A++ work ( No Plagairism, Quality work on time)

To tweet or not to tweet: the effects of social media endorsements on unfamiliar sport brands and athlete endorsers

Natasha T. Brisona*, Kevin K. Byonb and Thomas A. Baker IIIc

aDepartment of Health & Kinesiology, Texas A&M University, 4243 TAMU, College Station, TX 77845, USA; bDepartment of Kinesiology, School of Public Health, Indiana University, 1025 E. 7th street, SPH 112D, Bloomington, IN, 47405, USA; cDepartment of Kinesiology, University of Georgia, 359 Ramsey Center, 330 River Road, Athens, Georgia 30602, USA

(Received 9 September 2016; accepted 12 September 2016)

Unfamiliar sport brands are consistently searching for ways to gain the interest of consumers and to differentiate the brand’s messages from the huge amount of adver- tising communications consumers receive on a daily basis. One of the most powerful methods used to distinguish one brand from others is an alliance with athlete endor- sers. Companies are actively courting athlete endorsers and their social media accounts to reach consumers. However, little is known about the effects of these alli- ances. Therefore, two studies were conducted: (1) to explore the effectiveness of Twitter endorsement campaigns for unfamiliar sport brands, and (2) to investigate the influence that Twitter endorsements for unfamiliar sport brands have on athlete endorsers. The study revealed that attitude toward the unfamiliar sport brand was positively influenced by the athlete endorser’s Twitter message and that trustworthi- ness was the dominant characteristic influencing credibility of an endorser through Twitter.

Keywords: Twitter; social media; athlete endorsements; brand attitudes; marketing

Introduction

Social media (i.e. social networks and microblogs) is rapidly replacing traditional media as a source to which consumers turn for information about products and brands (Bruhn, Schoenmueller, & Schäfer, 2012). This growing trend is not lost on sport brands as bil- lions have been spent by companies on the implementation and support of social media communications. The extant literature identifies social media as a valuable relationship marketing tool (Eagleman, 2013) that facilitates the formation of ‘meaningful relation- ships [with consumers] through opportunities for communication, interaction, and value’ (Williams & Chinn, 2010, p. 436). With an average of 304 million monthly active users (Statista.com, 2015), Twitter is one of the more popular social media platforms (Jansen, Zhang, Sobel, & Chowdury, 2009). It is a microblogging site that allows users to follow other users for the purpose of sharing information via messages called ‘tweets.’

Twitter’s potential for fostering information-sharing relationships with consumers may be of particular value to unfamiliar sport brands in developing brand knowledge. Unfamiliar brands are those that lack brand knowledge in the minds of consumers because the brands are either new to the market or ones to which consumers have not

*Corresponding author. Email: [email protected]

© 2016 Informa UK Limited, trading as Taylor & Francis Group

Innovation: Management, Policy & Practice, 2016 Vol. 18, No. 3, 309–326, http://dx.doi.org/10.1080/14479338.2016.1237304

been exposed (Campbell & Keller, 2003). Twitter can be a low-cost marketing method for an unfamiliar brand to shape consumer attitudes. However, a problem for unfamiliar sport brands looking to Twitter as a resource for reaching consumers with communica- tions is found in how the platform works. The reach of Twitter-based communications is limited to those who have opted-in to follow the brand and to those following Twitter users who elect to retweet brand-based tweets. Unfamiliar sport brands may lack the Twitter reach needed to foster the creation of brand knowledge in the minds of large numbers of new consumers.

To increase communicative reach and inform more consumers through Twitter, unfa- miliar sport brands may align their products with athlete endorsers. Athlete endorsers are able to reach large numbers of consumers via Twitter, as a top 500 athlete endorser tends to have more than 300,000 followers (Twitter Counter, 2015); while, the top 10 most-followed athletes have between 9.6 million and 38 million followers (Fan Page List, 2015). Additionally, consumers have been found to enjoy the two-way line of com- munication with celebrity athletes provided by Twitter (Cunningham & Bright, 2012), thus allowing brands to reach potential consumers who follow athlete endorsers and to garner consumer support by associating with athlete endorsers (Li & Bernoff, 2011). The question, nonetheless, is whether celebrity athlete Twitter endorsements effectively improve brand attitudes for unfamiliar sport brands.

While scholarly research on social media platforms such as Twitter is expanding (Eagleman, 2013), most studies have examined the influence of social media advertising from a brand management (through marketing) perspective. Three studies stand out as providing a basis for Twitter’s service as a valuable marketing tool for brands. Jansen et al. (2009) analyzed more than 150,000 Twitter postings related to 50 brands over a 13-week period and found that Twitter is an effective online tool for brand-to-consumer (B2C) marketing strategies. Bruhn et al. (2012) compared the influence of traditional media and social media (using Twitter) on brand equity in the tourism, telecommunica- tions, and pharmaceutical industries. The authors also distinguished between firm-created content and user-generated content on Twitter and found different effects on consumers’ perceptions as well as on brand equity based on the source of the infor- mation. They recommended that companies initiate consumer word of mouth about their brands and adopt marketing strategies and tactics that foster communications through social media. Similarly, Wood and Burkhalter (2014) explored brand promotions on Twitter based on the source of the information and compared consumer responses to celebrity-generated tweets and company-generated tweets for an unfamiliar and a famil- iar brand. Their findings revealed that celebrity tweets were successful in acquiring attention and spreading information for brands. They also found that social media celebrities are influential in eliciting interest in unfamiliar brands. Their research, how- ever, did not examine how Twitter impacts the athlete’s credibility. Thus, gaps still exist in the academic literature regarding the effects of social media endorsement campaigns on unfamiliar brands and the athletes who endorse them.

Through two investigations, this study was conducted to bridge the growing body of academic literature focused on: (a) brand decisions to use social media in marketing campaigns, and (b) the literature regarding the effects of celebrity athlete endorsements. The first investigation explores the influence of celebrity athlete Twitter endorsements on attitudes toward unfamiliar sport brands. The second study examines the influence that unfamiliar sport brands have on the credibility of celebrity athlete endorsers.

310 N.T. Brison et al.

Theoretical background

Brand associations and brand attitudes

Brand awareness is frequently achieved through a company’s marketing and advertising efforts. Creating and cultivating brand awareness increases the likelihood that a brand will be part of a consumer’s consideration set, which may include several brands that are contemplated when making a purchase decision (Hutter, Hautz, Dennhardt, & Füller, 2013). This consideration set is developed by creating associations with the brand in consumers’ minds (Keller, 2009). Brand associations can vary in strength, favorability, and uniqueness and are contingent on the marketing campaigns that are linked with a product. Strength is defined by both the quality and quantity of the information pre- sented into the consumer’s memory (Keller, 1993). Strength is also dependent upon the context in which the consumer initially receives the product information and whether the information is actually retained. Brand associations viewed by consumers as positive assist brands in distinguishing themselves from competing brands and thereby influence consumer purchase decisions (Faircloth, Capella, & Alford, 2001). Accordingly, the goal of any marketing campaign for an unfamiliar sport brand should be to improve brand awareness and to create strong, positive brand associations.

Brand associations are categorized into three main types: attributes, benefits, and attitudes (Keller, 1993). Attributes are the descriptive features of a product which may be product-related or non-product related, i.e. internal and external aspects of the pro- duct, which may include the product ingredients and its packaging or types of users, respectively. The benefits are ‘what the consumer thinks the product … can do for them’ (Keller, 1993, p. 4). Benefits can be feelings evoked from using a product or how the product adds to a consumer’s self-concept. The last type of association (brand atti- tudes) is the consumer’s complete assessment or overall evaluation of the brand (Bruhn et al., 2012). Brand attitudes are of great consequence because they often form the foun- dation for consumer behavior (e.g., brand choice and purchase decision) (Keller, 1993). Brand attitudes can also be related to consumer beliefs about product attributes and the benefits received from purchasing the product (Zeithaml, 1988). When consumers believe the brand has attributes and benefits that satisfy their needs and wants, a positive overall brand attitude is formed.

For an unfamiliar sport brand that has minimal or even non-existent associations in the consumers’ mind, garnering consumer attention and developing positive brand atti- tudes may be difficult and costly. Traditionally, associations are established through (1) a consumer’s personal experience with the product or service or (2) information commu- nicated by the company through commercial channels. Associations, additionally, can be created by the brand tying itself to other information in the consumer’s memory that will generate an indirect or secondary connection for the brand. Keller (1993) coined these memories as inferred or secondary associations that allow the characteristics or associations of one brand to transfer to another. Distinctively, ‘secondary associations can be leveraged to create favorable, strong, and unique associations that otherwise may not be present’ (Keller, 1993, p. 12). Secondary associations often arise from a brand utilizing a celebrity spokesperson or an athlete endorser to promote the product or ser- vice. The simultaneous exposure to the two stimuli builds an associative link in the minds of consumers for both the endorser and the brand (Till & Shimp, 1998). This associative link with an endorser enables the unfamiliar brand to leverage the status of the athlete endorser. Thus, allowing consumers to form attitudes toward the unfamiliar brand which can facilitate positive consumer behaviors.

Innovation: Management, Policy & Practice 311

Endorsers, unfamiliar sport brands, and Twitter

The assertion behind the use of athlete endorsers is that these individuals will not only generate awareness and attention to the product/service being advertised, but the charac- teristics associated with them will also be transferred to the product (Till & Shimp, 1998). McCracken’s Meaning Transfer Model (1989) was developed to explain how the characteristics are transferred from the endorser to the consumer, and then to the brand. The author identified that the endorsement process is grounded in the symbolism of the endorser, and distinguished three stages to the process: (a) development of celebrity/ athlete image; (b) transference of meaning from celebrity/athlete to product; and then (c) transference from the product to consumers (McCracken, 1989). The transfer of information from the celebrity/athlete endorser to the product, as well as the brand, is what makes celebrity/athlete endorsements effective (Martin, 1996). The celebrity/athlete endorsement, in turn, allows the unfamiliar brand to build instant brand recognition, to increase the efficiency of the company’s marketing communications, and to obtain high levels of brand awareness (Till & Busler, 2000).

Still, there are limitations with unfamiliar sport brands utilizing traditional athlete endorsers in their marketing campaigns. Even a less popular athlete can be an expensive marketing option for an unfamiliar sport brand, and unfamiliar sport brands need cost- effective options to create brand awareness. Social media (and Twitter in particular) has become a widespread marketing tool to expose consumers to a brand and to develop brand awareness (Hutter et al., 2013). It is also a cost-effective option for unfamiliar sport brands with limited marketing budgets who seek to partner with athlete endorsers. Rather than contracting with the endorser for a multi-level marketing campaign, unfa- miliar sport brands can choose to pay an athlete endorser per Twitter message or tweet. Depending on their popularity, an athlete endorser could be paid as little as $250 (Kornowski, 2013) to more than $300,000 (Milord, 2015) per tweet. For example, former NFL player Terrell Owens received $4,800 per tweet from retailer Old Navy (Brison, Baker, & Byon, 2013).

Twitter allows a consumer to interact with an athlete as if the athlete were a close friend. Through this interaction, the sharing of information will undoubtedly occur, and research indicates that consumers prefer to obtain information from their peers or other known contacts instead of receiving information from the company itself (Harrison- Walker, 2001). The mere fact that an athlete tweets about using a particular brand pro- vides consumers with information regarding product benefits. This, in turn, positively affects the influence of Twitter as a marketing tool (Jin & Phua, 2014) providing the path through which meanings about the athlete endorser will transfer to the product and change consumer attitudes toward the unfamiliar brand. Accordingly, by utilizing an ath- lete endorser’s social media account, unfamiliar brands can increase the probability of positive consumer brand associations and attitudes.

Therefore, the following hypothesis is proposed.

H1: Consumer perceptions of attitude toward the unfamiliar brand will change based on exposure to endorser messages through Twitter.

Brand alliances and endorser credibility through Twitter

Brand alliances can produce positive and negative effects for both the sport brand and the athlete endorser. In the case of the athlete endorser, the possibility may exist that the qualities of the unfamiliar sport brand could negatively influence consumer perceptions

312 N.T. Brison et al.

of the endorser. This possibility is grounded in the reciprocal transfer of meanings that flow through paths established in consumer minds via brand alliances. The possibility that negative information about the brand may transfer to the endorser is well estab- lished in the extant literature; as evidenced in studies by Aaker and Keller (1990) and Halonen-Knight and Hurmerinta (2010). Furthermore, a study by Arsena, Silvera, and Pandelaere (2014) revealed that an endorser’s perceived attributes are influenced by the products they endorse. Thus, there may be negative effects on the endorser’s credibility through a partnership with a brand with which consumers are not familiar.

Factors influencing an endorser’s credibility through Twitter

An athlete’s credibility influences the persuasiveness of the information or message (Seno & Lukas, 2007), and fans typically view endorsers as credible sources of informa- tion about the products they endorse (Goldsmith, Lafferty, & Newell, 2000). The credi- bility and effectiveness of the endorser is ascribed to the endorser’s attractiveness, expertise, and trustworthiness (Shank, 2009). These salient characteristics of the endor- ser are also essential in developing consumer attitudes toward the unfamiliar brand.

Athlete endorser attractiveness

Athlete endorsement literature has revealed that attractiveness is a significant gauge of endorser effectiveness (Amos, Holmes, & Strutton, 2008; Till & Busler, 2000). Attrac- tiveness is multi-dimensional in nature, encompassing physical characteristics as well as non-physical characteristics (Erdogan, 1999; Shank & Langmeyer, 1994). Kahle and Homer (1985) researched endorser attractiveness and found that attractive endorsers influenced consumer attitudes more than less attractive endorsers, while, Kim and Na (2007) found that attractiveness is contingent upon the familiarity, similarity, and liking of an endorser. If consumers closely identify with or like the athlete endorser, they will often consider the athlete as attractive (Amos et al., 2008). In the context of social media, consumers are more inclined to follow athletes based on some degree of famil- iarity or liking (i.e., attractiveness) (Witkemper, Lim, & Waldburger, 2012). Therefore, the authors posit that:

H2a: Consumer perceptions of athlete endorser attractiveness will change based on exposure to athlete endorser messages through Twitter.

Athlete endorser expertise

Athlete endorser expertise affects consumer attitudes and decisions more than athlete endorser attractiveness or any other factor (Ohanian, 1990; Till & Busler, 2000). Source expertise is the degree to which a communicator is observed to be a source of valid claims (Hovland, Janis, & Kelley, 1953). If the receiver perceives the source to be an expert in the requisite subject, source effectiveness will be positively influenced. Hence, consumers subjected to an endorsement by an athlete perceived as having high expertise will display more accordance with the athlete’s recommendation (Ohanian, 1990), and the authors propose that:

H2b: Consumer perceptions of athlete endorser expertise will change based on exposure to athlete endorser messages through Twitter.

Innovation: Management, Policy & Practice 313

Athlete endorser trustworthiness

Endorser expertise is closely tied to trustworthiness, and an endorser’s objective for pre- senting information is a key component in the consumer’s assessment of the informa- tion’s trustworthiness (Romani, 2006). Source trustworthiness is the consumer’s belief that the source will provide impartial and truthful information (Ohanian, 1990). If the athlete endorser is trustworthy, consumers are less likely to scrutinize the communica- tion and will accept the information as accurate. Trustworthiness is also linked to a con- sumer’s belief that the statements about a product or brand in the endorsement reflect the actual opinion of the athlete endorser. Priester and Petty (2003) examined the influ- ence of endorser trustworthiness on the persuasiveness of a message and found that information provided by trustworthy endorsers was more likely to be accepted without skepticism. Additionally, endorsers who are perceived as trustworthy have been shown to be more effective in eliciting attitude changes in consumers than untrustworthy endor- sers (Ohanian, 1990). As a result, the authors posit that:

H2c: Consumer perceptions of athlete endorser trustworthiness will change based on expo- sure to athlete endorser messages through Twitter.

To test the hypotheses, two studies were conducted. Study 1 tests the two-way non-directional hypotheses assessing the effectiveness of a fictitious athlete endorser’s Twitter messages. Study 2 tests the hypotheses via a real athlete endorser’s Twitter messages. Both studies examine the potential changes in the consumer attitudes toward the unfamiliar brand and the potential changes to the endorser’s credibility based on an athlete endorser’s Twitter messages.

Study 1 (fictitious athlete)

Method

Research design

A within-group experimental design was chosen to evaluate differences in consumer atti- tudes toward the unfamiliar brand and the athlete endorser’s characteristics after the par- ticipant was exposed to the athlete endorser’s Twitter messages. Prior to the experiment, a pretest was conducted to create a profile of the fictitious athlete endorser, the stimuli (the tweets or Twitter messages), and the product for the messages.

Creation of a credible fictitious athlete. A fictitious athlete endorser was used for the first study to avoid the influence of pre-existing associations that participants have with celebrity athletes. This is in accordance with previous source credibility and adver- tising research studies, which found that the use of familiar endorsers in a study creates ‘a significant amount of variation in the subjects’ knowledge and attitude’ toward the endorser (Till & Busler, 2000, p. 4). The data collection for the study was based in the United States, where American football, basketball, and baseball are prevalent among sports fans. Soccer, however, is rapidly emerging as a top sport in the US, but most consumers are still unfamiliar with many of the sport’s best athletes (International Soccer, 2013). Consequently, using this sport best supported the ruse among US survey participants.

On the basis of pretesting, a Premier League reserve player for the Newcastle Club, Bradden Inman, was selected as the fictitious endorser. Inman was chosen through a two-step process. Seven photographs of players from the Newcastle United reserves

314 N.T. Brison et al.

were presented to a focus group of five sport management professionals who had watched at least ten Premier League matches over the past 12 months. The group was instructed to rank the photographs according to familiarity. The top three least familiar athletes were then rated on attractiveness, trustworthiness, and expertise. The group was then charged with creating a fictitious name and profile for the athlete endorser to be used in the study. The group decided on the name Bradley Wolcott and an association set was created that described Wolcott as one of the brightest new stars in the Premier League. A separate focus group was created to evaluate Wolcott based on attractiveness (M = 6.25, SD = 1.83), trustworthiness (M = 5.96, SD = 1.51), and expertise (M = 6.59, SD = 1.87). Based on a nine-point scale, the descriptive statistics for each of the endor- ser characteristics demonstrated that overall participants believed Wolcott to be an actual soccer player.

Selection of the brand. The authors selected a brand that was under development by a company called Union Square. The authors chose an athletic product for endorsement to facilitate the formation of a match-fit between the athlete and the product in the minds of consumers. An association set was created for the product, Core, that described the product as a post-workout body cleansing soap that would be in stores in early 2016.

Stimulus material. For the social media manipulation, the promotional tweets were created using a focus group of five marketing professionals in the sport industry who were asked to draft a list of promotional tweets. A separate focus group was then asked to rank the tweets according to informational content and overall message appeal. The following three items were selected:

(1) Just tried @Core by #UnionSquare. #exhilarating#recharging (2) Hard practice. Just showered with @Core by #UnionSquare. #clean#cool#invig-

orating (3) Good match! Time to hit the showers @Core #coolandrefreshing#rejuvenating

Participants. Twitter users were recruited using undergraduate and graduate student databases from two sport management programs located in the southeastern US. Students were emailed a link requesting their participation in the study. A snowball sampling design was utilized, encouraging participants to forward the survey to their friends and acquaintances. This design was most efficient for targeting a broad population of Twitter followers, as college age students make up 37% of Twitter users (Duggan, Ellison, Lampe, Lenhart, & Madden, 2015).

Upon consenting to take the survey, each participant was asked basic demographic information. Given that this study was about Premier League athlete endorser effective- ness through Twitter, two screening questions were used to verify eligibility for the sur- vey: (a) Twitter use and (b) those who had watched at least one Premier League match within the past 12 months.

A total of 155 responses were collected; nine surveys were eliminated due to exces- sive missing values, resulting in 146 usable surveys. In terms of participants’ demo- graphic characteristics, gender was predominately male (68.5%). Approximately 92% were aged between 18 and 34. For the ethnicity, a little under half of participants were White/Caucasian (47.2%). For participant’s Twitter usage, over 43% stated that they vis- ited Twitter at least once a day.

Instruments. The questionnaire was designed to measure source attractiveness, source expertise, source trustworthiness, and attitude toward the unfamiliar brand. The

Innovation: Management, Policy & Practice 315

scales were adapted from existing studies in advertising and consumer behavior literature. The measures for source attractiveness, source expertise, source trustworthi- ness were developed from Ohanian’s (1990) 15-item, nine-point semantic differential scale to evaluate celebrity endorsers. Source attractiveness consisted of the following anchors: unattractive/attractive, not classy/classy, ugly/handsome, plain/elegant, and not sexy/sexy. An example of a source attractiveness item is, ‘Select the point between the two adjectives that best represents your attitude toward the athlete. Bradley Wolcott is:’. Source expertise included: not an expert/expert, inexperienced/experienced, unknowl- edgeable/knowledgeable, unqualified/qualified, and unskilled/skilled. The items measur- ing source trustworthiness consisted of the following anchors: undependable/dependable, dishonest/honest, unreliable/reliable, insincere/sincere, and untrustworthy/trustworthy. The reliability of the scale was assessed by calculating Cronbach’s alpha coefficients for each construct. The results were: attractiveness = 0.93, expertise = 0.93, and trustworthi- ness = 0.98, indicating the constructs were found to be reliable. Seven items from Till and Shimp’s (1998) nine-point semantic differential scale were utilized to measure attitude toward the unfamiliar brand on Twitter. An example of an attitude toward the brand item is ‘My overall attitude about Core by Union Square is:’. The anchors included: favorable/unfavorable, positive/negative, strongly like/dislike, good/bad, high quality/low quality, superior/inferior, and effective/ineffective. The Cronbach’s alpha was 0.96, indicating that the items representing the construct were internally consistent.

Procedure

Participants were informed that a company had developed an athletic body cleansing system called Core which would be in stores early 2016. The company was also inter- ested in using Premier League players to promote the new product through social media. The survey participants’ opinions would aid the company in determining the effectiveness of using the players and social media to promote the product. This ruse formed the basis for participants to (1) read initial information about the athlete endorser and Core; (2) read Twitter posts by the athlete about Core; and (3) complete pre and post measures of attitude toward the athlete and attitude toward the unfamiliar brand.

After reading the association set, participants were asked to recall specific informa- tion about the athlete and the product, such as ‘Bradley Wolcott plays for what sport club?’ and ‘Core is looking to sign players from what sport?’ The ‘fictitious athlete’ participants recalled 89% of the facts about the athlete and 95% of the facts about the brand, indicating that overall the participants had acquired the necessary information about the athlete and the brand.

The first evaluation was completed before the participants were exposed to the Twit- ter campaign. Participants were asked to read the association set about Core and assess their overall attitudes about the brand based on the Till and Shimp’s (1998) scale, and on the athlete endorser based on Ohanian’s (1990) 15-item, nine-point semantic differen- tial scale to assess celebrity endorsers. Additionally, participants were asked to evaluate athlete and product congruence (‘As an endorser for Core by Union Square, I think Bradley Wolcott is …’ and ‘I think the combination of Core by Union Square and Bradley Wolcott is …’) based on five questions with bi-polar endpoints and nine-point scales (i.e., inappropriate/appropriate, ineffective/effective, do not belong together/belong together, do not go together/go together, do not fit together, fit together) (Till & Busler, 2000). The average scores for the fictitious athlete ranged from 6.31 (SD = 1.90) to 6.54

316 N.T. Brison et al.

(SD = 1.84). These results revealed that most of the participants believed the product to be a good match with the athlete.

The social media campaign for Core by Union Square with the athletes was estab- lished by providing participants with a series of Twitter messages (tweets). Participants were asked to review the three tweets and then re-evaluate the athlete and the unfamiliar brand based on the aforementioned scales.

Data analysis

Using SPSS 20.0 a one-way repeated measures multivariate analysis of variance (RM-MANOVA) was conducted on source expertise, source attractiveness, source trust- worthiness, and attitude toward the unfamiliar brand. The pre-evaluations by the partici- pants were treated as independent variables, and the post-evaluations (after exposure to the tweets) were examined as dependent variables. Table 1 includes the means and standard deviations for the four variables evaluated with the fictitious athlete endorser.

Results

When the a priori alpha was set at 0.05, there was not a statistically significant effect of the Twitter message on the endorsement effects (attractiveness, trustworthiness, exper- tise) and attitude toward the unfamiliar brand (Wilks’ Lambda = 0.941, F(2, 142) = 2.24, p = 0.067). However, a statistically significant effect of the Twitter message on the variables was observed in a more liberal condition (i.e., p < 0.10).

Four follow-up univariate ANOVAs (one for each dependent variable) were calcu- lated. For attitude toward the unfamiliar brand, the result was statistically significant for within subjects effect after exposure to the Twitter messages (F(1, 145) = 5.13, p = 0.025, partial η2 = 0.034). This supports H1 that exposure to athlete Twitter mes- sages will change consumer attitudes toward the unfamiliar brand. For source attractive- ness, there was no significant difference (F(1, 145) = 1.71, p = 0.193, partial η2 = 0.012). This result disconfirms H2a that exposure to endorser Twitter messages will change consumer perceptions of athlete endorser attractiveness. For source expertise, there was no significant difference within subjects effect after exposure to the Twitter messages (F(1, 145) = 2.77, p = 0.098, partial η2 = 0.019), identifying that exposure to athlete Twitter messages did not change consumer perceptions of athlete endorser

Table 1. Means, standard deviations (SD), Cronbach’s alpha, and mean differences for the fictitious endorser.

Fictitious athlete n = 146

Pre (SD) Post (SD) Cronbach’s

alpha

Mean difference (Std. Error) Sig.

Source attractiveness 6.03 (1.57) 6.16 (1.76) 0.937 0.136 (0.104) 0.193 Source trustworthiness 5.98 (1.44) 6.26 (1.73) 0.980 0.282 (0.123) 0.023* Source expertise 6.18 (1.72) 6.36 (1.75) 0.931 0.180 (0.108) 0.098 Attitude toward the unfamiliar brand

6.36 (1.61) 6.64 (1.41) 0.967 0.281 (0.124) 0.025*

Note: *p < 0.05.

Innovation: Management, Policy & Practice 317

expertise (H2b not supported). Only one endorser characteristic was found to be statistically significant between the pre- and post-tests: trustworthiness (F(1, 145) = 5.30, p = 0.023, partial η2 = 0.035). Hence, this provides support for H2c that exposure to athlete Twitter messages will change consumer perceptions of athlete endorser trustworthiness and attitude toward the brand.

Discussion

Study 1 provides some interesting results related to the fictitious athlete endorser’s effec- tiveness on an unfamiliar brand through Twitter. The consumer’s attitudes toward the unfamiliar brand did change after being exposed to the athlete endorser’s tweets. Con- sumers had more favorable attitudes toward the unfamiliar brand after being exposed to the athlete endorser’s tweets about the brand. Therefore, unfamiliar brands should strongly consider using Twitter and endorsers to influence consumer perceptions about the brand. The consumer’s perceptions of the athlete’s trustworthiness was also statisti- cally significant. The endorser appeared more trustworthy after tweeting about the prod- uct. This supports previous literature that these alliances can be advantageous for an athlete endorser’s image (Brison et al., 2013; Jin & Phua, 2014).

Study 2 (real athlete)

Method

The purpose of the second study was to test the hypotheses 1, 2a, 2b, and 2c from Study 1 with a real athlete endorser who is active on Twitter using the same methods from Study 1 with a different sample. The use of a real athlete endorser in Study 2 was used to provide better ecological validity than the use of the fictitious athlete endorser in Study 1.

Real athlete selection

Robin Van Persie, a player for Manchester United Football Club, was selected on the basis of a pretest, in which six Premier League players were rated on the three credibil- ity characteristics of an endorser: attractiveness; trustworthiness; and expertise. The test was administered at a southeastern university in the US using undergraduate and gradu- ate sport management students, who had watched at least one Premier League match within the past 12 months. The six athletes had previously been identified by sport ana- lysts as having high credibility (i.e., Robin Van Persie, Gareth Bale, Steven Gerrard, Luis Suarez, Frank Lampard, and Wayne Rooney) (‘Our panel of experts count down the top 50 Premier League players ahead of 2013–2014 season,’ 2013’). Participants were asked to rank the athletes according to their perceptions of who was most credible. Two athletes were observed to be high in attractiveness, trustworthiness, and expertise; Steven Gerrard and Robin Van Persie.

Based on the fact that this study aimed to analyze athlete endorser effectiveness through Twitter, and Steven Gerrard did not have a personal Twitter account, Robin Van Persie was chosen for this study. Researchers also confirmed that Robin Van Persie was actively using social media for endorsement purposes. For example, Van Persie tweeted to his 3.4 million Twitter followers about his new football (soccer) boots: ‘Really happy with my new @adidasfootball boots. Designed them myself on miadidas’ (‘Robin van

318 N.T. Brison et al.

Persie (Persie_Official) on Twitter,’ 2013). This fact supported the study ruse that Van Persie would actually tweet about a newly endorsed product or brand. Information from Robin Van Persie’s Manchester United player profile was then used to create the association set for the study.

Participants

As with Study 1, Twitter users were recruited using undergraduate and graduate student databases from two sport management programs located in the Southeastern US. The authors adopted the two same screening questions and snowball sampling used in Study 1 to collect data. As a result, a total of 168 responses were collected; 11 surveys were eliminated due to an excessive number of incomplete responses, resulting in 157 usable surveys. In terms of participants’ demographic characteristics, gender was predominately male (70.7%). Approximately 90% were aged between 18 and 34. For the ethnicity, a little over half of participants were White/Caucasian (52.9%). For participant’s Twitter usage, over 50% stated that they visited Twitter at least once a day.

Instruments and procedure

The same instrument for Study 1 was used in Study 2, with the substitution of the real athlete’s association set and name in the survey items. Identical data collection proce- dures were implemented. The reliability of the scales via the Cronbach’s alpha coeffi- cients for each construct were: attitude toward the unfamiliar brand = 0.96, attractiveness = 0.90, expertise = 0.97, and trustworthiness = 0.96, signifying the con- structs were found to be reliable for the study.

To gauge prior knowledge about the real athlete, participants were asked questions about their familiarity with Robin van Persie. Approximately 60% of the participants rated Van Persie with at least a ‘5’ or above on a nine-point scale regarding familiarity, indicating that many of the participants were entering the study with some familiarity and prior information about the athlete. Even though most of the participants were familiar with Robin van Persie, all participants were provided with an association set for both the athlete and the unfamiliar brand. After reading the association sets, partici- pants were asked to recall specific information about the athlete and the product such as ‘Robin van Persie plays for what sport club?’ and ‘Core is looking to sign players from what sport?’ The ‘real athlete’ participants recalled 87% of the facts about Robin van Persie and 94% of the facts about Core, denoting overall the participants had acquired the necessary information about the athlete and the unfamiliar brand.

The participants were then asked questions about the fit of the real athlete with the unfamiliar brand. The mean scores for the real athlete ranged from 6.64 (SD = 2.02) to 7.04 (SD = 2.12), indicating that most of the participants believed that brand to be a good match with the athlete. Next, participants were provided with a series of Twitter messages (tweets) by Robin van Persie about the brand Core. Participants were asked to review the three tweets and then re-evaluate the athlete and the product based on the scales from Study 1.

Data analysis

As with Study 1, data were analyzed using SPSS 20.0, and a one-way repeated measures multivariate analysis of variance (RM-MANOVA) was conducted on the four

Innovation: Management, Policy & Practice 319

variables. The pre-evaluations were treated as independent variables, and the post- evaluations were dependent variables. Table 2 includes the means and standard deviations for the four variables examined with the real athlete endorser.

Results

The results revealed a statistically significant effect of the Twitter messages on the endorsement characteristics (i.e., attractiveness, trustworthiness, expertise) and attitude toward the unfamiliar brand (Wilks’ Lambda = 0.226, F(4, 153) = 130.75, p < 0.001; see Table 2).

Four follow-up univariate ANOVAs (one for each dependent variable) were calcu- lated. For attitude toward the unfamiliar brand, there was a significant difference for within subjects effects after exposure to the Twitter messages (F(1, 156) = 9.86, p = 0.002, partial η2 = 0.059), confirming H1 that exposure to athlete Twitter messages will change consumer perceptions of attitude toward the unfamiliar brand. For source attractiveness, source expertise, and source trustworthiness, the results showed signifi- cant differences within subjects effects after exposure to the Twitter messages: attractive- ness (F(1, 156) = 25.97, p < 0.001, partial η2 = 0.143), expertise (F(1, 156) = 425.96, p < 0.001, partial η2 = 0.732), and trustworthiness (F(1, 156) = 27.73, p < 0.001, partial η2 = 0.151). These results provide support for H2a, H2b, and H2c that exposure to athlete Twitter messages will change consumer perceptions of athlete endorser attractiveness, expertise, and trustworthiness.

Discussion

Upon examining the real athlete endorser’s effectiveness through Twitter, the results determined that the consumer’s attitudes toward the unfamiliar brand changed after being exposed to the athlete endorser’s tweets. This confirms the results of Study 1 that unfamiliar brands can greatly benefit from partnering with an athlete endorser through Twitter. All of the consumer’s perceptions of the athlete endorser’s characteristics (at- tractiveness, trustworthiness, and expertise) also changed after consumers were exposed to the tweets. The results were statistically significant for each of the endorser’s credibil- ity characteristics, supporting prior research that brand attributes can influence consumer perceptions of the endorser (Halonen-Knight & Hurmerinta, 2010).

Table 2. Means, standard deviations (SD), Cronbach’s alpha, and mean differences for the real endorser.

Real athlete n = 157

Pre (SD) Post (SD) Cronbach’s

alpha Mean difference (Std. Error) Sig.

Source attractiveness 6.01 (1.76) 5.35 (.924) 0.900 −0.656 (0.129) 0.000** Source trustworthiness

5.99 (1.96) 6.43 (2.07) 0.961 0.441 (0.084) 0.000**

Source expertise 7.58 (1.81) 5.42 (.975) 0.975 −2.156 (0.104) 0.000** Attitude toward the unfamiliar brand

6.61 (1.55) 6.96 (1.56) 0.965 0.357 (0.114) 0.002*

Note: *p ˂ 0.05; **p ˂ 0.001.

320 N.T. Brison et al.

General discussion

The purpose of this study was to examine the effect of an athlete endorser Twitter mes- sages on an unfamiliar brand and the endorser’s attractiveness, trustworthiness, and expertise. The results indicated that incorporating athlete endorsements with social media campaigns can have a positive effect on the endorsed unfamiliar brand. This study supported previous studies which found that social network marketing is an acceptable medium to engage consumers to influence perceptions of a brand (Bolotaeva & Cata, 2010; Du, Bhattacharya, & Sen, 2010).

With both the fictitious athlete and the real athlete endorsers, the attitudes toward the unfamiliar brand were favorable prior to exposing the consumers to the Twitter mes- sage, and unfamiliar brand attitudes significantly increased after the consumer was exposed to the athlete endorser’s Twitter messages (see Table 1). This supported H1, that exposing consumers to an unfamiliar brand through an athlete’s Twitter feed can lead to increased consumer perceptions of the brand. When a consumer chooses to fol- low an athlete on Twitter, a relationship is forged with the athlete. Depending on the strength of the tie that the athlete has with the consumer, the athlete can positively influ- ence consumer perceptions about a brand (Gupta, 2007; Lafferty, Goldsmith, & Newell, 2002). Moreover, the findings also advocated using an athlete to introduce the consumer to new or unfamiliar brands. Till, Stanley, and Priluck (2008) found that individuals who were exposed to a new or unfamiliar brand through an endorser had more positive attitudes toward the brand than those who were introduced to the brand without an endorser. If the consumer has positive perceptions of the athlete regarding attractiveness, trustworthiness, and expertise, these attitudes can transfer from the athlete to the unfa- miliar brand based on messages conveyed by the athlete endorser on Twitter, which is in line with results from Ohanian’s (1990) study.

Following examination of the athlete endorser’s credibility characteristics, both stud- ies found that trustworthiness was a predominant characteristic that influenced athlete credibility. When an endorser is recognized to be trustworthy, information presented by the endorser is more useful in altering attitudes about the athlete and the brand (Stevens, Lathrop, & Bradish, 2003). Similar findings were reported by Cunningham and Bright (2012); trustworthiness was an important factor that consumers rely upon in determining athlete endorser credibility. The findings of the study were not as predictive for athlete endorser expertise and attractiveness after consumers were exposed to the Twitter mes- sages. However, Parker and Fink (2012) determined that an athlete endorser can still be effective, even if he/she is deficient of one or more of the credibility characteristics. Consumers will, in the end, identify the athlete endorser as credible based on the total effect of the characteristics.

For the fictitious, or up-and-coming athlete, the findings indicated that consumers thought more favorably of the athlete endorser after he was tied to a brand, even a brand that was unfamiliar. There were statistically significant differences for the credibil- ity characteristics. So, for unfamiliar athletes, partnering with an appropriate brand can enhance the image of the athlete. The relationship, conversely, may not always be bene- ficial to the athlete endorser, as our results showed that consumers thought less of the real athlete after he posted the messages on Twitter. This may have been due to the pre- existing knowledge that consumers had regarding the athlete. Approximately 60% of the participants were familiar with van Persie, and it is possible that their scores greatly influenced the means of those who were not familiar with the athlete. Additionally, par- ticipants may not have believed the athlete had the expertise to promote such a product,

Innovation: Management, Policy & Practice 321

although a majority of the participants believed the athlete and product to be a good match when presented with product congruency questions. Matching the characteristics of the athlete to those of the product should bolster the qualities of both the endorser and the product or brand in the mind of the consumer (Braunstein & Zhang, 2005), and a stronger sport-related product, e.g., an athletic shoe, may have revealed different results. Another potential reason why the real athlete was seen as less credible after he posted the Twitter messages may be due to consumer’s believing that the athlete was only endorsing the product for financial reasons. Halonen-Knight and Hurmerinta (2010) cautioned that such a belief by consumers could be dangerous to the credibility of the athlete endorser, and when choosing endorsement partnerships the athlete should strive to not only maintain their credibility but also their integrity with consumers. Although the opportunity may appear to be mutually beneficial to the athlete and the brand, there may be positive and negative implications to the athlete’s image and brand equity.

Implications

The findings from this study produced several theoretical, practical and regulatory impli- cations. Theoretically, this study expanded on current research into social media, sport consumer behavior, and athlete endorsers. Social media has been heralded as an effec- tive marketing tool for brands (Eagleman, 2013), and this study supports the notion that consumer attitudes toward a brand can be positively influenced through social media. Additionally, this study expands on the Meaning Transfer Theory in its application to social media. The effectiveness of an athlete endorser is based on the meanings he/she carries to the endorsement process. This may be due to both on-field and off-the-field performance (Koo, Ruihley, & Dittmore, 2012). These meanings are then transferred to the endorsed product and the brand. In the case of social media, as with traditional advertising mediums, meanings are also transferred from the athlete to the brand and then to the consumer. In particular, the perceptions of the endorser transfer from the endorser to the unfamiliar brand and these perceptions become a component of the unfa- miliar brand’s association set. Through the application of the Meaning Transfer Theory, the athlete endorser can be a catalyst to develop and reinforce unfamiliar brand atti- tudes.

From a practical perspective, unfamiliar sport brands should actively engage in prac- tices to incorporate endorsers in their advertising and marketing campaigns. The linking of an athlete endorser with an unfamiliar or new brand allows both parties to leverage positive consumer perceptions and enhance consumer attitudes of the athlete and the brand (Boyd & Shank, 2004). In general, athlete endorsers who actively use Twitter are influential in drawing attention to unfamiliar brands. With the unfamiliar brand, the information provided by the athlete endorser may be the only information provided to the consumer, and a well-placed tweet from the athlete endorser will aid in distinguish- ing the unfamiliar brand from its competitors (Wood & Burkhalter, 2014). Sport mar- keters should, therefore, incorporate social media campaigns into their marketing and promotional activities, with the design to partner with athlete endorsers and drive con- sumers to their Twitter pages.

Sport marketers for unfamiliar brands also need to focus on endorser–brand congru- ence. As a result of high endorser–brand congruence, consumers will attribute the perti- nent attributes of the endorser with the pertinent attributes of the brand (Ilicic & Webster, 2011). If there is low endorser–brand congruence, the appeal of the athlete to

322 N.T. Brison et al.

the sport consumer will likely be diminished (Lee & Koo, 2015), and the association will be contrary to the initial rationale for the alliance: i.e., drawing attention to adver- tising and marketing messages, increasing consumer awareness of a product, enhancing brand recall, and influencing consumer purchase decisions regarding the product (Zafar & Rafique, 2012). Overall, the endorser–brand relationship is rooted in the unfamiliar brand believing that the endorser can increase consumer attitudes toward the unfamiliar brand, which will lead to purchase intention and purchase behavior (Spears & Singh, 2004). Twitter can provide additional insight into this relationship and intensify the like- lihood of purchase; this, undoubtedly, can have a significant impact on advertising and marketing practices.

The study also shed light on the positive and negative repercussions of brand alli- ances for athlete endorsers. Through endorsement opportunities, athletes can increase their credibility with consumers which, in turn, will increase the marketability and potential value of the athlete on and off the field. Yet there is also potential for con- sumers to develop negative perceptions of the athlete due to the brand alliance. Athletes should be diligent in researching the brands they choose to endorse. The matching of an athlete to a brand is important for the athlete, and both endorser-related and unrelated products can have effects on the endorser’s appropriateness for the alliance (Wright, 2016).

From a regulatory perspective, it is important for both athletes and unfamiliar brands to maintain a level of trustworthiness with consumers by providing accurate information about the products and brands. Since trustworthiness was the dominant endorser credi- bility characteristic, companies should regularly monitor athlete endorser social media campaigns and seek to educate the endorsers regarding the Federal Trade Commission’s (FTC) Guides to Endorsements and Testimonials (Brison et al., 2013; McKelvey & Masteralexis, 2013). The FTC continues to be vigilant in monitoring social media mar- keting campaigns that utilize athlete endorsers, and the last thing a company wants, or needs, is a false advertisement investigation by the FTC. The FTC consistently provides companies with access to information to prevent violations of the Federal Trade Com- mission Act. In 2014, the FTC updated its Dot.com Disclosure Guidelines to emphasize that disclosures must be clear and conspicuous to avoid deception. This also includes an endorser’s disclosure that he/she has been paid to provide comments or testimonials about a brand through social media. Violations could result in modification or discontin- uance of a campaign as well as monetary fines (Federal Trade Commission, 2014). Based on the potential influence that athlete endorsers could have on consumers, mar- keters should be cognizant of the FTC rules and guidelines regarding disclosure of these relationships.

Limitations and further research

Although the study expanded on the current research regarding unfamiliar brands, endorser effectiveness, and social media, several limitations existed. First, a student sample was used in the study, and this may affect the generalizability of the results. However, college students make up close to 40% of Twitter users (Duggan et al., 2015), which provides a justification for the selection of this population. Since a screening question was used to identify Twitter users and those familiar with the Premier League, the results of the study are of strong ecological validity, but future studies should focus on collecting data from a more generalizable population to further enhance population validity (Hair, Black, Babin, & Anderson, 2010). Second, approximately 40% of the

Innovation: Management, Policy & Practice 323

participants reported that they checked Twitter daily or several times a day; nonetheless, participants were not asked about whether they followed an athlete on Twitter. They were also not asked about their attitudes toward Twitter or toward advertising, either of which may have explained the decrease in participant perceptions of athlete attractive- ness and expertise. Future research should also examine the effects of positive and nega- tive brand perceptions on the athlete endorser so as to increase understanding on the influence of brand reputation and image on endorser credibility. Third, this study only examined one type of brand. Additional studies should be conducted with varying brands as well as types of athlete endorsers to broaden the generalizability of the find- ings. Lastly, the study only examined the use of Twitter for endorsement effectiveness. Future studies should include other forms of social media, such as Facebook and Insta- gram, as well as other variables that may influence endorser effectiveness and attitudes toward unfamiliar brands.

References Aaker, D. A., & Keller, K. L. (1990). Consumer evaluations of brand extensions. The Journal of

Marketing, 54, 27–41. Amos, C., Holmes, G., & Strutton, D. (2008). Exploring the relationship between celebrity endor-

ser effects and advertising effectiveness. International Journal of Advertising, 27, 209–234. Arsena, A., Silvera, D. H., & Pandelaere, M. (2014). Brand trait transference: When celebrity

endorsers acquire brand personality traits. Journal of Business Research, 67, 1537–1543. Bolotaeva, V., & Cata, T. (2010). Marketing opportunities with social networks. Journal of Inter-

net Social Networking and Virtual Communities, 2010, 1–8. Boyd, T. C., & Shank, M. D. (2004). Athletes as product endorsers: The effect of gender and pro-

duct relatedness. Sport Marketing Quarterly, 13, 82–93. Braunstein, J. R., & Zhang, J. J. (2005). Dimensions of athletic star power associated with Generation

Y sports consumption. International Journal of Sports Marketing and Sponsorship, 6, 242–267. Brison, N. T., Baker, T. A., III, & Byon, K. K. (2013). Tweets and crumpets: Examining U.K.

and U.S. regulation of athlete endorsements and social media marketing. Journal of Legal Aspects of Sport, 23, 55–71.

Bruhn, M., Schoenmueller, V., & Schäfer, D. B. (2012). Are social media replacing traditional media in terms of brand equity creation? Management Research Review, 35, 770–790.

Campbell, M. C., & Keller, K. L. (2003). Brand familiarity and advertising repetition effects. Journal of Consumer Research, 30, 292–304.

Cunningham, N., & Bright, L. F. (2012). The tweet is in your court: Measuring attitude towards athlete endorsements in social media. International Journal of Integrated Marketing Commu- nications, 4, 73–87.

Du, S., Bhattacharya, C. B., & Sen, S. (2010). Maximizing business returns to Corporate Social Responsibility (CSR): The Role of CSR Communication. International Journal of Manage- ment Reviews, 12, 8–19.

Duggan, M., Ellison, N. B., Lampe, C., Lenhart, A., & Madden, M. (2015, January 9). Demo- graphics of key social networking platforms. Retrieved from http://www.pewinternet.org/2015/ 01/09/demographics-of-key-social-networking-platforms-2/

Eagleman, A. N. (2013). Acceptance, motivations, and usage of social media as a marketing com- munications tool amongst employees of sport national governing bodies. Sport Management Review, 16, 488–497.

Erdogan, B. Z. (1999). Celebrity endorsement: A literature review. Journal of Marketing Manage- ment, 15, 291–314.

Faircloth, J. B., Capella, L. M., & Alford, B. L. (2001). The effect of brand attitude and brand image on brand equity. Journal of Marketing Theory and Practice, 9, 61–75.

Fan Page List. (2015). Top athletes on Twitter. Fanpagelist.com. Retrieved from http://fanpage list.com/category/athletes/view/list/sort/followers/

Federal Trade Commission. (2014). FTC issues changes to some civil penalty amounts. FTC.gov. Retrieved from https://www.ftc.gov/news-events/press-releases/2014/03/ftc-issues-changes-some- civil-penalty-amounts

324 N.T. Brison et al.

Goldsmith, R. E., Lafferty, B. A., & Newell, S. J. (2000). The impact of corporate credibility and celebrity credibility on consumer reaction to advertisements and brands. Journal of Advertis- ing, 29, 43–54.

Gupta, D. K. (2007). Impact of celebrity endorsement on consumer buying behaviour and brand building. SSRN eLibrary. Retrieved from http://papers.ssrn.com/sol3/papers.cfm?abstract_id= 1203322

Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010) Multivariate data analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.

Halonen-Knight, E., & Hurmerinta, L. (2010). Who endorses whom? Meanings transfer in celebrity endorsement. Journal of Product & Brand Management, 19, 452–460.

Harrison-Walker, L. J. (2001). The measurement of word-of-mouth communication and an investigation of service quality and customer commitment as potential antecedents. Journal of Service Research, 4, 60–75.

Hovland, C. I., Janis, L. I., & Kelley, H. H. (1953). Communication and persuasion. New Haven, CT: Yale University Press.

Hutter, K., Hautz, J., Dennhardt, S., & Füller, J. (2013). The impact of user interactions in social media on brand awareness and purchase intention: The case of MINI on Facebook. Journal of Product & Brand Management, 22, 342–351.

Ilicic, J., & Webster, C. M. (2011). Effects of multiple endorsements and consumer–celebrity attachment on attitude and purchase intention. Australasian Marketing Journal (AMJ), 19, 230–237.

International Soccer. (2013). Soccer fans info. Retrieved from http://www.soccer-fans-info.com/in ternational-soccer.html

Jansen, B. J., Zhang, M., Sobel, K., & Chowdury, A. (2009). Twitter power: Tweets as electronic word of mouth. Journal of the American Society for Information Science and Technology, 60, 2169–2188.

Jin, S. A., & Phua, J. (2014). Following celebrities’ tweets about brands: The impact of twitter- based electronic word-of-mouth on consumers’ source credibility perception, buying intention, and social identification with celebrities. Journal of Advertising, 43, 181–195.

Kahle, L. R., & Homer, P. M. (1985). Physical attractiveness of the celebrity endorser: A social adaptation perspective. Journal of Consumer Research, 11, 954–961.

Keller, K. L. (1993). Conceptualizing, measuring, and managing customer-based brand equity. Journal of Marketing, 57(1), 1–22.

Keller, K. L. (2009). Building strong brands in a modern marketing communications environment. Journal of Marketing Communications, 15, 139–155.

Kim, Y. J., & Na, J. H. (2007). Effects of celebrity athlete endorsement on attitude towards the product: The role of credibility, attractiveness and the concept of congruence. International Journal of Sports Marketing & Sponsorship, 8, 310–320.

Koo, G. Y., Ruihley, B. J., & Dittmore, S. W. (2012). Impact of perceived on-field performance on sport celebrity source credibility. Sport Marketing Quarterly, 21, 147–158.

Kornowski, L. (2013). Celebrity sponsored tweets: What the stars get paid for advertising in 140 characters. Retrieved from http://www.huffingtonpost.com/2013/05/30/celebrity-sponsored- tweets_n_3360562.html

Lafferty, B. A., Goldsmith, R. E., & Newell, S. J. (2002). The dual credibility model: The influence of corporate and endorser credibility on attitudes and purchase intentions. Journal of Marketing Theory & Practice, 10, 1–12.

Lee, Y., & Koo, J. (2015). Athlete endorsement, attitudes, and purchase intention: The interaction effect between athlete endorser-product congruence and endorser credibility. Journal of Sport Management, 29, 523–538.

Li, C., & Bernoff, J. (2011). Groundswell: Winning in a world transformed by social technologies. Boston, MA: Harvard Business Press.

Martin, J. H. (1996). Is the athlete’s sport important when picking an athlete to endorse a nonsport product? Journal of Consumer Marketing, 13, 28–43.

McCracken, G. (1989). Who is the celebrity endorser? Cultural foundations of the endorsement process. Journal of Consumer Research, 16, 310–321.

McKelvey, S., & Masteralexis, J. T. (2013). New FTC guides impact use of social media for companies and athlete endorsers. Sport Marketing Quarterly, 22, 59.

Innovation: Management, Policy & Practice 325

Milord, J. (2015, March 10). These are the absurd amounts of money athletes get paid per tweet. Retrieved from http://elitedaily.com/sports/athletes-get-paid-to-tweet/961383/

Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers’ per- ceived expertise, trustworthiness, and attractiveness. Journal of Advertising, 19, 39–52.

Our panel of experts count down the top 50 Premier League players ahead of 2013-2014 season. (2013, August 16). FoxSports. Retrieved from http://www.foxsports.com.au/football/premier-league/our- panel-of-experts-count-down-the-top-50-premier-league-players-ahead-of-2013-2014-season/story- e6frf4a3-1226693616997

Parker, H. M., & Fink, J. S. (2012). Arrest record or openly gay: The impact of athletes’ personal lives on endorser effectiveness. Sport Marketing Quarterly, 21, 70.

Priester, J. R., & Petty, R. E. (2003). The influence of spokesperson trustworthiness on message elaboration, attitude strength, and advertising effectiveness. Journal of Consumer Psychology, 13, 408–421.

Robin van Persie (Persie_Official) on Twitter. (2013, August 27). Twitter. Retrieved from https:// twitter.com/persie_official

Romani, S. (2006). Price misleading advertising: Effects on trustworthiness toward the source of information and willingness to buy. The Journal of Product and Brand Management, 15, 130–138.

Seno, D., & Lukas, B. A. (2007). The equity effect of product endorsement by celebrities, A con- ceptual framework from a co-branding perspective. European Journal of Marketing, 41, 121– 134.

Shank, M. D. (2009). Sports marketing: A strategic perspective (4th ed.). Upper Saddle River, NJ: Pearson Prentice Hall.

Shank, M. D., & Langmeyer, L. (1994). Does personality influence brand image? The Journal of Psychology, 128, 157–164.

Spears, N., & Singh, S. N. (2004). Measuring attitude toward the brand and purchase intentions. Journal of Current Issues & Research in Advertising, 26, 53–66.

Statista.com. (2015). Twitter: number of monthly active users 2015. Retrieved from http://www.sta tista.com/statistics/282087/number-of-monthly-active-twitter-users/

Stevens, J. A., Lathrop, A. H., & Bradish, C. L. (2003). “Who is your hero?” Implications for ath- lete endorsement strategies. Sport Marketing Quarterly, 12, 103.

Till, B. D., & Busler, M. (2000). The match-up hypothesis: Physical attractiveness, expertise, and the role of fit on brand attitude, purchase intent and brand beliefs. Journal of Advertising, 29 (3), 1–13.

Till, B. D., & Shimp, T. A. (1998). Endorsers in advertising: The case of negative celebrity infor- mation. Journal of Advertising, 27, 67–82.

Till, B. D., Stanley, S. M., & Priluck, R. (2008). Classical conditioning and celebrity endorsers: An examination of belongingness and resistance to extinction. Psychology & Marketing, 25, 179–196.

Twitter Counter. (2015). The Twitaholic.com top 500 Twitterholics based on followers. Twitaholic. Retrieved from http://twitaholic.com/top500/followers

Williams, J., & Chinn, S. J. (2010). Meeting relationship-marketing goals through social media: A conceptual model for sport marketers. International Journal of Sport Communication, 3, 422– 437.

Witkemper, C., Lim, C. H., & Waldburger, A. (2012). Social media and sports marketing: Examining the motivations and constraints of Twitter users. Sport Marketing Quarterly, 21, 170–183.

Wood, N. T., & Burkhalter, J. N. (2014). Tweet this, not that: A comparison between brand promotions in microblogging environments using celebrity and company-generated tweets. Journal of Marketing Communications, 20, 129–146.

Wright, S. A. (2016). Reinvestigating the endorser by product matchup hypothesis in advertising. Journal of Advertising, 45, 26–32.

Zafar, Q., & Rafique, M. (2012). Impact of celebrity advertisement on customers’ brand percep- tion and purchase intention. Asian Journal of Business and Management Sciences, 1, 53–67.

Zeithaml, V. A. (1988). Consumer perceptions of price, quality, and value: a means-end model and synthesis of evidence. The Journal of Marketing, 52, 2–22.

326 N.T. Brison et al.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.