Revisiting Sport Brand Personality: Scale Development and Validation

Abstract

Comprehensive brand personality conceptualizations may not be applicable to a single category goods or service brand personality because such approaches are too broad and overstated. Furthermore, different brand categories need to be specified and conceptualized according to their own distinctive features. With those observations in mind, as well as the absence of a valid and reliable sport brand personality measurement tool, the purpose of this study was to create a sport brand personality scale centered on sports apparel and footwear. Using Korean sport consumers, six sport brand personality factors were identified: Competence, Creativity, Ruggedness, Excitement, Sincerity, and Energy. A valid and reliable sport-centric brand personality scale also was developed. The instrument includes a total of 18 items, with each of the six brand personality factors consisting of 3 items. Included in the discussion section is an explanation of the research implications, managerial applications, and limitations of the developed brand personality scale. [PUBLICATION ABSTRACT]

Headnote

Abstract

Comprehensive brand personality conceptualizations may not be applicable to a single category goods or service brand personality because such approaches are too broad and overstated. Furthermore, different brand categories need to be specified and conceptualized according to their own distinctive features. With those observations in mind, as well as the absence of a valid and reliable sport brand personality measurement tool, the purpose of this study was to create a sport brand personality scale centered on sports apparel and footwear. Using Korean sport consumers, six sport brand personality factors were identified: Competence, Creativity, Ruggedness, Excitement, Sincerity, and Energy. A valid and reliable sport-centric brand personality scale also was developed. The instrument includes a total of 18 items, with each of the six brand personality factors consisting of 3 items. Included in the discussion section is an explanation of the research implications, managerial applications, and limitations of the developed brand personality scale.

Keywords

sport brand personality, personality traits, creativity, footwear, sports apparel, sport consumers

Introduction

Sport organizations throughout the world have given greater priority to brand management strategies, investing hundreds of millions if not billions of dollars on such activities. The European soccer club Manchester United, for example, enjoys the most lucrative jersey sponsorship deal in soccer history. The deal with Chevrolet (a U.S. automaker) is worth more than US$559 million over 7 years (Cutler, 2012). Additionally, Nike, the sport apparel and footwear giant, spent more than two billion dollars on the marketing and promotion of its products in 2010. To put that in perspective, Under Armour, a rival to Nike, generated more than a billion dollars in revenue that same year (a first for the company in its 15 year history), and yet its total revenue was still less than half of what Nike spent on promoting its brand alone (Burke, 2011).

Along with sport teams and firms, sport scholars also have embraced brand management strategies, devoting increasing amounts of attention toward understanding how to build strong, favorable, and distinctive brands in order to positively affect sport consumer attitudes and patronage behaviors (e.g., Bauer, Stokburger-Sauer, &Exler, 2008; Gladden &Funk, 2002; Ross, 2006). In particular, an area of branding research that has both piqued the interest of and proven challenging for sport scholars to comprehend sufficiently is brand personality. A key reason brand personality research has proven challenging for sport researchers is that, despite several research attempts, a reliable measure of brand personality has yet to be produced. Braunstein and Ross (2010), for example, developed an adapted sport brand personality scale (BPS) that was met with mixed success as only two of the six dimensions were found to be reliable (i.e., good estimates in both Cronbach's alpha and average variance explained).

Therefore, the primary objective of this study is to improve upon past research efforts and develop a scale that can be used to measure brand personality in the context of sport products (i.e., sport apparel and footwear). In doing so, this study not only creates a brand personality measurement tool but also highlights the benefits and inherent limitations of a general scale development approach to sport brand personality research. Preceding the research method and results sections of this study is a brief review of the relevant brand personality literatures as well as a more detailed explanation of how this study makes an important contribution to sport marketing research and practice.

Literature Review

The brand personality concept tends to be depicted as a key measure for understanding consumer perceptions of and attitudes toward brands, in addition to serving as a communication guideline for brand strategists to reference when engaging consumers (Aaker, 1996). More specifically, in her seminal work on brand personality, Aaker (1997) held the concept to reflect the associations of human characteristics with a specific brand. Plummer (2000) likewise defined brand personality as "the characterization aspects of the brand" (p. 81). However, Azoulay and Kapferer (2003) viewed such approaches as too broad and vague. As a result, they defined brand personality as "the set of human personality traits that are both applicable to and relevant for brands" (Azoulay & Kapferer, 2003, p. 151). Of the various definitions found in the extant literature, Azoulay and Kapferer's revised conceptualization appears to provide a clearer and more precise explanation of brand personality, and it, therefore, represents the version employed in this study.

Although by no means as extensive as the body of research established in the business literatures, several relevant approaches to examining brand personality research have emerged in the sport marketing literature. Forming the foundation for the first approach is the work of Aaker (1997). She developed the concept of brand personality and then, looking to the personality research of Goldberg (1992), developed a five dimensional scale for assessing the construct that subsequently has been used and adapted across a multitude of cultures and situations (Aaker, 1997; Aaker, Benet-Martinez, &Garolera, 2001; Beldona & Wysong, 2007; Bosnjak, Bochmann, & Hufschmidt, 2007; Supphellen & Gr0nhaug, 2003). By extension, one approach undertaken by sport marketing scholars has been the application of Aaker's brand personality scale to a sport context (Cho, 2004; Rosenberger & Donahay, 2008; Ross, 2008; Smith, Graetz, & Westerbeek, 2006). However, despite repeated attempts, sport scholars who have undertaken this approach to sport brand personality have yet to demonstrate sufficiently that Aaker's original scale is fully applicable to sports teams.

Next, a growing number of studies (Austin, Siguaw, & Mattila, 2003; Azoulay & Kapferer, 2003; Bosnjak et al., 2007; Gladden & Funk, 2002; Parent & Séguin, 2008; Ross, 2008; Smith et al, 2006) lend support to the notion that general brand personality conceptualizations, such as Aaker's brand personality framework, may not be automatically applicable to sports. The reason this may be the case is that only three sport brands were selected as representative examples of sport brands (ESPN station, Nike athletic shoes, and Reebok athletic shoes) in Aaker's study (1997). Therein, she developed four brand groups, and each group consisted of nine brands. Three of the groups had one sport brand; however, one of the four groups did not have a sport brand. Moreover, when she confirmed the brand personality dimensions she generated, no sport brand was selected and matched to a confirmatory sample of brands. Consequently, sport scholars have sought to develop and individualize the brand personality "concept on their own turf' (Braunstein & Ross, 2010, p. 8).

Of note, Braunstein and Ross (2010) made an effort to develop the BPS, which consisted of 6 brand personality factors and 41 items. Even though this attempt was met with mixed success, their adapted scale nevertheless reflected a much needed divergence from the more frequently employed approach of applying Aaker's (1997) scale to sport brands. Therefore, given potential problems with using Aaker's scale, as well as the absence of a reliable and valid sportspecific brand personality scale, the development of a usable scale to evaluate the unique personalities of sport brands has great research and practical value for the study and execution of branding activities in sport marketing.

Accordingly, one main contribution of the present study is the development of a sport brand personality scale specific to sport goods. This aspect of the study holds two-fold significance to the sport marketing literature. First, though Aaker's (1997) seminal work represents a prominent cornerstone of brand personality research, it has come under increasing scrutiny in recent years. At the core of these criticisms are issues of scale validity and applicability to various contexts and products (Austin et al, 2003; Azoulay & Kapferer, 2003; d'Astous & Lévesque, 2003), a point that several sport-based studies also have observed (e.g., Heere, 2010; Ross, 2008; Smith et al, 2006). This study, therefore, is advantageous to sport scholars because it does not attempt to fit Aaker's scale to a sport context. It seeks, instead, to develop a unique sport brand personality scale.

Second, and in tandem with the first point, sport management scholars have paid minimal attention to the development of distinctive sport brand personality scales. One contemporary exception is Braunstein and Ross (2010); they focused on professional sport teams and developed the BPS. However, when discussing their newly adapted scale, they noted "the reliability and validity of the BPS scale, in its current state, do not provide a sufficiently sound instrument" (Braunstein & Ross, 2010, p. 13). Although the approach of this study does not address all of the limitations of the extant sport branding literature, the creation of a valid and reliable sport-based brand personality scale represents a significant and positive advancement.

Method

Overview

In the present study, the development of a brand personality scale centered on apparel and footwear as represented by the well-known sport companies Nike and Adidas in 2010 (Korean Management Association Consulting) (see Table 1). The rationale for the apparel and footwear was grounded in the following information. The Korean sport industry is classified into three areas: (a) sport facility, (b) sport product, and (c) sport service. The sport product category consists of five subcategories: sport equipment, sport apparel and shoes, sport food and drink, wholesale and retail trade, and online trade. In 2009, the market size of the apparel and sport footwear subcategory was approximately $1.2 billion US dollars (34% of the total sport product market); this was the largest of the five subcategories (Ministry of Culture, Sports, and Tourism, 2009) , which is the reason the researchers conducting this study selected it as the focal product category.

Participants

The target population for this study was University students attending colleges and universities located in the Seoul and Gyeonggi Provinces in Korea. Out of 500 questionnaires, 55 survey questionnaires contained incomplete responses. This resulted in 445 (a response rate of 89.4%) questionnaires being suitable for data analysis. The respondents consisted of 303 male (67.8%) and 144 female (32.2%). Approximately, 59.5% of the respondents were between 20 and 25 years old, 27.5% were in the 25 or higher age range, and 13% were between 18 and 20 years old.

Measures

The researchers conducted two separate surveys as part of a two-stage process to develop a sport brand-specific personality measurement tool. Similar to the research of Ross, James, and Vargas (2006), the first stage was the generation of items stage; it included a free-thought listings questionnaire. A convenience sample of 200 students who were enrolled in sport, recreation management, and physical education courses at a large university in Seoul, Korea, completed the free-thought listing questionnaires. This resulted in a total of 107 sport brand personality traits.

The second stage was "the purification stage." In this stage, sport brand apparel and footwear consumers completed the questionnaire consisting of 107 sport brand personality traits. The second stage of the survey involved developing the final arrangement of sport brand personality traits. To improve the appropriateness of the identified sport brand personality traits, a total of 38 teaching assistants, research assistants, and instructors who were majoring in sport management participated in the survey. They responded to the questionnaire items on a 7 -point Likert-type scale ranging from 1 (very inappropriate) to 7 (very appropriate). Next, an expert panel of scholars, including four sport marketing professors and one brand management researcher, selected only sport brand personality traits averaging 5.0 or higher. The panel, using the criteria of 5.0 or higher, selected 47 sport brand traits for inclusion in the final sport brandcentric personality questionnaire.

Data Analysis

During the data analysis process, questionnaires deemed to be redundant, insufficient, and/or conceptually unreliable were excluded and only the remaining questionnaires were restructured. To start, an exploratory factor analysis (EFA) was conducted using predictive analytics software and solutions (PASW) statistics 18.0. This was done in order to identify and consolidate potential underlying factors of the sport brand-centric personality scale (Tabachnick & Fidell, 2007). Principal axis factoring with varimax rotation was used in order to reduce the number variables in a model to the smallest number as possible. The varimax method is also focused on the interpretation of factors while at the same time maintaining their independence from one another.

In order to determine an appropriate number of factors, three criteria were used: (a) The Kaiser criterion (eigenvalue are greater than 1; Kaiser, 1970), (b) scree test (Catell, 1966), and (c) interpretable criteria (loading with excess .45 is fair; Comrey & Lee, 1992). Additionally, in order to verify the extent to which items consistently reflect a single construct, an internal consistency reliability analysis was conducted using Cronbach's a. This was done to indicate the degree to which the items are interrelated (Johnson & Christensen, 2008). For purposes of the present study, the cutoff criterion was set at .70 (Nunnally, 1978).

After this step, a confirmatory factor analysis (CFA) was conducted to evaluate the measurement model representing "how measured variables reflect certain latent variables" (Thompson, 2010, p. 110) using Mplus 6.0 software (Muthén & Muthén, 2008). According to Hu and Bender (1999), in order to evaluate the overall fit of a model, several goodness of fit indices also need to be estimated: (a) chi-square (statistically non-significance is favorable), (b) the comparative fit index (greater than .90 or .95 demonstrates "reasonable" fit), (c) the root mean square error of approximation (equal to or less than .05 corresponds to a "good" fit), (d) the standardized root mean square residual (smaller than .08 is "favorable" fit), and (e) the TuckerLewis index (greater than .90 or .95 demonstrates "reasonable" fit).

Results

Exploratory Factor Analysis

Three criteria (i.e., eigenvalue, scree test, and factor loading) were assessed with the 47 sport personality traits generated from the two separate item generation processes. The analysis revealed that six factors were initially structured based on eigenvalues greater than 1. This sixfactor model was also supported by the scree plot test. Among the six factors, 19 items were eliminated based on factors loadings lower than .45. The first factor was Competence; it consisted of 10 items. The second and third factors were Creativity and Ruggedness; they consisted of 7 items. The fourth factor was Excitement, which consisted of 4 items. The fifth factor was Sincerity, and it included 3 items. The sixth and final factor was Energy; it consisted of 6 items.

Cronbach's a for each factor was as follows: .876 (Competence), .849 (Creativity), .844 (Ruggedness), .843 (Excitement), .863 (Sincerity), and .818 (Energy). All these values were greater than .70, which was the criterion established by Nunnally (1978). The EFA results indicate that Korean sport consumers recognized the six -factor model with 3 7 -items of sport brand personality (see Table 2) .

Confirmatory Factor Analysis

As a follow up to the EFA, a CFA was conducted to evaluate whether the proposed model of 6-factors with 3 7 -items adequately measured sport brand personality. Prior to conducting the CFA, the multivariate normality assumption was examined using LISREL Version 8.8. Since the normality assumption was not met (p < .05), Satorra-Bentler (1994) scaling correction method was utilized to address violation of the normality assumption. Based on the result of the CFA, however, the proposed measurement model through EFA did not fit the data well. Goodness of fit indices indicated that chi-square was significant (?2 =2238.19, ? < .001), which suggests the model was not favorable. Both the CFI and TLI were not greater than .90 or .95 (CFI=.794, TLI=.776), which also indicated a poor fit (Hu & Bender, 1999). Further, the RMSEA (RMSEA =.078 >.05) value revealed an unreasonable fit (Hu & Bender, 1999). Lastly, a favorable SRMR is less than .08 (Hu & Bender, 1999). In the present study, the reported value was .068 (SRMR = 0.68 < .080). Only this fit index suggests a good model fit to the data (Hu & Bender, 1999).

Overall, the results of the CFA for the 6-factor model with 37-items revealed that except for the SRMR, other global fit indices were not acceptable. Based on this evidence, the model, with particular reference to the items, needed to be readjusted. Several items were therefore eliminated based on the factor loadings, model modification indices, theoretical considerations (i.e., excessively high correlation between two items and low factor loadings). This process resulted in the removal of 19 items and the creation of a 6-factor instrument with 18-items. The revised instrument consisted of the following factors and item totals: Competence (3 items), Creative (3 items), Ruggedness (3 items), Excitement (3 items), Energy (3 items), and Sincerity (3 items). Once again, before conducting the CFA, the multivariate normality assumption was analyzed. In this case, the normality of this model was still violated (p < .05). Thus, the SatorraBentler collection was also used in this model, and a CFA was conducted to evaluate if the reproposed model of 6-factors with 18-items appropriately measured sport brand personality.

From the CFA results it is evident the revised measurement model was a good fit for data once the low factor loading items and highly correlated items were removed. By and large, the goodness of fit indices also provided values supportive of this observation. Chi-square was significant (?2 =181.92, ? < .001), which implies the model was not favorable. However, the Chi-square statistic is both functionally and statistically affected by a large sample sizes. As a result, the variety of goodness of fit indices was carefully examined.

Both the CFI and TLI were greater than .90 or .95 (CFI = .983; TLI = .979), which indicates a good fit (Hu & Bender, 1999). The RMSEA was less than .05 (RMSEA =.034 <.05); this indicates reasonable fit (Hu & Bender, 1999). The SRMR value also indicated a good model fit (SRMR = .034 < .08) according to Hu and Bender' (1999) criteria (SRMR < .080 or .010).

The discriminant validity and reliability of all factors were measured. Reliability was evaluated using Cronbach's alpha. Average variance explained (AVE) was also calculated to indicate to the extent which the items on an underlying factor correlated with one another. The AVE values showed that Cronbach's a for each factor was: .817 (Competence), .855 (Creativity), .727 (Ruggedness), .836 (Excitement), .863 (Sincerity), and .827 (Energy). Each factor demonstrated alpha levels that were greater than .70. In addition, Hair, Anderson, Tatham, and Black (1998) suggested that AVE value above .50 have good convergent validity. In this study, each factor indicated AVE values that were greater than .50, thus providing evidence for convergent validity. Discriminant validity was evaluated using AVE values. Fornell and Larcker (1981) suggested that if the AVE value is larger than the squared correlation between two referent factors, it is evidence of discriminant validity. Thus, the revised model's discriminant validity was satisfied (see Figure 1).

Discussion

Overall, the results of the present study make several contributions to the extant sport marketing and strategic brand management literatures. One promising development is the creation of a sport-centric brand personality scale. Rather than just extend Aaker's (1997) scale to a sport context, we sought to embrace and explore the concept of brand personality specific to the realm of sports. Also, whereas the Braunstein and Ross (2010) scale had mixed reliability results, the final scale generated in this study was found to be a valid and reliable measurement tool.

Another interesting contribution surfaces after cross-examining the current results with other brand personality research examples. In this study, six personality dimensions were identified (i.e., competence, creativity, ruggedness, excitement, sincerity, and energy). While several of these factors, such as ruggedness and sincerity, have been reported with sport and nonsport examples (Aaker, 1997, Aaker et al, 2001; Braunstein & Ross, 2010; Smith et al., 2006; Yang, 2002), other factors, such as creativity and energy, have not been examined previously.

On the one hand, this finding is in line with recent criticism regarding sport brand personality scale development. Specifically, brand personality scales (such as Aaker's scale) may be too tailored to a specific context or product to be fully applied outside of their primary setting (Heere, 2010). On the other hand, such criticism may be hasty in its dismissal of brand personality research because, as a greater number of sport brand personality scales are developed, a commonality among brand factors may emerge. As this research line matures, it could lead to the emergence of a core set of brand personality factors (similar to what has been seen with motives research in sport management). This core set of factors would then provide sport researchers and practitioners with a valuable brand personality foundation. Indeed, this foundation could provide them with the opportunity to: (a) investigate these "core" (and assumed to be generalizable) brand personality factors across a myriad of contexts and sport products and (b) discover supplementary brand personality factors that may be context specific.

Finally, another intriguing result of this study is that the multinational sport corporations, Nike and Adidas, were identified as possessing "creative" and "imaginative" personality traits. These personality traits differ from the stereotypical personality trait of "athletic," which is commonly associated with sport brands (Aaker, 1996). This result may be linked to global sport companies' strategic marketing efforts to "think globally, act locally" (Nederveen Peterse, 2003, p. 49). This kind of tandem marketing operation of local/global dynamics, global localization or glocalization, appears to be at work in the case of Korean sport brand consumers.

Nike and Adidas are actively involved in sponsorship partnership through Korea wherein they make creative and imaginative advertisements with either globally or locally wellknown/popular sport organizations, teams, or athletes in order to create a brand image and brand personality that best suits their local consumers. These giocai communication efforts by Nike and Adidas appear to influence Korean sport consumers to perceive the sport brand personalities of these companies as both creative and imaginative. Moreover, it also appears to demonstrate that Korea sport consumers possess a more nuanced evaluation of Nike and Adidas. That is, they may be more aware of non-product-related characteristics, such as imaginative, in addition to being aware of more traditional product-related characteristic (e.g., athletic, vigorous, lively).

Managerial Implications

Given the continual dawning of new media technologies, sports advertisement and sponsorship communications enable multinational brands to enhance brand associations and brand awareness in rapidly changing sport marketplaces. Based on congruent theory referring to the image/personality of the endorsers or teams with the image/personality of the brand is promoted, the more effective the message (Gwinner & Eaton, 1999), the sport brand personality scale developed in this study can play an important role in (1) determining whether there is a good fit between the sponsor and the desired personality of a particular team. Specifically, there is no doubt that Samsung is perceived as a giant global electronics brand and their logo on the Chelsea FC's uniform is instantly recognized as official corporate sponsor of Chelsea FC by audiences (Kerr & Gladden, 2008). This synergy, whereby there is congruent between two highquality products reflecting the high degree of characteristic similarity, reinforces each other's personality that would be seen as creative, prestigious, hardworking, and leadership. Hence, sport brand personality scale provides a way to evaluate the sponsorship effectiveness between the sponsors and the sport entity before making sponsorship agreement.

Achieving a better understanding of sport brand personality enables sport organizations (2) to monitor the key brand personalities to effectively market and position to their target consumers and differentiate them from the competitors. Measuring sport brand personality, as (3) a roadmap for sport organizations to build an emotional connection with their consumers who make their purchase decisions based on their self-expressions.

Limitations and Future Research

One lingering challenge associated with brand personality research is that brand personality scale factors have been found to vary according to both product and context (Austin et al, 2003; Azoulay & Kapferer, 2003). The current study is no exception. The participants of this study were Korean consumers, and the sport brand personality category was sport products (i.e., apparel and footwear). As a result, the factors identified herein may not be generalizeable to other consumer populations and brands, such as professional and intercollegiate sport teams. Be that as it may, it remains worthwhile to conduct additional research about brand personality in sport in order to determine if a core set of generalizable brand personality factors exists.

In closing, the main objective of the present study was the development of a brand personality scale that could be used to measure a specified sport-related product category (i.e., apparel and footwear). More precisely: (a) a total of 47 sport brand personality traits were chosen through the first and second survey stages, (b) through EFA, a 6-factor model with 3 7 -items was developed, and (c) after re-examining the model through CFA, an improved 6 factor model with 18 items was confirmed. The development of usable brand personality tool represents an important continuation as well as advancement of previous research attempts in this line of sport marketing inquiry, and we therefore hope it spurs continued interest in this important research area.

Sidebar

Discussion Questions

1. According to d'Astous and Lévesque (2003), a single category goods or service brand personality scale is imperative because of its own market uniqueness. This viewpoint stands in contrast to Aaker's (1997) brand personality scale (BPS). The BPS is a general measure that was developed to evaluate all different types of brands. If that is the case, can Aaker's brand personality scale be used to effectively measure distinctive characteristics of sport team brands?

2. What are several challenges associated with sport brand personality research? Discuss several reasons sport brand personality research has proven challenging for sport scholars. For example, in addition to the development of a reliable and valid brand personality scale, what other research developments would improve greatly the study of brand personality in sport?

3. In this study, the companies of Nike and Adidas were identified as possessing "creative" and "imaginative" personality traits. These personality traits stand in contrast to the conventional personality trait of "athletic," which is customarily associated with sport brands (Aaker, 1996). Do you view Nike and Adidas as being creative and imaginative companies? Explain why or why not. Also, what sort of strategic marketing efforts would you recommend to a sport company that is looking to be viewed by consumers as creative and imaginative?

To Cite this Article

Kim, Y. D., Magnusen, M., & Kim, Y. (2012, Fall). Revisiting sport brand personality: Scale development and validation. Journal of Multidisciplinary Research, 4(3), 65-80.

References

References

Aaker, D. A. (1996). Building strong brands. New York: The Free Press.

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

Aaker, J. L, Benet-Martinez, V., & Gorolera, J. (2001). Consumption symbols as carriers of culture: A study of Japanese and Spanish brand personality constructs. Journal of Personality and Social Psychology, 8(3), 492-508.

Austin, J. R., Siguaw, J. ?., & Mattila, A. S. (2003). An assessment of the Aaker brand personality framework: Method, measurement and conceptual issues. Journal of Strategic Marketing, i i (2), 77-92.

Azoulay, ?., & Kapferer. J. (2003). Do brand personality scales really measure brand personality? Journal of Brand Management, 11(2), 143-155.

Bauer, H. H., Stokburger-Sauer, N. E., & Exler, S. (2008). Brand image and fan loyalty in professional team sport: A refined model and empirical assessment. Journal of Sport Management, 22, 205-226.

Beldona, S., & Wysong, S. (2007). Putting the 'brand' back into store brands: An exploratory examination of store brands and brand personality. Journal of Product and Brand Management, 16, 226-235.

Bosnjak, M., Bochmann, V., & Hufschmidt, T. (2007). Dimensions of brand personality attributions: A person- centric approach in the German cultural context. Social Behavior and Personality, 35, 303-316.

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

Burke, M. (2011, February 14). Under Armour strikes deal with major league baseball. Forbes. Retrieved August 10, 2012, from http://www.forbes.com/sites/monteburke/2011/02/14/ under-armour-strikes-deal-with-major-league-baseball/

Cattell, R. B. (1966). The Scree test for the number of factors. Multivariate Behavioral Research, I, 245-276.

Cho, S. H. (2004). Structural model of brand personality and its application: Comparing brand personality of sport properties and sponsor by using equality constraint modeling. Unpublished doctoral dissertation, University of Connecticut-Storrs.

Comrey, A. L, & Lee, H. B. (1992). A first course in factor analysis (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Association, Inc.

Cutler, M. (2012, August 6). Chevrolet deal set to generate world record $559 million for United. Sport Business. Retrieved August 10, 2012, from http://www.sportbusiness.com/ news/186053/ chevrolet-deal-set-to-generate-world-record-559-million-for-united

d'Astous, ?., & Lévesque, M (2003). A scale for measuring store personality. Psychology & Marketing, 20(5), 455-469.

Fornell, C, & Larcker, D. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.

Gladden, J. M., & Funk, D. C. (2002). Developing an understanding of brand associations in team sport: Empirical evidence from consumers of professional sport. Journal of Sport Management, 1 6(1), 54-81.

Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4(1), 26-42.

Gwinner, K. P., & Eaton, J. (1999). Building brand image through event sponsorship: The role of image transfer. Journal of Advertising, 28(4), 47-57.

Joneson, B., & Christensen, L. (2008). Educational research: Quantitative, qualitative, and mix approaches. Thousand Oaks, CA: Sage Publications.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis. Englewood Cliffs, NJ: Prentice Hall.

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

Hu, L., & Bender, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55.

Kaiser, H. (1970). A second generation little jiffy. Psychometrika, 35, 401-415.

Kerr, A. K., & Gladden, J. M. (2008). Extending the understanding of professional team brand equity to the global marketplace. International Journal of Sport Management and Marketing, 3(1/2), 58-77.

Ministry of Culture, Sports, and Tourism. (2009). 2009 Sport white paper. Seoul, Korea: Ministry of Culture, Sports, and Tourism.

Muthén, L. K., & Muthén, ?. M. (2008). Mplus (version, 5.2). Los Angeles, CA: Muthén & Muthén.

Nederveen Pieterse, J. (2003). Globalization and culture: Global mélange. Lanham, MD: Rowman and Littlefield.

Nunnally, J. C. (1978). Psychometric theory. New York: McGraw-Hill, Inc.

Parent, M., & Séguin, ?. (2008). Toward a model of brand creation for international large-scale sporting events. Journal of Sport Management, 22, 526-549.

Plummer, J. T. (2000). How personality makes a difference. Journal of Advertising Research, 40(6), 79-83.

Rosenberger III, P. J., & Donahay, B. (2008). Brand personality differentiation in Formula One motor racing: An Australian view. Marketing Bulletin, 19, Article 2.

Ross, S. D. (2006). A conceptual framework for understanding spectator-based brand equity. Journal of Sport Management, 20(1), 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 (1), 23-28.

Ross, S. D., James, J. D., & Vargas, P. (2006). Development of a scale to measure team brand association in professional sport. Journal of Sport Management, 20, 260-279.

Satorra, ?., & Bender, P. M. (1994). Corrections to test statistics and standard errors in covariance structure analysis (pp. 399-419). In A. von Eye, & Clogg, C. C. (eds.), Latent variables analysis: Applications for developmental research. Thousand Oaks, CA: Sage

Smith, A. C, Graetz, B. R., & Westerbeek, H. M. (2006). Brand personality in a membershipbased organization. International Journal of Nonprofit and Voluntary Sector Marketing, 11(3), 251-266.

Supphellen, M., & Gr0nhaug, ?. (2003). Building foreign brand personalities in Russia: The moderating effect of consumer ethnocentrism. International Journal of Advertising, 22(2), 203-226.

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

Thompson, B. (2010). Exploratory and confirmatory factor analysis: Understanding concepts and applications. Washington, DC: American Psychological Association.

Yang, Y., & Jo, ?. H. (2002). The study of Korean brand personality scale development and validation. The Korean Journal of Consumer and Advertising Psychology, 3(2), 25-53.