives with the lowest means, those which did the worst job describing the program's BP. These were done for fans and non-fans independently. Furthermore, paired-samples t tests were run to determine if there were statistically significant differences between the top rated adjectives and bottom rated adjectives within the fans and non-fans groups independently.

To address [RQ.sub.4], respondents' status (fan or non-fan) acted as the independent variable while the source of the adjectives (administrators, fans, or non-fans) acted as the dependent variable. As noted above, all respondents assessed the applicability of each of the 42 adjectives to describe the BP of the men's basketball program. To assess the difference in adjectives provided by each source, three aggregate mean scores were created. The mean scores of the 21 adjectives provided by fans were averaged to create an aggregate fan adjective score. Similarly, the mean scores of the 16 adjectives provided by non-fans were averaged to create a non-fan adjective score. Finally, the means of the nine adjectives provided by administrators were averaged to create an administrator adjective score. These aggregate means represent the ability of each source to provide adjectives deemed appropriate by the respondents (fans and non-fans). Therefore, the source with the greatest average mean provided the collection of adjectives respondents (fans and non-fans) believed best described the BP of the program under investigation.

To address within group (fans and non-fans) differences between the three sources of adjectives (administrators, fans, and non-fans) three paired samples t tests were run on the responses provided by each of the groups (fans and non-fans). A total of six paired sample t-tests were run to assess differences between the adjectives provided by the administrators, fans, and non-fans as perceived by fan respondents and then by non-fan respondents separately.

Due to non-normal data (positively worded adjectives were typically negatively skewed and negatively worded adjectives were typically positively skewed) and unequal cell sizes data were transformed. First, all cases with missing data were removed. Next, all 114 remaining responses by fans were retained and 114 of the 197 non-fan responses were randomly selected using SPSS' Random sample of cases function. These final 228 cases were utilized for all subsequent analyses.

Results

A MANOVA was conducted to assess RQ1, the 42 BP adjectives were the dependent variables and the independent variable was the respondent's fan status (fan vs. non-fan). Prior to assessing the fan status on BP, a check of assumptions found Box's test of equality of covariance matrices was found to be significant (F(946, 153847) = 1.48, p < .01) indicating the assumption homogeneity of covariance matrices was not met. According to both Stevens (2002) and Field (2009) Box's test is very sensitive to non-normality. Field (2009) and Stevens (2002) both noted that MANOVA is quite robust to this violation so long as group sizes are approximately equal: cells in the current study were exactly equal. Results indicated a statistically significant multivariate difference F(43, 184) = 9.03, p < .001, between the fans of the program and the non-fan respondents on the BP adjectives. Post-hoc analysis revealed 41 of the 42 adjectives were reported to be statistically significantly different between fans and non-fans. The only adjective which was not viewed significantly differently between fans and non-fans was the adjective dangerous. See Table 1 for complete results.

To answer [RQ.sub.2] the 10 adjectives with the greatest means and the 10 adjectives with the lowest means, as reported by fans of the program under investigation, were identified. Those adjectives with the highest means represent the adjectives fans believed best described the BP of the organization. Conversely, the adjectives with the lowest means were the least representative of the organization's BP. Results of a paired-samples t tests indicated that each of the top 10 rated adjectives were statistically significantly greater (p <.001) than the 10 lowest rated adjectives. The same analysis was conducted to answer [RQ.sub.3] utilizing responses from non-fans only. Again, the top 10 adjectives were statistically significantly greater than the bottom 10 adjectives at the p <.001 level. See Table 2 for full results of [RQ.sub.2] and [RQ.sub.3].

A MANOVA was utilized to assess [RQ.sub.4] on the differences between the fan and non-fan respondents on adjectives provided from the three different sources. As noted above, aggregate adjective scores were created for each of the three sources and were included as the dependent variables. Adjectives provided by more than one source were included in each source for analysis (i.e., determined was proved by administrator and fans). Hence, the adjective determined was included in the aggregate score for both administrators and fans. Again, fan status acted as the independent variable. Results indicated a statistically significant multivariate difference between fans and non-fans with respect to the source of the adjectives F(3, 224) = 86.37, p < .001. A follow up post-hoc found that fans rated the adjectives provided by all three sources at a statistically significantly higher rate compared to non-fans. See Table 3 for the full post-hoc results. Furthermore, when investigating the within group differences of the fan respondents, statistically significant differences were discovered between the adjectives provided by administrators, fans, and non-fans. Fans reported the adjectives provided by the administrators were most representative of their perceived BP of the program under investigation, followed by the adjectives provided by fans, and the adjectives provided by non-fans were the least representative. However, within the non-fan respondents there were no statistically significant differences between the adjectives provided by any of the three sources (administrators, fans, and non-fans).

Discussion

As noted earlier, without empirical investigation of an organization's BP it is nearly impossible for the team to know where their brand fits in the minds of their consumers/potential consumers. Without this knowledge, marketers cannot effectively shape or change their organization's BP. The findings of the current study assist in advancing this burgeoning facet of the sport marketing literature base. The importance of understanding BP within the dynamic sport industry has been well established. The sport marketing literature base has evolved with respect to the way BP is measured. The transition from utilizing Aaker's (1997) brand personality scale, to the more organizationally-focused methodology provided by Heere (2010) was a significant advancement. Yet, results of the current study indicate BP measurement within the sport industry is still evolving. Specifically, the current study demonstrated a need to broaden the ways in which BP adjectives are collected prior to investigation, as well as the need to broaden the sample when investigating BP to include non-fans of the organization.

Findings from the current study paint two distinctly different pictures. Stark differences were found in the ability of the three different sources (fans, non-fans, and administrators) to provided adjectives closely resembling the perceived BP of their organization in the eyes of fan and non-fans. The differences were highlighted when looking at the lists provided by the different sources collectively versus looking at the most and least representative adjectives as perceived by fans and non-fans individually (See Table 2).

When the responses from fans and non-fans were assessed for all 42 adjectives, there were clear distinctions between the way these two groups viewed the BP of the same organization. Fans and non-fans within the current study were found to differ in several ways with respect to their BP perceptions of the organization under investigation (Table 1). Further solidifying this difference was the fact that fans and non-fans were found to view the brand significantly differently on 41 of the 42 BP adjectives. Therefore, it appears to be very short sighted to study only one of these groups and be able to provide a holistic review of the brand's personality as well as suggestions on ways the organization can improve their personality.

Not only did fans view the brand differently than non-fans holistically, these two groups also differed in their assessment of the BP adjectives provided by the three different sources. Fans of the program reported much greater agreement in the applicability of the adjectives provided by all three sources collectively. Overall, fans also found the adjectives provided by the administrators to be most representative of their team, followed by those provided by fans, and then finally the adjectives provided by non-fans were found to be the least representative when fans of the program were doing the evaluation. However, non-fans found no statistically significant difference between the adjectives provided by each source. Perhaps these results are akin to the issue described earlier where administrators of a program are likely to only provide adjectives that are positive in connotation, subsequently providing more of an ideal BP rather than a realistic BP. Fans evaluating the BP of the program agreed most with the adjectives provided by the administrators which, like the administrators in previous studies (i.e., Heere, 2010; Walsh et al., 2013) were all positive in nature.

However, non-fans viewed the team through a less idealistic lens. Non-fan respondents were not found to distinguish between the three sources as there were no statistically significant differences between the three sources in the non-fan responses. From a pragmatic perspective these findings demonstrate that marketers truly looking to receive candid feedback on their organization's BP would be wise to collect adjectives from a variety of sources and solicit feedback on the applicability of those adjectives from a diverse group of fans and non-fans.

When the data were split to evaluate fans and non-fans separately, a very different interpretation could be made. While 41 of the 42 adjectives were statistically significantly different between fans and non-fan, it appears as though the difference really lies in the fact that fans' scores were significantly inflated when compared to non-fans (or one could argue non-fans scores were suppressed). Interestingly, there was overlap on six of the top 10 adjectives fans and non-fans believed best described the BP of the organization in the current study (see Table 2). While fans indicated the adjectives that best describe the BP of the organization were overwhelmingly positive, the top 10 adjectives reported by non-fans were also mostly positive. It appears as though the BP of the team in the current study is relatively stable considering the consistency between the top 10 lists of both fans and non-fans. However, the presence of adjectives provided by administrators found within the bottom 10 list of the non-fans may be indicators of areas the marketing department should focus as these adjectives are clearly not resonating within the non-fan group.

Surprisingly, the top three adjectives as rated by fans within the current study were provided by non-fans during the adjective solicitation phase of the study. This means that fans of the organization determined the adjectives which best described the BP of their organization were provided by non-fans of that organization. This further supports the suggestion to include a variety of stakeholders at all phases of BP measurement in sport marketing research. While Heere (2010) revolutionized BP research methodology within sport marketing by suggesting administrators provide the adjectives fans evaluate with respect to their ability to describe an organizations BP, the current study revealed this method may need to be taken a step further.

This study aimed to investigate alternatives to BP measurement within sport marketing. Findings revealed a stakeholder approach would be very beneficial for future BP research. Not only were the adjectives provided by fans, non-fans, and administrators beneficial to providing a more holistic view of the organization's BP, sampling fans and non-fans proved very valuable. As noted above, the importance of attracting non-fans impacts organizations at the box office, within media rights negotiations, and sponsorship solicitation. Finally, the current study revealed the need to look beyond collective mean scores. Results presented in Table 3 indicate administrators were most successful in providing adjectives fans and non-fans perceived to be accurate descriptors of the organization's BP. However, the results presented in Table 2 clearly indicate that non-fans provided several adjectives fans and non-fans found to be most representative of the organization's BP. It is quite clear that in order to gain the clearest view of an organization's BP a variety of perspectives are needed.

Limitations and Future Research

The current study illuminated numerous future research possibilities. However, the study was not without limitations. Most clearly, the current study is a reflection on the BP of one collegiate men's basketball program. While this study took a more holistic view of BP by soliciting adjectives from three separate sources (administrators, fans, and non-fans) the collection of adjectives from these three sources could have been more broad. It should also be noted that the BP of the program under investigation in this study was not remarkable, but rather very typical of a collegiate men's basketball program. This may make it ripe for BP research or may be a limitation as there may have been confusion amongst respondents as the BP was not clearly defined. Furthermore, the rivalry intensity between the team under investigation and the non-fans who provided BP adjectives and the non-fans who completed the survey may also bias results. Furthermore, the ability to accurately measure the best source for BP adjectives proved difficult. While social media has been utilized by numerous social scientists, the inability to report a response rate is a limitation to the ability to infer these results to other populations. Selection bias is always a concern in sample based research. The use of social media sampling strategies can minimize sampling bias but not eradicate this error. In an attempt to mitigate sampling bias in the current study the research team offered an incentive of a tablet device which is non-sport nor team affiliated. Finally, the sample was relatively homogeneous with the majority of respondents being educated, older, Caucasian males.

The concepts presented and supported in this manuscript provide ample research opportunities moving forward. Further understanding the importance of a more holistic approach to BP adjective solicitation will provide academicians more refined research methods while providing practitioners with a more broad perspective from which they can manage their brand. Utilization of the stakeholder approach needs greater attention in the sport management BP literature base. Future researchers may expand the stakeholders to include alumni, donors, sponsors, media rights holders, or even players. Research in the sport team BP vein may also thrive if we are able to best identify the various stakeholders, beyond the current fan base or those in attendance at a game, who should be targeted for inclusion when evaluating the BP of a team or sport organization. Overall, the current study found that Heere's (2010) method of moving beyond trying to utilize an all-encompassing BP in the sport context was a vital first step in progressing the literature base. Yet, this was but the first step in improving the sport-focused BP research design, and the team stakeholder approach is the logical next step as this literature base continues to evolve. Critical insights may also be revealed if future researchers were to assess BP adjective creation and assessment by fans across varying levels of fan avidity or attachment. These insights could prove very beneficial for sport marketers.

References

Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34, 347-356.

Aggarwal, P. (2004). The effects of brand relationship norms on consumer attitudes and behavior. Journal of Consumer Research, 3, 87-101.

Aggarwal, P., & McGill, A. L. (2007). Is that car smiling at me? Schema congruity as a basis for evaluating anthropomorphized products. Journal of Consumer Research, 34, 468-479.

Anandkumar, V., & George, J. (2011). From Aaker to Heere: A review and comparison of brand personality scales. Research Journal of Social Science & Management, 1(3), 30-51.

Austin, J. R., Siguaw, J. A., & Mattila, A. S. (2003). A re-examination of the generalizability of the Aaker brand personality measurement framework. Journal of Strategic Marketing, 11, 77-92.

Avis, M. (2012). Brand personality factor based models: A critical review.

Australasian Marketing Journal, 20, 89-96.

Azoulay, A., & Kapferer, J. N. (2003). Do brand personality scales really measure brand personality? The Journal of Brand Management, 11, 143-155.

Bhutta, C. B. (2012). Not by the book: Facebook as a sampling frame. Sociological Methods & Research, 4, 57-88.

Braunstein, J. R., & Ross, S. D. (2010). Brand personality in sport: Dimension analysis and general scale development. Sport Marketing Quarterly, 19, 8-16.

Carlson, B. D., & Donovan, D. T. (2013). Human brands in sport: Athlete brand personality and identification. Journal of Sport Management, 27, 193-206.

Carlson, B. D., Donavan, D. T., & Cumiskey, K. J. (2009). Consumer-brand relationships in sport: Brand personality and identification. International Journal of Retail & Distribution Management, 37, 370-384.

Clavio, G. (2011). Social media and the college football audience. Journal of Issues in Intercollegiate Athletics, 4, 309-325.

Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational images and member identification. Administrative Science Quarterly, 39, 239-63.

Dwyer, B., Mudrick, M., Greenhalgh, G. P., LeCrom, C. W., & Drayer, J. (2015). The tie that blinds? Developing and validating a scale to measure emotional attachment to a sport team. Sport Management Review, 18, 570-582.

Esch, F., Langer, T., Schmitt, B. H., & Geus, P. (2006). Are brands forever? How brand knowledge and relationships affect current and future purchases. Journal of Product & Brand Management, 15, 98-105.

Field, A. (2009). Discovering statistics using SPSS (3rd ed.). Thousand Oaks, CA: Sage.

Freeman, R.E. (1984). Strategic management: A stakeholder approach. Boston: Pitman.

Freling, T. H., & Forbes, L. P. (2005). An empirical analysis of the brand personality effect. Journal of Product & Brand Management, 14, 404-413.

Friedmann, R. (1986). Psychological meaning of products: Identification and marketing applications. Psychology & Marketing, 3, 1-15.

Friedmann, R., & Jugenheiner, D. (1985). Copytesting through psychological meaning. In N. Stephens (Ed.), Proceedings of the 1985 Conference of the American Academy of Advertising.

Gladden, J. M., & Funk, J. C. (2002). Developing an understanding of brand associations in team sports: Empirical evidence from consumers of professional sport. Journal of Sport Management, 16, 54-81.

Graham, T. (2014, December 21). Raiders now have battered reputation. The Buffalo News. Retrieved from http://bills.buffalonews.com/2014/12/21/raiders-now-have-battered reputation/

Heere, B. (2010). A new approach to measure perceived brand personality associations among consumers. Sport Marketing Quarterly, 19, 17-24.

Hosany, S., Ekinci, Y., & Uysal, M. (2006). Destination image and destination personality: An application of branding theories to tourism places. Journal of Business Research, 59, 638-642.

Kang, C., Bennett, G., & Welty Peachey, J. (2016). Five dimensions of brand personality traits in sport. Sport Management Review, 19, 441-453.

Kaufman, L., & Rousseeuw, P. J. (2005). Finding groups in data: An introduction to cluster analysis. Hoboken, NJ: John Wiley & Sons.

Kim, Y. D., Magnusen, M., & Kim, Y. (2012). Revisiting sport brand personality: Scale development and validation. Journal of Multidisciplinary Research, 4, 65-80.

Larkin, B. (2015). An examination of fantasy sport participation motives and substitution versus attendance intention. Sport Marketing Quarterly, 24, 120-133.

McCormick, T. H., Lee, H., Cesare, N., Shojaie, A., & Spiro, E. S. (2015). Using Twitter for demographic and social science research: Tools for data collection and processing. Sociological Methods & Research, 1, 1-32.

Mizik, N., & Jacobson, R. (2008). The financial value impact of perceptual brand attributes. Journal of Marketing Research, 45, 15-32.

Mosely, M., Seifert, K., Walker, J., & Graham, T. (2009, July 6). The great debate: "America's team." ESPN. Retrieved from http://espn.go.com/blog/nflnation/post/_Zid/6329/the-great-debateamerica-s-team

Pappu, R., Quester, P. G., & Cooksey, R. W. (2005). Consumer-based brand equity: Improving the measurement-empirical evidence. Journal of Product & Brand Management, 14, 143-154.

Robinson, M. J., & Trail, G. T. (2005). Relationships among spectator gender, motives, points of attachment, and sport preference. Journal of Sport Management, 19, 58-80.

Romaniuk, J. (2008). Comparing methods of measuring brand personality traits. Journal of Marketing Theory and Practice, 16, 153-161.

Ross, S. (2006). A conceptual framework for understanding spectator-based brand equity. Journal of Sport Management, 20, 22-38.

Ross, S. D. (2008). Assessing the use of the brand personality scale in team sport. International Journal of Sport Management and Marketing, 3, 23-38.

Ruihley, B. J., Billings, A. C., & Rae, C. (2014). As time goes by: Deciphering the fantasy sport playing teenager. Sport Marketing Quarterly, 23, 189-197.

Schade, M., Piehler, R., & Burmann, C. (2014). Sport club personality scale (SCBPS): A new brand personality scale for sport clubs. Journal of Brand Management, 21, 650-663.

Scott, S. G., & Lane, V. R. (2000). A stakeholder approach to organizational identity. Academy of Management Review, 25, 43-62.

Smith, A. C., Graetz, B. R., & Westerbeek, H. M. (2006). Brand personality in a membership-based organisation. International Journal of Nonprofit and Voluntary Sector Marketing, 11, 251-266.

Staples, A., (2015, July 1). What's your program's brand? What should it be? A look across the nation. Sports Illustrated. Retrieved from http://www.si.com/college-football/2015/ 07/01/college-football-teamsbrand?page=6&devicetype=default

Stevens, J. (2002). Applied multivariate statistics for the social sciences. (4th ed.). Mahawah, NJ: Lawrence Erlbaum Associates.

Stinnett, R. C., Hardy, E. E., & Waters, R. D. (2013). Who are we? The impacts of anthropomorphism and the humanization of nonprofits on brand personality. International Review on Public and Non-profit Marketing, 10, 31-48.

Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (5th ed.). Boston, MA: Pearson Education.

Tsiotsou, R. (2012). Developing a scale for measuring the personality of sport teams. Journal of Services Marketing, 26, 238-252.

Walsh, P., Clavio, G., Lovell, M. D., & Blaszka M. (2013). Differences in event brand personality between social media users and non-users.

Weick, K. E. (1995). What theory is not, theorizing is. Administrative Science Quarterly, 40, 385-390.

Wood, L. (2000). Brands and brand equity: Definition and management. Management Decision, 38, 662-669.

Yoon, C., Gutchess, A. H., Feinberg, F., & Polk, T. A. (2006). A functional magnetic resonance imaging study of neural dissociations between brand and person judgments. Journal of Consumer Research, 33, 31-40.

Greg Greenhalgh, PhD, is the director of student services and outreach of the Center for Sport Leadership at Virginia Commonwealth University. His research interests include marketing of niche sports, sport sponsorship, and sport and the natural environment.

Brendan Dwyer, PhD, is the director of research and distance learning of the Center for Sport Leadership at Virginia Commonwealth University. His research focuses on sport consumer behavior with a distinct focus on the media consumption habits of fantasy sport participants and behavioral patterns of ticket purchasers.

Carrie LeCrom, PhD, is the executive director of the Center for Sport Leadership at Virginia Commonwealth University. Her research focuses on sport for development. Table 1 Source, Sentiment, Means, and Standard Deviations for All Adjectives, and p value Differences Between Fans and Non-Fans Adjective Source Sentiment Fan (n=114) Athletic NF Positive 6.19 (1.01) Boring NF Negative 2.07 (1.58) Charismatic A Positive 5.46 (1.28) Competitive NF Positive 6.31 (1.03) Consistent NF Positive 5.32 (1.37) Dangerous NF Positive 4.47 (1.97) Declining NF Negative 1.98 (1.63) Defensive F Neutral 5.67 (1.49) Determined A, F Positive 6.14 (0.98) Disciplined NF Positive 5.86 (1.14) Disillusioned NF Negative 2.31 (1.57) Driven A Positive 6.13 (0.99) Eager A Positive 6.05 (1.03) Encouraging F Positive 5.92 (1.12) Energetic A, F Positive 6.03 (1.13) Exciting A Positive 6.08 (1.10) Fearless F Positive 5.63 (1.21) Frustrating F Negative 3.16 (1.83) Fun F Positive 6.11 (1.15) Gritty F Positive 5.54 (1.38) Hard-Working NF Positive 6.32 (0.99) Hungry F Positive 6.05 (1.10) Inconsistent F Negative 3.09 (1.73) Intelligent F Positive 5.69 (1.18) Irrelevant NF Negative 2.00 (1.61) Knowledge-able A Positive 5.79 (1.09) Methodical NF Positive 5.21 (1.27) Poised F Positive 5.80 (1.11) Potential NF Positive 6.25 (0.99) Proud A, F Positive 6.21 (1.15) Relentless F Positive 5.61 (1.27) Resilient F Positive 5.90 (1.12) Respectful F Positive 6.04 (1.18) Strong F Positive 5.92 (1.13) Surprising F Neutral 5.69 (1.34) Talented A Positive 6.00 (1.00) Tenacious F Positive 5.79 (1.06) Tough NF Positive 5.96 (1.16) Traditional F Neutral 5.31 (1.28) Transitional F, NF Negative 5.04 (1.40) Urban NF Neutral 4.91 (1.32) Weak NF Negative 1.88 (1.55) Adjective Non-Fan (n=121) p value Athletic 4.53 (1.22) < .001 Boring 4.06 (1.47) < .001 Charismatic 3.48 (1.14) < .001 Competitive 4.75 (1.28) < .001 Consistent 3.41 (1.29) < .001 Dangerous 4.16 (1.47) .171 Declining 3.71 (1.50) < .001 Defensive 4.04 (1.27) < .001 Determined 4.51 (1.21) < .001 Disciplined 3.99 (1.47) < .001 Disillusioned 3.88 (1.48) < .001 Driven 4.13 (1.12) < .001 Eager 4.30 (1.29) < .001 Encouraging 3.71 (1.24) < .001 Energetic 4.18 (1.20) < .001 Exciting 3.57 (1.32) < .001 Fearless 4.00 (1.12) < .001 Frustrating 4.27 (1.24) < .001 Fun 3.52 (1.36) < .001 Gritty 4.39 (1.27) < .001 Hard-Working 4.55 (1.26) < .001 Hungry 4.33 (1.27) < .001 Inconsistent 4.75 (1.19) < .001 Intelligent 3.68 (1.22) < .001 Irrelevant 4.11 (1.66) < .001 Knowledge-able 3.95 (1.17) < .001 Methodical 3.91 (1.13) < .001 Poised 4.05 (1.14) < .001 Potential 4.49 (1.32) < .001 Proud 4.58 (1.38) < .001 Relentless 3.95 (1.23) < .001 Resilient 4.28 (1.23) < .001 Respectful 3.23 (1.41) < .001 Strong 4.19 (1.28) < .001 Surprising 4.46 (1.48) < .001 Talented 4.26 (1.26) < .001 Tenacious 4.09 (1.22) < .001 Tough 4.26 (1.34) < .001 Traditional 4.01 (1.29) < .001 Transitional 4.11 (1.19) < .001 Urban 4.55 (1.39) <.001 Weak 3.56 (1.46) < .001 Table 2 Adjective Ratings Fans Top 10 Adjectives Adjective Source Sentiment Mean(SD) Hardworking NF Positive 6.32 (0.99) Competitive NF Positive 6.31 (1.03) Potential NF Positive 6.25 (0.99) Proud A, F Positive 6.21 (1.15) Athletic NF Positive 6.19 (1.01) Determined A, F Positive 6.14 (0.98) Driven A Positive 6.13 (0.99) Fun F Positive 6.11 (1.15) Exciting A Positive 6.08 (1.10) Eager A Positive 6.05 (1.03) Bottom 10 Adjectives Transitional F, NF Negative 5.04 (1.40) Urban NF Negative 4.91 (1.32) Dangerous NF Negative 4.47 (1.97) Frustrating F Negative 3.16 (1.83) Inconsistent F Negative 3.09 (1.73) Disillusion NF Negative 2.31 (1.57) Boring NF Negative 2.07 (1.58) Irrelevant NF Negative 2.00 (1.61) Declining NF Negative 1.98 (1.63) Weak NF Negative 1.88 (1.55) Non-Fans Top 10 Adjectives Adjective Source Sentiment Mean(SD) Inconsistent F Negative 4.75 (1.19) Competitive NF Positive 4.75 (1.28) Proud A, F Positive 4.58 (1.38) Urban NF Negative 4.55 (1.39)) Hardworking NF Positive 4.55 (1.26) Athletic NF Positive 4.53 (1.22) Determined A, F Positive 4.51 (1.21) Potential NF Positive 4.49 (1.32) Surprising F Neutral 4.46 (1.48) Gritty F Positive 4.39 (1.27) Bottom 10 Adjectives Disillusion NF Negative 3.88 (1.49) Encouraging F Positive 3.71 (1.24) Declining NF Negative 3.71 (1.50) Intelligent F Positive 3.68 (1.22) Exciting A Positive 3.57 (1.32) Weak NF Negative 3.56 (1.46) Fun F Positive 3.52 (1.36) Charismatic A Positive 3.48 (1.14) Consistent NF Positive 3.41 (1.29) Respectful F Positive 3.23 (1.41) Table 3 Adjective Means and Differences Between Fans and Non-Fans Based on Source Respondents Adjective Source Administrators Fans Non-Fans Fans(n = 114) 5.98 (0.92) 5.54 (0.74) 4.50 (0.59) Non-Fans (n = 114) 4.11 (0.94) 4.11 (0.79) 4.12 (0.47) P < .001 < .001 <.001 Note. All items measured on a 7-point Likert type scale.