Research in sport has examined the use of social media by organizations and athletes, but little research has assessed the effectiveness of social media marketing, especially in terms of relationship-building. When relationship marketing is used as a framework for using social media, it should also be used to guide assessment. The purpose of this paper is to re-examine a model for measuring the impact of fans’ Facebook interactions with their favorite sport team on relationship quality, purchase intentions and referral intentions.
After conducting a survey of professional sport fans through Amazon Mechanical Turk, data were analyzed using structural equation modeling. The total sample size was 425.
Results indicated the more a fan interacts with their favorite team on Facebook, the higher their relationship quality and intentions to purchase. There was no significant effect on referral intentions. Additionally, the indirect effect of Facebook interaction on purchase intention as mediated by relationship quality was positive and significant.
Based on results, it appears that relationship marketing can be used as a framework for assessing social media marketing effectiveness in sport, and that as suggested by relationship marketing theory, social media interaction does improve relationships between fans and teams. Researchers should continue to explore this model and include other variables, such as team identification, to gain a thorough understanding of social media marketing effectiveness.
Sport marketers should focus strategy on Facebook on building relationships through content that encourages interactions between the team and fans.
While research in sport has suggested social media be used to build relationships, little research measuring whether it actually does so exists. This study extends social media research in sport by modeling and testing the relationships between social media interaction, relationship quality and consumer behavioral intentions.
Achen, R.M. (2019), "Re-examining a model for measuring Facebook interaction and relationship quality", Sport, Business and Management, Vol. 9 No. 3, pp. 255-272. https://doi.org/10.1108/SBM-10-2018-0082Download as .RIS
Emerald Publishing Limited
Copyright © 2019, Emerald Publishing Limited
New media provide innovative marketing channels for team marketers, allowing them to communicate with and create added value for customers (Buhler and Nufer, 2010). Brands across many industries now use social media networks to connect with consumers and create valuable relationships (Nisar and Whitehead, 2016). Push marketing is becoming less effective, and instead, consumers are becoming co-creators of value, who are actively involved with brands (Constantinides, 2014). For sport teams, social media can be used to enhance the brand, encourage social interaction, promote sales and enhance fans’ online experiences (Miranda et al., 2014). Because social media allow sport fans to build on their sport experience and identities and express this connection to others (Stavros et al., 2013), they are potential relationship marketing tools, and thus should be continually studied to explicate their utility as such. These channels are interactive and can facilitate conversations between customers and businesses, allowing businesses to better serve customers’ needs (Sashi, 2012). If used to understand customer needs, increase satisfaction and enhance relationships, social media channels can provide value to businesses from a marketing standpoint (Williams and Chinn, 2010; Abeza et al., 2013).
Armstrong et al. (2016) suggested sport teams could create brand community by prioritizing the relationship with the active audience and engaging fans in communication on Twitter. Based on a case study of the Los Angeles Kings twitter account, they suggested sport brands move beyond traditional marketing strategies on social media and try to build relationships with customers using these networks. Effectively encouraging interactions on social media to increase identification is one benefit social media provide (Stavros et al., 2013).
While social media provide a unique channel for marketers to reach customers, they are similar to other marketing tools in one important way; measurement is critical for success as a marketing channel (Murdough, 2009; Fisher, 2009). If sport teams are going to expend resources to market on social media networks, then they must have a deep understanding of the impacts on customers. Although researchers have determined social media managers in college sport view social media to be an effective marketing tool (Dixon et al., 2015), very little is known about whether social media marketing achieves business objectives in sport, precipitating a need for the current study. Furthermore, Abeza et al. (2015) suggested sport management researchers expand research on social media in sport by conducting more empirical analyses, which requires moving beyond describing what teams and players are doing on social media. Similarly, Filo et al. (2015) suggested social media researchers move beyond descriptive data and content analyses and use more sophisticated analytical methods. To extend social media research in sport, this study will utilize structural equation modeling (SEM) to test the impact of Facebook interaction between fans and teams on relationship quality, purchase intentions and referral intentions. Additionally, the impacts of Facebook interaction on the outcome variables among US professional sport leagues, including the National Basketball Association (NBA), National Football League (NFL), National Hockey League (NHL), Women’s National Basketball Association (WNBA), Major League Baseball (MLB) and Major League Soccer (MLS) will be explored.
The services environment, in which sport is situated, is characterized by intangibility, inseparability, variability and perishability, which makes building relationships with customers more logical than relying on traditional transaction-based marketing (Egan, 2004). Relationship marketing is defined by Morgan and Hunt (1994) as a marketing strategy that focuses on building, enhancing and maintaining relationships between customers and organizations. Serving customer needs and building strategy around those needs is central to relationship marketing strategy (Ferrand and McCarthy, 2009). At its core, relationship marketing strategy strives to build relationships between organizations and customers where both sides actively participate and benefit (Gummesson, 1999). As the two parties interact, relationships, involving mutual exchanges between both sides, grow (Ferrand and McCarthy, 2009). The main goal of relationship marketing is to retain customers because it is less expensive and more beneficial long term than acquiring new ones (Buhler and Nufer, 2010; Egan, 2004). Realized benefits of relationship marketing include improved financial performance, lower costs and higher revenues (Buhler and Nufer, 2010; Gummesson, 1999), and higher customer retention rates (Berry, 1995, Kim and Trail, 2011). In sport specifically, cultivating higher relationship quality with fans can result in greater intentions to attend a game, consume team-related media and purchase licensed merchandise (Kim, Trail and Ko, 2011).
Building connections to your brand by engaging customers is essential to using relationship marketing successfully (Drury, 2008). Marketing on social media should focus on meeting relationship marketing goals because of the two-way nature of social media, which encourages interaction and communication, fitting with Grönroos’ (2004) description of relationship marketing. He suggests relationship marketing includes interactions, two-way communications and providing added value. Social media platforms allow businesses to co-create value with customers, which could result in more meaningful relationships between them (Kao et al., 2016). Instead of simply sharing information through social media, organizations should focus on engaging customers because higher levels of engagement can lead to stronger relationships (Rishika et al., 2013).
Within the relationship marketing framework, social media provide organizations an advanced customer–organization interaction opportunity, increase their knowledge of the consumer, offer effective customer engagement, are an efficient use of resources and allow for a quicker evaluation of the customer–organization relationship by examining the intensity of involvement on these networks (Abeza et al., 2013). Companies would benefit from implementing a relationship-building focus on their social networking sites to enhance users’ participation (Pentina et al., 2013).
Organizations should use strategies on social media that nurture relationships with customers (Rishika et al., 2013). Although not a definitive indicator of relationships between fans and organizations in sport, interaction on social media can provide an indication of relationship status between fans and teams (Abeza et al., 2013). Abeza et al. (2013) cited the importance of social media for improving interaction and communication with customers in a two-way, real-time format. From the fan perspective, Thompson et al. (2017) found social media followers recognized the importance of conversations and engagement on social media that helps them connect with the sport event by building community. This indicates fans value the relationship with the team forged through social networks. The preceding literature highlights the importance of utilizing social media as relationship marketing tools.
Measuring social media marketing
One of the first steps in using social media as marketing tools is to determine the objective behind their use (Murdough, 2009). Then, an organization can determine how to measure effectiveness based on this objective. If, as Abeza et al. (2013) and Williams and Chinn (2010) have suggested, social media channels in sport can be used to build relationships and meet relationship marketing goals, then they should be measured based on their ability to do so. Thus, relationship marketing provides the theory for guiding measurement of effectiveness.
The effectiveness of social media marketing in sport has been examined in a myriad of ways. In a study of how content type of professional athletes on social media may lead to relationship-building, Watkins (2017) determined providing information fans perceived as useful was just as effective in terms of engaging fans in relationship-building activities as two-way dialogue. Respondents stated they were more likely to engage on social media if they found the information to be useful, than if they perceived its purpose to be creating conversation. Although this study provided suggestions for improving relationships using social media, it did not evaluate whether there was an actual impact on the relationship itself.
Researchers have suggested social media can be used to increase team identification, fan connection to the team and fan affiliation with the team (MacIntosh et al., 2017; Park and Dittmore, 2014). Specifically, MacIntosh et al. (2017) found the more time a fan spent reading the tweets of their team, the higher their level of affiliation with that team. However, the individual’s interaction with their favorite team did not enhance their social identity related to that team. Instead, they determined a fan’s interaction with the team’s community as a whole on social networks, which includes other fans, is what impacted their level of identification with the team. In addition to their finding that using sport teams’ social media accounts increased team identification, Park and Dittmore (2014) found the frequency with which fans used sport teams’ official social media accounts had a direct, positive effect on word-of-mouth intention, but not on attendance intention. However, when mediated by team identification, social media consumption positively impacted word-of-mouth intention and attendance intention.
In other industries, social media participation has been found to impact the frequency of shopping visits (Rishika et al., 2013), improve brand awareness, result in positive word-of-mouth, intensify purchase intentions (Hutter et al., 2013) and increase purchases (Goh et al., 2013). However, little is known about how social networking sites impact purchases in sport (Hong and Rhee, 2016). In one study, Popp et al. (2017) determined the number of Facebook page likes and Twitter followers did not increase ticket revenues or attendance, thus suggesting it might not be an effective sales tool. Conversely, other studies advocate that teams with higher revenues have more Facebook fans (Parganas et al., 2017), and teams with more Facebook fans have higher operating incomes and attendance (Achen, 2015), which provide correlational evidence that a relationship exists, but do not try to determine if consumer behavior is impacted by interacting with teams on social networks. Potentially interaction, in the form of comments made by fans on a team’s Facebook page, could have positive impacts on other consumers’ purchase intentions (Seng and Keat, 2014). Research specifically asking consumers about their social media and purchase behaviors would add to the literature exploring social media marketing’s impact on business outcomes.
Deciding how to measure the effectiveness of social media marketing is challenging; however, if relationships are built through interactions and communications, it makes sense to measure the impacts of interactions and communications on relationships. Hoffman and Fodor (2010) advocated measuring how social media affects customer relationships in the long term. Interacting with brands on social media can impact relationship quality and positive word of mouth by increasing emotional attachment (Hudson et al., 2015). Also, firm-generated content on social media networks can have a direct, positive impact on relationships in customer-firm interactions (Kumar et al., 2016). Huntley (2006) suggested that if relationship quality is increased, it then positively impacts recommendations to others and purchase behaviors.
While Risius and Beck (2015) stated, “social media activity has proven to be a viable solution for influencing relational outcomes” (p. 829), evidence of this in sport is limited to only a few studies. Watkins (2014) determined Twitter and Facebook involvement for NBA fans positively affected brand relationship and brand equity, although these effects were weak. After building a model to measure social media marketing from the literature, Achen (2016) examined how interaction on Facebook impacted relationship quality, referral intentions and purchase intentions in the NBA. Results suggested higher levels of interaction on Facebook resulted in stronger relationships and higher intentions to purchase. However, higher Facebook interaction did not have a significant effect on referral intentions. Finally, the study revealed a significant, positive indirect effect of Facebook interaction on purchase and referral intentions, as mediated by relationship quality.
While Achen’s (2016) study was important for building a model to measure marketing on Facebook, it focused solely on one league. The current study seeks to expand this research and test the model with a broader sample across multiple professional sport leagues. The expansion to multiple leagues is essential for four reasons. First, Egan (2004) suggested a one-size-fits-all relationship marketing strategy would not work because of differences across organizations. In sport, leagues differ on the demographic profiles of their fans, the business strategy, season length and number of games, which could lead to differences in the use of social media to build relationships, and results of this strategy. Second, researchers have suggested the NBA actively adopted relationship marketing (Cousens et al., 2001; Lachowetz et al., 2001), making the league and its teams potentially inherently different in their marketing strategy. Third, the NBA is known for their active presence on social media and ability to engage fans on social networks in ways different from, and more effective than, other leagues (Maese, 2018), which could lead to differences between the connection of relationship quality and interaction on social networks. Finally, there is evidence that fans may interact differently on social networks. Through content analysis and multivariate modeling, Achen et al. (2018) found fans of teams in different leagues interacted on Facebook in distinctive ways. Teams in the NFL had the most comments, MLB teams had the most shares and teams in the NBA garnered the most likes. These differences did vary by the type of content teams posted. Additionally, differences were found in the types of content being posted by teams in different leagues. For example, teams in the NFL posted a lower percentage of content promoting games or merchandise sales, while teams in the MLS and WNBA posted a greater percentage of content designed to encourage interaction and conversation with fans. If content for teams in one league is more focused on relationship-building and building conversation, then differences in impacts on relationship quality could exist.
Tangible measures of social media marketing effectiveness are important to show top managers the value of marketing on social media (Inversini and Sykes, 2013); however, evaluating the value of social media is difficult because there is a lack of a direct link to ROI (Jiang et al., 2016), and a lack of understanding how to measure these channels. This study endeavored to provide further support of the model in Figure 1, which was created and tested by Achen (2016), after an extensive literature review across multiple business industries. Thus, the following hypotheses and research questions are proposed:
Higher levels of interaction on Facebook will lead to greater relationship quality.
Higher levels of interaction on Facebook will lead to greater intentions to purchase.
Higher levels of interaction on Facebook will lead to greater intentions to refer.
Relationship quality will mediate the relationship between Facebook interaction and referral intentions.
Relationship quality will mediate the relationship between Facebook interaction and purchase intentions.
Are relationships between Facebook interaction, relationship quality, referral and purchase intentions different across the NBA, WNBA, NFL, NHL, MLS and MLB?
A cross-sectional survey research design was used to examine the model in Figure 1 across all professional sport leagues. The participants, measures, procedure and analysis are described below. Survey methodology, which is an essential method for examining consumer perceptions, attitudes and beliefs (Rea and Parker, 2012), has been used sparingly in sport social media research. This study took a survey approach to develop an understanding of interaction on social media channels and was part of a larger data collection that included multiple social media networks. Facebook was chosen as the network for analysis of interaction because it had the most users in the sample of respondents.
The population for this study was professional sport fans in the USA who use Facebook. Convenience sampling was used in this study. Participants were recruited via Amazon Mechanical Turk (MTurk), a site that allows an individual to post a task for workers and provide compensation for their participation. The study was limited to 500 respondents, and participants were required to be at least 18 years of age, sport fans and social media users to be able to start the questionnaire. Completed questionnaires were examined and quality control questions were used to determine if respondents met the criteria for payment. Of the 563 who started the questionnaire, 508 completed it. Data were downloaded and examined for patterns and validation questions were re-checked. After this second round of data cleaning and the deletion of incomplete questionnaires, 425 useable responses were collected. Of the sample, 59.5 per cent were male. Consumer characteristics are reported in Table I. Favorite league was determined by determining what league their self-reported favorite team played in.
Facebook interaction was measured using the scale previously used in a similar study by Achen (2016). The scale was created after examining literature measuring social media engagement, with a focus on behavioral engagement measures. Additionally, previous researchers have measured behavioral engagement on Facebook using Likert-type scales (Pöyry et al., 2013; Gummerus et al., 2012). Respondents were asked to rank the frequency of their Facebook interactions including likes, comments, page visits, reads and shares measured on an eight-point scale (never, a few times a year, once a month, a few times a month, once a week, a few times a week, once a day and a few times a day).
Another existing scale, created by Kim, Trail, Woo and Zhang (2011), was used to measure relationship quality. The Sport Consumer–Team Relationship Quality Scale includes five constructs; trust, commitment, intimacy, identification and reciprocity. Each construct was measured using three items. The scale has been found to be reliable with Cronbach’s α coefficients ranging from 0.82 to 0.95 and average variance extracted (AVE) values from 0.61 to 0.86 (Kim, Trail, Woo and Zhang, 2011; Kim, Trail and Ko, 2011). Additionally, the scale demonstrated discriminant and concurrent validity (Kim, Trail, Woo and Zhang, 2011; Kim, Trail and Ko, 2011).
The final two measures used in this study, purchase and referral intentions, were adapted from Pöyry et al. (2013). Each construct was measured using three items each. In their study, because all factor loadings were greater than the 0.60 standard used by Fornell and Larcker (1981), the scale demonstrated convergent validity. Also, AVE values were greater than 0.50, and composite validity measures were greater than 0.70, which demonstrated internal consistency.
The questionnaire was created using measures from previous studies and coded into Qualtrics. Then, an anonymous link was posted to MTurk and participants were invited to complete the questionnaire, which began with the qualification questions. Once 500 questionnaires were completed, the researcher reviewed responses to be sure individuals passed the validation questions, completed the questionnaire and had no initial noticeable patterns in their responses. All questionnaires that met the criteria were approved for a payment of $0.87, then the remaining spots open from the questionnaires that were not approved were posted until 500 questionnaires meeting baseline criteria were received.
Once questionnaires were completed, the task was closed on MTurk and data were downloaded. Data were cleaned a second time by re-checking the validation questions, reviewing the responses for patterns and removing data with responses that were out of bounds. Data were then uploaded into SPSS Version 22 and MPlus Version 7.4 for analysis.
Descriptive statistics were calculated in SPSS Version 22. Then, MPlus Version 7.4 was used to conduct confirmatory factor analysis and SEM. Reliability analysis was conducted using McDonald’s omega and AVE for each scale. Once reliability was assessed, the two-step modeling process advocated by Kline (2011) was used, where the measurement model was run initially to examine fit, followed by the structural models. First, the measurement model was run using robust-maximum likelihood estimation. Then, the structural models were run to test the relationships between constructs. The first structural model run was the partial mediation model, which included all direct and indirect effects in the model. Each path in the model was deleted and likelihood ratio tests using the scaling correction factor were run to determine the best-fitting, most-parsimonious model. The differences across leagues were tested in SPSS Version 22 using linear regression.
The indicator means and standard deviations are listed in Table II. The indicators for Facebook interaction were measured on a 1 (never) to 8 (a few times a day) scale and all other indicators were measured on a 1 (strongly disagree) to 7 (strongly agree) Likert-type scale.
Evaluation of assumptions
To assess normality prior to analysis, Q-Q plots and kurtosis and skew statistics for each indicator were examined. Because some skew and kurtosis values were greater than the accepted ± 2, it was determined that the data could not be treated as multivariate normal. To adjust for non-normality, the robust-maximum likelihood estimator (MLR) with a scaling correction factor was used. Cases with missing data were deleted listwise.
Reliability was assessed using McDonald’s (1999) ω coefficient, which is reported in Table III. ω coefficient values greater than 0.80 demonstrated good reliability of scales (Kline, 2011). Convergent validity was supported by an AVE of all constructs greater than 0.50 (reported in Table III) (Fornell and Larcker, 1981). Convergent validity was further supported by significant factor loadings (reported in Table IV) (Hair et al., 2005). However, because of the high correlation between referral intentions and relationship quality, discriminant validity was not supported. Thus, referral intentions were removed from the model. All other constructs passed discriminant validity with their square root of the AVE for each construct being greater than the correlation between two constructs.
The measurement model including Facebook interaction with five indicators, relationship quality with five individual constructs each with three indicators and purchase intentions with three indicators was run first and model fit was adequate, = 795.28, scaling correction factor = 1.2375, p <0.001; RMSEA(0.072, 0.084) = 0.08; CFI = 0.90; TLI = 0.89; SRMR = 0.09; however, examination of modification indices and consideration of model statements led to model re-specification. Based on modification indices, review of the statements, and a review of literature, correlated residuals between comment and share indicators and between purchase intentions 1 and purchase intentions 2 were added and the model was re-run. Model fit improved and was deemed acceptable, = 589.36, scaling correction factor = 1.2385, p <0.001; RMSEA(0.057, 0.069) = 0.06; CFI = 0.94; TLI = 0.93; SRMR = 0.07. Thus, this model was used to test the structural relationships. Factor loadings are listed in Table IV and factor variances and covariances are listed in Table V.
Next, the partial mediation structural model testing all direct and indirect effects was run and model fit was acceptable, = 589.36, scaling correction factor = 1.2385, p <0.001; RMSEA(0.057, 0.069) = 0.06; CFI = 0.94; TLI = 0.93; SRMR = 0.07. Next, to test whether there was partial or full mediation, the path from Facebook interaction to purchase intentions was removed from the model, resulting in a degradation of model fit, = 595.559, scaling correction factor = 1.2380, p <0.001; RMSEA(0.057, 0.069) = 0.063; CFI = 0.93; TLI = 0.92; SRMR = 0.07. The likelihood ratio test using the scaling correction factor suggested the change in model fit was significant, Δ = 6.55, p = 0.01, thus the partial mediation model was retained. Table VI reports the direct and indirect effects, standardized values, standard errors and p-values for the partial mediation model. The retained model is shown in Figure 2:
Higher levels of interaction on Facebook will lead to greater relationship quality.
H1 was supported by the significant positive direct path from Facebook interaction to relationship quality. One standard deviation increase in Facebook interaction led to a 0.42 standard deviation increase in relationship quality:
Higher levels of interaction on Facebook will lead to greater intentions to purchase.
H2 was supported by the positive, significant relationship between purchase intentions and Facebook interaction. Intentions to purchase increase by 0.14 standard deviations for every standard deviation increase in Facebook interaction:
Higher levels of interaction on Facebook will lead to greater intentions to refer.
Relationship quality will mediate the relationship between Facebook interaction and referral intentions.
Relationship quality will mediate the relationship between Facebook interaction and purchase intentions.
A positive significant indirect path from Facebook interaction to purchase intentions as mediated by relationship quality illustrated support for H5. The indirect effect was 0.35, indicating that an increase in Facebook interaction increased relationship quality, which in turn increased purchase intentions:
Are relationships between Facebook interaction, relationship quality, referral and purchase intentions different across leagues?
Originally, group differences were going to be explored using multiple-group SEM; however, small group sizes prevented this. To provide initial insights into how league might impact Facebook interactions to influence the three outcome variables, linear regressions were conducted for each outcome variable. Tables VII–IX report the results of regression analyses. In the model for relationship quality, the R2 (0.172) was significant; however, the only significant coefficient was Facebook interaction. When the interaction term between Facebook interaction and league was added to the model, it resulted in a non-significant change in R2 (0.017, F = 1.769, p = 0.18). The model for purchase intentions also resulted in a significant R2 (0.153), although none of the league coefficients were significant at the p = 0.01 level. Adding the interaction variable of Facebook interaction and league to the model did not explain a significant additional amount of variance (0.004, F = 0.403, p = 0.847). Finally, the R2 (0.100) for the referral intentions model was significant; however, none of the league coefficients were significant. When the interaction term between Facebook interaction and league was entered into the model, it did not explain a significant amount of additional variance (0.011, F = 1.021, p = 0.405). This information suggested there was not a significant difference in the outcome variables based on league.
Researchers have suggested social media be used to build relationships between customers and organizations by approaching strategy from a relationship marketing perspective (Abeza et al., 2013; Williams and Chinn, 2010). Thus, this study endeavored to explore whether interaction on Facebook facilitated relationships between sport teams and fans across multiple professional sport leagues. Results indicate that interaction and communication on Facebook did result in higher levels of relationship quality in professional sport fans. Although conceptual research in sport has suggested social media are relationship marketing tools, this study provides empirical evidence that social media do build relationships with fans. Also, from a theoretical standpoint, increased relationship quality further supports using relationship marketing as a guide for social media marketing strategy and measurement, because according to Grönroos (2004), interactions and communications are two of the three essential pieces of the relationship marketing process. However, it should be noted that the relationship between interaction and relationship quality could be bi-directional, or it could be that stronger relationships lead to more interaction, which should be examined further.
From a practical perspective, a positive impact of Facebook interaction on relationship quality, similar to the finding of Achen (2016) in the NBA and Rishika et al. (2013) in retail, should lead social media managers in sport to optimize their content for interaction. Doing so involves integrating the final piece of the relationship marketing process advocated by Grönroos (2004), which is providing added value for customers. Providing added value requires teams to understand what their customers want on social media, which is something some researchers have explored. For example, research from Achen (2015) suggests that content providing behind-the-scenes information and player interest stories garners more interaction. The recommendation to humanize the team, or brand, is similar to that of Thompson et al. (2017) who suggested a humanistic approach on social media could result in stronger consumer–brand relationships. Additionally, Watkins (2017) suggested useful information increased interaction. Future research in sport should explore why fans are interacting and what added value they feel they are getting from following teams on social media by directly asking them. An exploration of added value could help teams improve their social media strategy to reach more fans and result in a greater and broader impact on relationships, especially because the most effective way to encourage fans to engage is unclear (Thompson et al., 2017).
Another finding similar to that of Achen (2016) in the NBA is the direct positive impact of Facebook interaction on purchase intentions. In other industries, interaction on Facebook led to increased visits to a store (Rishika et al., 2013) and increased purchases (Goh et al., 2013). The impact on purchase intentions provides evidence that marketing on Facebook is potentially effective for selling tickets, giving teams a direct financial justification for putting time and resources into managing Facebook, even though Popp et al. (2017) suggested this was not an effective strategy in college sports. To capitalize on these findings, teams should provide training or seek out experts to train their social media marketing staff how to post content that encourages likes, comments and shares from fans, while still conveying effective sales messages.
Although Maese (2018) suggested NBA teams were more effective at engaging fans on social networks than teams in other leagues, regression analyses did not indicate that league impacted the relationships between interaction on Facebook and relationship quality, purchase intentions or referral intentions. The lack of differences is surprising when leagues vary on the demographic profiles of their fans, business strategy, season length and number of games. This may indicate that fans across leagues may be more similar than different when it comes to their desire to connect to their favorite teams online. Additionally, Achen et al. (2018) determined content strategies, and the way fans of teams in different leagues interacted with content, were different across leagues, which would seem to lead to differences in outcomes of social media based on the relationship-orientation of content. However, the lack of differences may signify that the type of interaction and the relationship-orientation of the content are not important for building relationships with fans or encouraging behavioral intentions. Instead, it is important that fans interact at any level (passive or active), and they are encouraged to do so more often. Thus, instead of focusing on the type of content created, marketers should explore what encourages fans to interact. For example, it might not be imperative that content is designed to build relationships with customers, such as content asking fans to participate in polls, but instead that content leads individuals to like, comment on, or share it, even if it is simply posting the results of the most recent game. This finding provides more evidence that the most essential part of creating content may be its perceived added value to customers. Additionally, it may be that factors impacting interaction with content are related to personality, identification or consumer motivations. However, because this exploration of differences across leagues was exploratory, more research should be conducted. If the lack of differences holds across studies, it might allow social media practitioners in sport to create a universal best practice list for encouraging relationship-building on social media. Also, it would allow social media managers to easily transition across teams in all leagues with a more gradual learning curve in a new organization.
Building and strengthening relationships with customers is an important end goal of social media marketing in itself because, as relationship marketing research suggests, stronger relationships lead to improved financial performance, reduced costs (Gummesson, 1999; Buhler and Nufer, 2010), higher retention rates (Berry, 1995; Kim and Trail, 2011; Bush et al., 2007), lower price sensitivity and increased brand loyalty (Williams and Chinn, 2010). A broader theoretical implication supports the foundation of relationship marketing as a guiding framework for social media marketing, in that purposeful building relationships with consumers through interaction on social media actually resulted in increased quality of relationships.
It is important to note that, from an integrated marketing communications perspective, it is necessary to use social media marketing within the greater context of marketing communications in organizations (Kumar et al., 2016). Thus, while marketing on social media networks should not replace other channels, it can supplement an already strong marketing communications plan. Interactions are important for building and solidifying long-term relationships, and social media allow organizations to co-create value (Kao et al., 2016), making social media an essential channel for the expenditure of resources in sport marketing.
Limitations and future research
Although the sample size of this study was respectable, the number of individuals in some leagues was limited. Future studies should use quota sampling, and potentially multiple sampling methods, to reach smaller customer bases, including WNBA and MLS fans. Then, the use of social media and its impact on relationships can be more effectively modeled across leagues. Because of the fervor for soccer in countries worldwide, this study should also be replicated with an international fan base. Another limitation of this research is its focus on Facebook, and future research should examine interaction on other networks, such as Twitter, especially as other researchers have suggested Twitter could be used to build relationships with fans and build connections with new fans (Yoon et al., 2017). Although the results here for Facebook provide insight into how interaction on social media networks in general could encourage stronger relationships, each network has different purposes, capabilities and user bases, which makes it is essential to first explore networks individually. Then comparisons between different social media networks should be conducted. Comparative analysis might provide insight into effective social media strategy and how it is similar or different across platforms.
Additional variables may also be impacting the results; thus, another limitation of this study is the inability to account for all variables. For example, team identification could account for the increased relationship quality. Moreover, while there is evidence that interaction on Facebook increases relationship quality, the relationship could also go in the opposite direction. In fact, it is likely that it is bi-directional. Researchers should consider ways to understand this relationship by partnering with teams to examine fans’ relationship quality prior to social media campaigns and after. Although research in social media on sport has examined how teams and athletes are utilizing social media networks, there are a plethora of research opportunities in consumer behavior and measurement in regards social media in sport.
Finally, social media changes and evolves very quickly, leading to concerns over the long-term usefulness of study results. However, this study is unique in that it does not focus on specific content or strategies used by teams and instead focuses on fans’ attitudes, behaviors and behavioral intentions. While the types of interaction and uses of networks might change, it is likely that fans will still utilize these networks and engage with teams, even if the modes for doing so change. Additionally, while the focus was on Facebook, a network many believe will be obsolete, it is still the most widely used network. In addition, Facebook has continued to purchase competitors, such as Instagram, which likely means both networks will remain relevant for different reasons, precipitating the use of both networks widely.
In conclusion, social media networks can be used to encourage interaction with sport fans, thus improving relationship quality. Measuring social media by their ability to enhance and maintain relationships connects return on investment with the purported objective of using social media as relationship-marketing tools. Additionally, the results of this study advance theory by providing evidence that social media do function as relationship marketing tools, which previous conceptual research has advised. Marketers in sport should continue to design content meant to encourage interactions while working with fans of their teams to determine what content adds value to the fan-team relationship.
Consumer characteristics of the sample
|Season ticket holders||12.7%|
|Average games attended||2.43|
|Attended no games||40.5%|
|Average pieces of team-licensed merchandise purchased||3|
|Purchased no team-licensed merchandise||15.1%|
|Liked Facebook page of favorite team||74.6%|
Note: n= 425
Indicator means and standard deviations
|I intend to attend the team’s games during the 2014–2015 season (Purchase 1)||4.82||1.84|
|It is likely that I purchase tickets to the team’s games during the 2014–2015 season (Purchase 2)||4.74||1.86|
|I intend to purchase the team’s team-licensed merchandise in the next year (Purchase 3)||5.27||1.48|
|I intend to say positive things about the team to other people (Referral 1)||5.79||1.17|
|I plan to recommend the team to other people (Referral 2)||5.40||1.41|
|I will encourage my friends to purchase tickets or attend games (Referral 3)||5.02||1.59|
|I trust this team (Trust 1)||5.35||1.32|
|This team is reliable (Trust 2)||5.16||1.46|
|I can count on this team (Trust 3)||5.12||1.43|
|I am committed to this team (Commit 1)||5.82||1.15|
|I am devoted to this team (Commit 2)||5.67||1.23|
|I am dedicated to this team (Commit 3)||5.74||1.21|
|I am very familiar with this team (Intimacy 1)||5.84||1.10|
|I know a lot about this team (Intimacy 2)||5.65||1.23|
|I feel as though I really understand this team (Intimacy 3)||5.33||1.28|
|This team reminds me of who I am (Identify 1)||4.66||1.66|
|This team’s image and my self-image are similar in a lot of ways (Identify 2)||4.57||1.55|
|This team and I have a lot in common (Identify 3)||4.72||1.49|
|This team unfailingly pays me back when I do something extra for it (Reciprocity 1)||3.88||1.64|
|This team gives me back equivalently what I have given them (Reciprocity 2)||4.46||1.56|
|This team constantly returns the favor when I do something good for it (Reciprocity 3)||4.22||1.64|
|How often do you visit your favorite team’s Facebook page (Visit)||3.05||1.60|
|How often do you read content posted by the team on Facebook (Read)||2.43||1.67|
|How often do you share content posted by the team on Facebook (Share)||2.68||1.60|
|How often do you like content posted by the team on Facebook (Like)||3.15||1.64|
|How often do you comment on content posted by the team on Facebook (Comment)||2.49||1.63|
Reliability coefficients of the latent constructs
Factor loadings and residuals for the measurement model
|Factor loadings||Error variances|
|Std. est.||SE||p||Std. est.||SE||p|
|Facebook relationship strength|
Note: Std. est., standardized Estimate
Factor and residual variances and covariances for the measurement model
|Relationship quality with Facebook interaction||0.42||0.05||<0.001|
|Relationship quality with purchase intentions||0.90||0.04||<0.001|
|Facebook interaction with purchase intentions||0.50||0.05||<0.001|
|Comment with share||0.57||0.05||<0.001|
|Purchase 1 with purchase 2||0.81||0.03||<0.001|
Note: Std. est., standardized estimate
Estimates of direct and indirect effects for the retained partial mediation model
|Facebook interaction → relationship quality||0.24||0.42||0.05||<0.001|
|Facebook interaction → purchase intentions||0.11||0.14||0.05||0.003|
|Relationship quality → purchase intentions||1.12||0.84||0.04||<0.001|
|Facebook interaction → relationship quality→ purchase intentions||0.27||0.35||0.04||<0.001|
Note: Std. est., standardized estimate
Regression results for the effects of league and Facebook interaction on relationship quality
Notes: R2 = 0.172, adjusted R2 = 0.172, SE = 13.58, F = 14.46, df = 2,667.84, p<0.001
Regression results for the effects of league and Facebook interaction on purchase intentions
Notes: R2=0.153, adjusted R2=0.141, SE=4.15, F=12.62, df=217.448, p <0.001
Regression results for the effects of league and Facebook interaction on referral intentions
Notes: R2=0.100, adjusted R2=0.087, SE=3.30, F=7.70, df=83.833, p<0.001
Abeza, G., O’Reilly, N. and Reid, I. (2013), “Relationship marketing and social media in sport”, International Journal of Sport Communication, Vol. 6 No. 2, pp. 120-142.
Abeza, G., O’Reilly, N., Seguin, B. and Nzindukiyimana, O. (2015), “Social media scholarship in sport management research: a critical review”, Journal of Sport Management, Vol. 29 No. 6, pp. 601-618.
Achen, R.M. (2015), “Likes, comments, and shares: a multivariate multilevel analysis of Facebook engagement”, Global Sport Business Journal, Vol. 3 No. 3, pp. 1-16.
Achen, R.M. (2016), “The influence of Facebook engagement on relationship quality and consumer behavior in the National Basketball Association”, Journal of Relationship Marketing, Vol. 15 No. 4, pp. 247-268.
Achen, R.M., Kaczorowski, J., Horsmann, T. and Ketzler, A. (2018), “Exploring off-season content and interaction on Facebook: a comparison of US professional sport leagues”, International Journal of Sport Communication, Vol. 11 No. 3, pp. 389-413.
Armstrong, C.G., Delia, E.B. and Giardina, M.D. (2016), “Embracing the social in social media: an analysis of the social media marketing strategies of the Los Angeles Kings”, Communication & Sport, Vol. 4 No. 2, pp. 145-165.
Berry, L.L. (1995), “Relationship marketing of services – growing interest, emerging perspectives”, Journal of the Academy of Marketing Science, Vol. 23 No. 4, pp. 236-245.
Buhler, A. and Nufer, G. (2010), Relationship Marketing in Sports, Butterworth-Heinemann, Burlington, MA.
Bush, R.P., Underwood, J.H. III and Sherrell, D.L. (2007), “Examining the relationship marketing, marketing productivity paradigm: establishing an agenda for current and future research”, Journal of Relationship Marketing, Vol. 6 No. 2, pp. 9-32.
Constantinides, E. (2014), “Foundations of social media marketing”, Procedia – Social and Behavioral Sciences, Vol. 148, pp. 40-57.
Cousens, L., Babiak, K. and Slack, T. (2001), “Adopting a relationship marketing paradigm: the case of the National Basketball Association”, International Journal of Sport Marketing & Sponsorship, Vol. 2 No. 4, pp. 331-354.
Dixon, A.W., Martinez, J.M. and Martin, C.L. (2015), “Employing social media as a marketing strategy in college sport: an examination of perceived effectiveness in accomplishing organizational objectives”, International Review on Public and Nonprofit Marketing, Vol. 12 No. 2, pp. 97-113.
Drury, G. (2008), “Social media: should marketers engage and how can it be done effectively?”, Journal of Direct, Data and Digital Marketing Practice, Vol. 9 No. 3, pp. 274-277, doi: 10.1057/palgrave.dddmp.4350096.
Egan, J. (2004), Relationship Marketing: Exploring Relational Strategies in Marketing, Pearson Education, Boston, MA.
Ferrand, A. and McCarthy, S. (2009), Marketing the Sports Organisation: Building Networks and Relationships, Routledge, New York, NY.
Filo, K., Lock, D. and Karg, A. (2015), “Sport and social media research: a review”, Sport Management Review, Vol. 18 No. 2, pp. 166-181.
Fisher, T. (2009), “ROI in social media: a look at the arguments”, Journal of Database Marketing & Customer Strategy Management, Vol. 16 No. 3, pp. 189-195.
Fornell, C. and Larcker, D.F. (1981), “Evaluating structural equation models with unobservable variables and measurement error”, Journal of Marketing Research, Vol. 18 No. 1, pp. 39-50.
Goh, K.-Y., Heng, C.-S. and Lin, Z. (2013), “Social media brand community and consumer behavior: quantifying the relative impact of user- and marketer-generated content”, Information Systems Research, Vol. 24 No. 1, pp. 88-107.
Grönroos, C. (2004), “The relationship marketing process: communication, interaction, dialogue, value”, Journal of Business & Industrial Marketing, Vol. 19 No. 2, pp. 99-113, doi: 10.1108/08858620410523981.
Gummerus, J., Liljander, V., Weman, E. and Pihlstrom, M. (2012), “Customer engagement in a Facebook brand community”, Management Research Review, Vol. 35 No. 9, pp. 857-877.
Gummesson, E. (1999), Total relationship Marketing: Rethinking Marketing Management: From 4Ps to 30Rs, Butterworth, Boston, MA.
Hair, J.F., Tatham, R.L., Anderson, R.E. and Black, W. (2005), Multivariate Data Analysis, Pearson Prentice Hall, Upper Saddle River, NJ.
Hoffman, D.L. and Fodor, M. (2010), “Can you measure the ROI of your social media marketing”, MIT Sloan Management Review, Vol. 52 No. 1, pp. 41-49.
Hong, S.P. and Rhee, Y.-C. (2016), “Effect of SNS on purchasing intention for sport product”, Sport Journal, Vol. 20, pp. 1-15.
Hudson, S., Roth, M.S., Madden, T.J. and Hudson, R. (2015), “The effects of social media on emotions, brand relationship quality, and word of mouth: an empirical study of music festival attendees”, Tourism Management, Vol. 47, pp. 68-76.
Huntley, J.K. (2006), “Conceptualization and measurement of relationship quality: linking relationship quality to actual sales and recommendation intention”, Industrial Marketing Management, Vol. 35 No. 6, pp. 703-714.
Hutter, K., Hautz, J., Dennhardt, S. and 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, Vol. 22 Nos 5/6, pp. 342-351.
Inversini, A. and Sykes, E. (2013), “An investigation into the use of social media marketing and measuring its effectiveness in the events industry”, in Xiang, Z. and Tussyadiah, I. (Eds), Information and Communication Technologies in Tourism 2014, Springer, Champaign, IL, pp. 131-144.
Jiang, H., Luo, Y. and Kulemeka, O. (2016), “Social media engagement as an evaluation barometer: insights from communication executives”, Public Relations Review, Vol. 42 No. 4, pp. 679-691.
Kao, T.-Y., Kao, T.-Y., Yang, M.-H., Yang, M.-H., Wu, J.-T.B., Wu, J.-T.B., Cheng, Y.-Y. and Cheng, Y.-Y. (2016), “Co-creating value with consumers through social media”, Journal of Services Marketing, Vol. 30 No. 2, pp. 141-151.
Kim, Y.K. and Trail, G. (2011), “A conceptual framework for understanding relationships between sport consumers and sport organizations: a relationship quality approach”, Journal of Sport Management, Vol. 25 No. 1, pp. 57-69.
Kim, Y.K., Trail, G. and Ko, Y.J. (2011), “The influence of relationship quality on sport consumption behaviors: an empirical examination of the relationship quality framework”, Journal of Sport Management, Vol. 25 No. 6, pp. 576-592.
Kim, Y.K., Trail, G.T., Woo, B. and Zhang, J. (2011), “Sports consumer-team relationship quality: development and psychometric evaluation of a scale”, International Journal of Sports Marketing & Sponsorship, Vol. 12 No. 3, pp. 254-271.
Kline, R.B. (2011), Principles and Practice of Structural Equation Modeling, Guilford Press, New York, NY.
Kumar, A., Bezawada, R., Rishika, R., Janakiraman, R. and Kannan, P.K. (2016), “From social to sale: the effects of firm-generated content in social media on customer behavior”, Journal of Marketing, Vol. 80 No. 1, pp. 7-25.
Lachowetz, T., McDonald, M., Sutton, W. and Clark, J. (2001), “The national basketball association: application of customer lifetime value”, Sport Marketing Quarterly, Vol. 10 No. 3, pp. 181-184.
McDonald, R.P. (1999), Test Theory: A Unified Treatment, Erlbaum, Mahwah, NJ.
MacIntosh, E., Abeza, G. and Lee, J. (2017), “Enriching identity in the ‘fan nation’ the role of social media in the case of a professional sport team”, Sport, Business and Management: An International Journal, Vol. 7 No. 3, pp. 315-331.
Maese, R. (2018), “NBA Twitter: a sports bar that doesn’t close, where the stars pull up a seat next to you”, The Washington Post, available at: www.washingtonpost.com/news/sports/wp/2018/05/31/nba-twitter-a-sports-bar-that-doesnt-close-where-the-stars-pull-up-a-seat-next-to-you/?noredirect=on&utm_term=.725c7ad0a2e0
Miranda, F.J., Chamorro, A., Rubio, S. and Rodriguez, O. (2014), “Professional sports teams on social networks: a comparative study employing the Facebook assessment index”, International Journal of Sport Communication, Vol. 7 No. 1, pp. 74-89.
Morgan, R.M. and Hunt, S.D. (1994), “The commitment-trust theory of relationship marketing”, Journal of Marketing, Vol. 58 No. 3, pp. 20-38.
Murdough, C. (2009), “Social media measurement: it’s not impossible”, Journal of Interactive Advertising, Vol. 10 No. 1, pp. 94-99.
Nisar, T.M. and Whitehead, C. (2016), “Brand interactions and social media: enhancing user loyalty through social networking sites”, Computers in Human Behavior, Vol. 62, pp. 743-753.
Parganas, P., Liasko, R. and Anagnostopoulos, C. (2017), “Scoring goals in multiple fields: social media presence, on-field performance and commercial success in European professional football”, Sport, Business and Management: An International Journal, Vol. 7 No. 2, pp. 197-215.
Park, J.-A. and Dittmore, S. (2014), “The relationship among social media consumption, team identification, and behavioral intentions”, Journal of Physical Education and Sport, Vol. 14, pp. 331-336.
Pentina, I., Gammoh, B.S., Zhang, L. and Mallin, M. (2013), “Drivers and outcomes of brand relationship quality in the context of online social networks”, International Journal of Electronic Commerce, Vol. 17 No. 3, pp. 63-86.
Popp, N., McEvoy, C. and Watanabe, N. (2017), “Do college athletics marketers convert social media growth into ticket sales?”, International Journal of Sports Marketing and Sponsorship, Vol. 18 No. 2, pp. 212-227.
Pöyry, E., Parvinen, P. and Malmivaara, T. (2013), “Can we get from liking to buying? Behavioral differences in hedonic and utilitarian Facebook usage”, Electronic Commerce Research and Applications, Vol. 12 No. 4, pp. 224-235.
Rea, L.M. and Parker, R.A. (2012), Designing and Conducting Survey Research: A Comprehensive Guide, Jossey-Bass, San Francisco, CA.
Rishika, R., Kumar, A., Janakiraman, R. and Bezawada, R. (2013), “The effect of customers’ social media participation on customer visit frequency and profitability: an empirical investigation”, Information Systems Research, Vol. 24 No. 1, pp. 108-127.
Risius, M. and Beck, R. (2015), “Effectiveness of corporate social media activities in increasing relational outcomes”, Information & Management, Vol. 52 No. 7, pp. 824-839.
Sashi, C.M. (2012), “Customer engagement buyer-selling relationships, and social media”, Management Decision, Vol. 50 No. 2, pp. 253-272.
Seng, C.S. and Keat, L.H. (2014), “Marketing sports products on Facebook: the effect of social influence”, Physical Culture & Sport Studies & Research, Vol. 61 No. 1, pp. 65-73.
Stavros, C., Meng, M.D., Westberg, K. and Farrelly, F. (2013), “Understanding fan motivation for interacting on social media”, Sport Management Review, Vol. 17 No. 4, pp. 455-469.
Thompson, A.J., Martin, A.J., Gee, S. and Geurin, A.N. (2017), “Fans’ perceptions of professional tennis events’ social media presence: interaction, insight, and brand anthropomorphism”, Communication & Sport, Vol. 5 No. 5, pp. 579-603.
Watkins, B. (2014), “An integrated approach to sports branding: examining the influence of social media on brand outcomes”, International Journal of Integrated Marketing Communications, Vol. 6 No. 2, pp. 30-40.
Watkins, B.A. (2017), “Experimenting with dialogue on twitter: an examination of the influence of the dialogic principles on engagement, interaction, and attitude”, Public Relations Review, Vol. 43 No. 1, pp. 163-171.
Williams, J. and Chinn, S.J. (2010), “Meeting relationship-marketing goals through social media: a conceptual model for sport marketers”, International Journal of Sport Communication, Vol. 3 No. 4, pp. 422-437.
Yoon, S., Petrick, J.F. and Backman, S.J. (2017), “Twitter power and sport-fan loyalty: the moderating effects of Twitter”, International Journal of Sport Communication, Vol. 10 No. 2, pp. 153-177.