Antecedents and consequences of sponsor-stadium fit Empirical evidence from a non-historic stadium context in Japan

Purpose – Integrating several streams of theoretical reasoning such as social identity theory, congruity theory and the customer gratitude approach, the purpose of this paper is to develop a model of the antecedents and consequences of sponsor-stadium fit and examine the hypothesised relationships. Design/methodology/approach – Data were collected from professional football spectators in a non-historic stadium context (n1⁄4 342). Through a confirmatory factor analysis and structural equation modelling, the authors assessed the antecedents and consequences of sponsor-stadium fit. Findings – Based on the results, team identification and prior sponsor attitude were found to be the dominant factors in enhancing sponsor-stadium fit. Furthermore, the indirect effects of team identification on purchase intentions through sponsor-stadium fit and gratitude towards the sponsor were positive and significant. Research limitations/implications –When renaming non-historic stadiums of relatively new sport teams, sponsors that present a team-related brand identity can create a preference and image fit with stadiums. The findings serve to advance the literature on stadium sponsorship particularly at non-historic stadiums. Originality/value – In its conceptualisation of sponsor-stadium fit, the current study extends previous research that has focused primarily on sponsor-event fit.

The number of stadium-naming-rights deals has increased significantly since the mid-1990s (Crompton and Howard, 2003).For example, as of 2012, more than 70 per cent of professional sport teams in the four major leagues of the USA and Canada (i.e.Major League Baseball, the National Basketball Association, the National Football League and the National Hockey League) had played in corporate-named facilities (Howard and Crompton, 2014).Stadium-naming-rights sponsorship has since local residents through community development by promoting sustainable sport culture and increase the awareness of local sports in the society at large.Therefore, the current context is an excellent illustration of stadium sponsorship of a relatively new local sport team.On the other hand, renaming historic stadiums entails risks, particularly regarding the reactions of devoted sport fans, and has the potential to damage distinctive characteristics such as history and tradition (Reysen et al., 2012).To avoid these factors, the focus of the current study is on fan reaction to the corporate renaming of a non-historic stadium.This study was conducted in the 2015 season.More details on the respondents are presented in the Method section.

Conceptual background and hypotheses
Defining sponsor fit It is widely acknowledged that fit plays a key role in sponsorship effectiveness (Becker-Olsen and Hill, 2006;Cornwell et al., 2005;Speed and Thompson, 2000).Fit (also called congruence or similarity) is defined as a strategic match between sponsoring firms and sponsored objects in business, mission, image and/or value (Becker-Olsen and Hill, 2006;Zdravkovic et al., 2010).To date, sponsor-event fit is the most widely studied concept (e.g.Grohs and Reisinger, 2014;Lacey and Close, 2013;Mazodier and Merunka, 2012;Olson, 2010;Speed and Thompson, 2000) and refers to consumers' perceptions of "the degree to which sponsors and the event match, belong to the same world, or seem likely to engage in joint business or communication efforts" (Mazodier and Merunka, 2012, p. 808).
Viewed broadly, fit is a strategic match between sponsoring firms and sponsored objects (Becker-Olsen and Hill, 2006).This definition indicates that the concept of fit is Stakeholders/ attributes Description The J. League The J. League is a professional football league established in 1993.The first season took place with ten clubs.In the five-year period between 1994 and 1998, eight clubs were added to the league.In 1999, the league switched to a two-division format including 16 Division I clubs and 10 Division II clubs and started using a system of promotion and relegation between the two divisions.In the last decade, the league has continued to expand both in the number of clubs and also in the number of divisions.For the 2015 season, the league had three divisions and 51 clubs (18 Division I clubs, 22 Division II clubs, and 11 Division III clubs) The Football Club The football club was established in 1994 and joined the J. League as a Division II club in 2000.In the year studied (2015 season), the club finished 19th in Division II.
In the 2015 season, the club's average attendance was 4,816 The Stadium The stadium is a multi-purpose facility located in the club's hometown.The stadium can be used not only for football, but also for rugby and track and field.
While the stadium was built in 1987, it was renovated in 2009 expanding the seating capacity from 5,000 to 12,000.For this renovation, the city spent 3.4 billion yen (approximately 28 million US dollars at an exchange rate of 1 US dollar ¼ 120 yen).The stadium has been owned by the city The Sponsor The naming-rights sponsor of the stadium is one of the largest consumer electronics retail chains in Japan.Its national headquarters is based in the hometown of the football club.When the stadium was renovated in 2009, the electronics company obtained the naming-rights to the stadium and attached its' brand name to the beginning of the stadium name The Mission The primary missions of the naming-rights sponsorship are to serve local residents through community development by promoting sustainable sport culture and increase the awareness of local sports in the society at large

409
Sponsorstadium fit not solely based on sponsor-event fit, but instead develops in reference to different sponsored objects, such as teams, players, leagues, sports and stadiums.Following Becker-Olsen and Hill's (2006) general definition of fit, the authors define sponsorstadium fit as consumers' perceptions of the degree to which a sponsor and the sponsored stadium create a strategic match in image, mission and value.

Hypotheses
Figure 1 shows the theoretical framework and research hypotheses underlying this study.The authors expect that team identification will directly influence sponsorstadium fit, feelings of gratitude towards the sponsor, and purchase intentions through the sponsorship.Also, team identification is expected to have indirect effects on purchase intentions through sponsor-stadium fit and feelings of gratitude towards the sponsor.The proposed framework will extend the existing literature by incorporating sponsor-stadium fit into a traditional model of sport sponsorship (e.g.Gwinner and Bennett, 2008;Speed and Thompson, 2000).In the following section, the authors develop hypotheses within this framework.
The impact of team identification.Team identification refers to a consumer's perceived oneness with a sport team and the tendency to experience the team's successes and failures as one's own (Gwinner and Swanson, 2003;Mael and Ashforth, 1992).Insights derived from social identity theory (SIT) (Tajfel andTurner, 1979, 1985) constitute the theoretical foundation for the hypothesised effects of team identification.SIT argues that a social group can be conceptualised as "a collection of individuals who perceive themselves to be members of the same social category, share some emotional involvement in this common definition of themselves, and achieve some degree of social consensus about the evaluation of their group" (Tajfel and Turner, 1979, p. 40).An individual's identification with a sport team creates a team identity that shapes the person's self-image deriving from the social category with which he or she identifies as a fan of the team (Tajfel and Turner, 1985).Sport psychologists have noted that team identification has a positive influence on a variety of cognitive outcomes such as team knowledge, self-esteem and in-group favouritism (Wann and Branscombe, 1990;

Note:
The most notable addition to the literature involves a conceptualization of the shaded construct Wann and Grieve, 2005).In the sponsorship context, previous research has shown greater team identification leads to higher perceptions of sponsor fit, more positive attitudes towards the sponsor and higher purchase intentions (Gwinner and Bennett, 2008;Gwinner and Swanson, 2003).The underlying rationale for these relationships is that as a consumer's identification with a sport team increases, greater involvement with the sponsor occurs, promoting the assimilation of the sponsor's image into the team's image and resulting in higher levels of sponsor-fit (Gwinner and Bennett, 2008;Gwinner and Swanson, 2003).According to SIT, when a consumer strongly identifies with his or her favourite sport team, he or she tends to have more favourable impressions for other in-group members who support the team.This biased evaluation is referred to as an ingroup favouritism effect (Ashforth and Mael, 1989;Tajfel and Turner, 1985) and also occurs when highly identified fans recognise sponsors as in-group members who support their favourite sport objects (Gwinner and Bennett, 2008).
In the context of historic stadiums (e.g.Fenway Park), research shows that corporate renaming of sport venues will be a threat to devoted fans because their attitudes towards traditional facilities are persistent and resistant to change (Reysen et al., 2012).On the other hand, the authors expect an in-group favouritism effect to arise when consumers find an image fit between a sponsor and a non-historic stadium.When consumers encounter an unfamiliar stadium name, important information about the stadium is missing.The literature on consumer inference making suggests consumers evaluate stadium sponsorship by drawing a connection between available sponsor characteristics (e.g. the sponsor's reputation in the local community) and the missing information (e.g. the reason why the sponsor supports the sport team; Brown and Dacin, 1997;Dick et al., 1990).When a reputable company becomes the naming-right sponsor of a non-historic stadium, consumers have a favourable impression of the stadium sponsorship by determining its appropriateness for the suitable sponsor.Thus, the authors propose that one way in which in-group favouritism influences sponsor-stadium fit is through renaming a non-traditional sport stadium.
Additionally, the authors attempt to extend the literature by linking team identification to a consumer's gratitude towards the sponsor, which has not been well studied in the sponsorship literature.A notable exception is Kim et al. (2010), who find consumers who appreciate the sponsor's support are more likely to purchase the sponsor's products.Drawing from SIT, the authors posit that consumers who are high in team identification will exhibit higher levels of involvement with the sponsor, form a reciprocal relationship with the sponsor, and return a favour to the sponsor (e.g.feelings of appreciation and gratefulness).In this study, a consumer's gratitude towards the sponsor is defined as feeling grateful, thankful, or appreciative towards the sponsor on the basis of the benefit the consumer receives.Collectively, based on the discussion above, the authors propose the following hypotheses: H1.Team identification will have a positive effect on sponsor-stadium fit.
H2. Team identification will have a positive effect on feelings of gratitude towards the sponsor.
H3. Team identification will have a positive effect on purchase intentions through the sponsorship.
The impact of sponsor fit.Congruity theory (Cacioppo et al., 1994;Jagre et al., 2001), a type of cognitive consistency theory, is particularly relevant to the sponsor-fit

411
Sponsorstadium fit approach.Two of its central tenets are that harmony among thoughts, feelings and behaviours is the most pleasant, desirable and stable state of equilibrium for people and individuals are motivated to maintain consistency among elements of information in order to avoid feeling unpleasant tension when imbalance exists (Cacioppo et al., 1994;Jagre et al., 2001).Applying congruity theory to the sponsorship context implies that consumers value consistency between the image of a sponsoring firm and the image of the sponsored object.Once the image of the sponsor is consistent with the image of the sponsored object, this image consistency creates a harmonious state where consumers hold more positive feelings towards the sponsor.Within the sponsorship domain, Pracejus (2004) proposes that the greater fit between a sponsor and the sponsored object, the more likely it is that consumers will infer that the sponsored object is endorsing the quality of the sponsor.
To date, researchers have investigated these tenets in the sponsorship context and found that sponsor fit had a positive effect on attitude towards the sponsor (Gwinner and Bennett, 2008), attitude towards the sponsorship (Mazodier and Merunka, 2012;Olson, 2010), perceived sponsor sincerity (Olson, 2010), sponsor brand trust (Mazodier and Merunka, 2012), sponsor brand image (Grohs and Reisinger, 2014), commitment to the sponsor (Lacey and Close, 2013), post-event response (Lobo et al., 2014) and purchase intentions of the sponsor's products (Speed and Thompson, 2000).Moreover, Crimmins and Horn (1996) state that "[a] link between a brand and a sponsored event or organisation leads, at minimum, to gratitude" (p.17).Consumers feel gratitude towards the sponsor if they realise that there is a reasonable connection between the sponsor and the sponsored object, and if they feel they are contributing to the sponsored object by associating with the sponsor (Crimmins and Horn, 1996).In the current study, the authors propose a model suggesting that sponsor-stadium fit has a positive impact on feelings of gratitude towards the sponsor and purchase intentions through the sponsorship.Therefore, the following hypotheses are derived: H4.Sponsor-stadium fit will have a positive effect on feelings of gratitude towards the sponsor.
H5. Sponsor-stadium fit will have a positive effect on purchase intentions through the sponsorship.
The impact of consumer gratitude towards the sponsor.Gratitude is the emotional appreciation for benefits received.Researchers across many disciplines have recognised that when an individual experiences benevolence or goodwill from a giver, the feeling of gratitude arises and motivates the recipient to help the giver (Bartlett and DeSteno, 2006;McCullough et al., 2001;Palmatier et al., 2009).Empirical research indicates that the feeling of gratitude drives helping behaviour towards the person's benefactor, even if the act is costly to him or her at the moment (Bartlett and DeSteno, 2006).In marketing, Palmatier et al. (2009) argue customers increase their intentions to repay the seller by engaging in gratitude-based reciprocal relationships.
The preceding evidence suggests that, in the sponsorship context, a consumer's expression of gratitude towards the sponsor should produce greater motivation to support the sponsor in the future.Pracejus (2004) framed this as reciprocity whereby a consumer feels that the sponsor supports the sport object he or she cares about, and then he or she will patronise the sponsor.This proposition is supported by an empirical test that finds that a sport consumer's gratitude towards the event sponsor affects his or her intentions to purchase the sponsor's products (Kim et al., 2010).Consistent with 412 SBM 6,4 past research, the authors attempt to confirm the impact of gratitude on purchase intentions even if the simultaneous effects of team identification and sponsor-stadium fit are examined.Therefore, the authors propose the following hypothesis: H6. Consumers' feelings of gratitude towards the sponsor will have a positive effect on purchase intentions through the sponsorship.
Control variables.In order to assess the robustness of the main effects of the proposed constructs, the authors include two control variables: past purchase visits and prior sponsor attitude.According to the theory of planned behaviour, a consumer's past behaviour can explain his or her actual and future behaviour (Ajzen, 1991), indicating that past purchases at stores of the sponsor influence the consumer's reactions to the sponsorship.Furthermore, consumers' attitudes towards the sponsor have been found to influence their attitudes towards and purchase intentions through the sponsorship (e.g.Lobo et al., 2014;Olson, 2010;Speed and Thompson, 2000).Therefore, the authors control for past purchase visits and prior sponsor attitude.

Data collection
This study was conducted as part of a league wide annual spectator survey of the J. League.Data were collected from spectators attending a J. League Division II game in a mid-sized city in east Japan.The authors gathered data in all seating sections except for the section of the opposing team's fans.Questionnaires were distributed in the stands before the game started.In order to collect data as systematically as possible, the authors used a mixture of convenience and proportionate sampling, stratified by gender and age.Before distributing the questionnaires, 21 trained surveyors observed an assigned block of the stands in order to estimate the percentage of those attending based on gender (1 ¼ male, 2 ¼ female) and age [1] (1 ¼ ages between 18 and 29, 2 ¼ ages between 30 and 49, 3 ¼ ages of 50 and above).Each surveyor was responsible for distributing 20 self-administered questionnaires according to the estimated percentages based on gender and age.Of the 417 questionnaires distributed, 414 were returned, yielding a response rate of 99.3 per cent.Among the 414 forms returned, 55 were rejected because many items were left blank.Furthermore, the authors eliminated seventeen respondents who were not aware of the electronics company (the stadium-naming-rights sponsor), yielding a final usable response rate of 82.0 per cent (n ¼ 342).Of the total sample, 61.4 per cent of the subjects were male.
The average age of the respondents was 40.28 years old.Age was further classified into five categories.Approximately one-third of the subjects were in the 40-49 age range (34.0 per cent), 23.9 per cent were between 30 and 39 years old, 21.2 per cent were between 18 and 29 years old, 11.3 per cent were between 50 and 59 years old and 9.6 per cent were 60 years old and above.
The representativeness of the study sample to the population was verified by comparing its demographic information with that of the 2014 league-wide annual survey (League, 2014).According to the survey report which was based on the data collected from 17,234 game attendees of clubs of Divisions I and II, the gender distribution of the entire league (male ¼ 61.5 per cent, female ¼ 38.5 per cent) corresponded to that of the current sample (male ¼ 61.4 per cent, female ¼ 38.6 per cent).Therefore, the sample of this study was considered to be an adequate representation of the overall population to generate data for this research.

Measurement
The items used to measure the proposed constructs were adapted from previous research (see Table II).Team identification was measured with a three-item scale capturing the cognitive aspects of sport fandom, perceived oneness and self-definition (Trail and James, 2001).In order to measure sponsor-stadium fit, a three-item scale was adapted from Speed and Thompson's (2000) sponsor fit scale which measured the elements of logical connection, similarity and image fit.The wording of the scale was modified to reflect consumers' perceived fit with the stadium.Feelings of gratitude towards the sponsor was measured based upon Palmatier et al.'s (2009) customer gratitude scale, which captured a consumer's feelings of thankfulness, appreciation and gratefulness.Purchase intentions through the sponsorship were measured with a three-item scale adapted from Speed and Thompson (2000).These items measured consumers' future intentions to make visits and purchases based on the electronics company's sponsorship.
Finally, the authors included two control variables that might influence the proposed endogenous constructs: past purchase visits and prior sponsor attitude.Past purchase visits were measured by the self-reported number of visits to the Construct Item λ CR AVE Team identification (Trail and James, 2001) 0.94 0.83 1.
I consider myself to be a "real" fan of (team name) 0.84 2.
I would experience a loss if I had to stop being a fan of (team name) 0.92 3.
Being a fan of (team name) is very important to me 0.97 Sponsor-stadium fit (Speed and Thompson, 2000) 0.91 0.77 1.
There is a logical connection between (team name)'s home stadium and (sponsor name) 0.82 2.
The image of (team name)'s home stadium and the image of (sponsor name) are similar 0.88 3.
I am thankful for (sponsor name) 0.89 2.
I am grateful for (sponsor name) 0.97 Purchase intentions through the sponsorship (Speed and Thompson, 2000) 0.97 0.90 1.
(Sponsor name)'s sponsorship would make me more likely to visit (sponsor name)'s stores 0.93 2.
(Sponsor name)'s sponsorship would make me more likely to consider (sponsor name)'s products the next time I buy 0.98 3.
I would be more likely to buy products from (sponsor name) as a result of (sponsor name)'s sponsorship 0.95 Prior sponsor attitude (Malär et al., 2011) 0.87 0.69 1.
I have a favourable opinion of (sponsor name) 0.81 2.

414
SBM 6,4 sponsor's stores over the last six months.Prior sponsor attitude was measured with a three-item scale adapted from Malär et al.'s (2011) scale of prior brand attitude.In order to control for possible item-order effects, the items for the two control variables were administered before all items of team identification, sponsor fit, gratitude and purchase intentions.

Back-translation
For this study, the survey items were developed in English.As a check of meaning equivalence between the original English instrument and the translated Japanese instrument, the survey questionnaire was first translated into Japanese by one of the authors and then back-translated into English by another native of Japan who is also fluent in English.To ensure the accuracy of the translation, a US-born American citizen was asked to assess differences in meaning between the original and back-translated instruments.The comparison of the two forms indicated that both instruments reflected the construct domain.

Measurement check
In order to assess the construct validity of the latent constructs, the authors conducted a confirmatory factor analysis (CFA) using Mplus version 7.31.Scale statistics, including factor loadings (λ), composite reliability (CR) and average variance extracted (AVE) values, are presented in Table II.Factor loadings ranged from 0.77 to 0.98.The CR values for the five constructs were greater than the recommended cut-off point of 0.60 (Bagozzi and Yi, 1988), indicating the proposed constructs were internally consistent.A further assessment of convergent and discriminant validity was conducted by examining AVE values.The computed AVE values for the five latent constructs ranged from 0.69 to 0.90, providing evidence of convergent validity (Fornell and Larcker, 1981).Discriminant validity was assessed by comparing the AVE estimate for each construct with the squared correlations between the constructs (see Table III).Of a total of ten correlations among the five constructs, the AVE values were greater than any squared correlations between all pairs of the constructs.Therefore, the authors found evidence of discriminant validity.
Table II also presents the results of the global fit indices for the measurement model.The ratio of χ 2 /df was smaller than Hu and Bentler's (1999)

415
Sponsorstadium fit the comparative fit index (CFI) and Trucker Lewis index (TLI) were greater than the cut-off point of 0.90 (Hu and Bentler, 1999).The root mean square error of approximation (RMSEA) and standardised root mean square residual (SRMR) were smaller than the required cut-off point of 0.08 (Hu and Bentler, 1999).The overall assessment of the fit indices indicated the measurement model was an acceptable fit to the data.

Hypothesis testing
The hypothesised relationships were examined by structural equation modelling using Mplus version 7.31.To test the direct and indirect effects of the exogenous variables on the endogenous variables, the authors used the bootstrapping method recommended by Preacher and Hayes (2008).As shown in Table IV estimation using 5,000 resamples confirmed that the direct effects of team identification on sponsor-stadium fit, gratitude towards the sponsor, and purchase intentions were positive and significant.H1-H3 were thus supported.In addition, the direct effects of sponsor-stadium fit on gratitude towards the sponsor and purchase intentions were positive and significant.Therefore, the authors found support for H4 and H5.Furthermore, the path identified between gratitude towards the sponsor and purchase intentions was positive and significant, in support of H6.
Moreover, via the bootstrapping technique, a mediation analysis revealed that the 95 per cent confidence intervals (CIs) were wholly greater than zero for the indirect effects of team identification on gratitude towards the sponsor (95 per cent CI H1×H4 : 0.06-0.15)and purchase intentions (95 per cent CI H1×H5 : 0.04-0.13)through sponsorstadium fit.The direct effects of team identification on gratitude towards the sponsor and purchase intentions were also significant (H2 and H3).Therefore, sponsor-stadium fit partially mediated the relationship between team identification and feelings of gratitude and the relationship between team identification and purchase intentions.Further, the authors examined the indirect effects of team identification and sponsorstadium fit on purchase intentions through gratitude towards the sponsor.In addition to the direct effects (H3 and H5), the 95 per cent CIs for the indirect effects of team identification and sponsor-stadium fit on purchase intentions did not include zero (95 per cent CI H2×H6 : 0.02-0.09;95 per cent CI H4×H6 : 0.08 to 0.22), indicating these effects were partially mediated by gratitude towards the sponsor.The results also indicated that the relationship between team identification and purchase intentions was sequentially mediated by sponsor-stadium fit and gratitude towards the sponsor (95 per cent CI H1 × H4 × H6 : 0.02-0.06).Additionally, the 95 per cent CIs for the aggregate indirect effects of team identification on purchase intentions were greater than zero (95 per cent CI H1 × H5+H2 × H6+H1×H4×H6 : 0.11-0.22),and the estimated coefficient was statistically significant at the 0.01 level (see Table IV).
In order to check the robustness of the hypothesised effects, the authors also examined whether the inclusion of two control variables (past purchase visits and prior sponsor attitude) influenced the proposed endogenous variables (see Table IV).According to the results, prior sponsor attitude had a positive effect on sponsorstadium fit (γ ¼ 0.31, p o 0.01), gratitude towards the sponsor (γ ¼ 0.23, p o 0.01) and purchase intentions (γ ¼ 0.15, p o 0.05), whereas the effects of past purchase visits on the three endogenous variables were not significant.The hypothesised effects were greater than or equal to the impact of prior sponsor attitude.Therefore, in regards to the inclusion of these control variables, the results were robust.The ability of the exogenous variables to explain variations in the endogenous variables was assessed by R 2 values.The variances explained in sponsor-stadium fit, gratitude towards the sponsor and purchase intentions were 29.7, 50.3 and 51.9 per cent, respectively.

Discussion
Companies spend millions of dollars annually seeking to improve brand image and increase sales through stadium naming-rights sponsorship.Due to the size of these expenditures, it is critical to have a clear understanding of how corporate-named sport venues influence sponsorship effectiveness.In order to provide the rationale for this prediction, the hypothesised model integrated several streams of literature: SIT, congruity theory and the customer gratitude approach.The crucial concept that ties the three theoretical perspectives into the proposed framework is sponsor-stadium fit.However, previous research has largely focused on sponsor-event fit and has largely 417 Sponsorstadium fit ignored sponsor-stadium fit.This study is one of the first attempts to empirically test the antecedents and consequences of sponsor-stadium fit and thereby advances the existing knowledge in several ways.
First, the authors introduced a new perspective in terms of conceptualising sponsorstadium fit.While past studies have built a large knowledge base regarding sponsorevent fit, this study extends the literature by defining the concept of sponsor-stadium fit and testing the reliability and validity of the proposed scale.In this study, the authors defined sponsor-stadium fit as the consumers' perceptions of the similarity between a sponsor and the sponsored stadium in terms of image, mission and value (Becker-Olsen and Hill, 2006).In empirical research, the authors adopted Speed and Thompson's (2000) scale items to measure sponsor-stadium fit and tested the adopted scale in the context of a naming-rights sponsor of a football stadium in Japan.An examination of the CFA results provided strong support for convergent and discriminant validity.Incorporating the conceptually relevant construct grounded in stadium-naming-rights will advance research on sponsorship effectiveness of corporate-named sport facilities.
The second major finding was that team identification and prior sponsor attitude were the dominant factors in enhancing sponsor-stadium fit.Previous studies have confirmed the impact of prior attitude towards the sponsor on consumer responses (e.g.favourability, interest and purchase intentions; Gwinner and Bennett, 2008;Olson, 2010;Speed and Thompson, 2000).On the other hand, the results indicated that sponsor-stadium fit can be also strengthened by consumers' identification with their favourite sport team.This finding is particularly appealing because it is grounded in a sound theoretical rationale provided by SIT in combination with congruity theory.As SIT suggests, this study demonstrated that an in-group favouritism effect emerged when examining the electronics company's image fit with the non-historic home stadium of the football club.In the current setting, fans identifying strongly with the team may have positively biased evaluations and consider the stadium sponsor as an in-group member of the same local community (Wann and Grieve, 2005).Because the image consistency between the sponsor and the stadium is desirable and pleasant for the consumer ( Jagre et al., 2001), it is reasonable to believe that team identification leads to higher levels of sponsor-stadium fit.While prior research provides support for the impact of sport identification on sponsor-event fit (Gwinner and Bennett, 2008), this study extends the literature by suggesting that team identification positively influences sponsor-stadium fit in the context of a non-historic stadium.
Third, the growing interest in stadium sponsorship favours an examination of its' effectiveness.In the current study, the hypothesised model accounted for a large amount of the variance in gratitude towards the sponsor (50.3 per cent) and purchase intentions (51.9 per cent).In terms of the impact of each predictor variable, the authors found that the effect of sponsor-stadium fit on gratitude towards the sponsor (γ ¼ 0.43) was greater than that of team identification (γ ¼ 0.21).Similarly, the findings revealed that the effects of sponsor-stadium fit (γ ¼ 0.27) and gratitude towards the sponsor (γ ¼ 0.27) on purchase intentions were larger than that of team identification (γ ¼ 0.19).These results suggest that sponsor-stadium fit and gratitude towards the sponsor are closer to purchase intentions than team identification.To enhance purchase intentions through stadium sponsorship, fostering identification with the sport team is not enough.Consumers' perceptions of sponsor-stadium fit and their feelings of gratitude towards the sponsor are important prerequisites of purchase intentions.In line with this thinking, this study adds to the existing knowledge by testing the simultaneous effects of team identification, sponsor-stadium fit, and gratitude towards the sponsor on purchase intentions, while previous research investigated these effects separately (Gwinner and Swanson, 2003;Kim et al., 2010;Lobo et al., 2014;Speed and Thompson, 2000).
Fourth, this research explains the important mediating roles of sponsor-stadium fit and gratitude towards the sponsor in the relationship between team identification and purchase intentions.As shown in Table IV, the mediation analysis demonstrated that both the direct (TID→PI) and indirect effects (TID→FIT→PI, TID→GRAT→PI, TID→FIT→GRAT→PI) of team identification on purchase intentions were statistically significant.It is also worth noting that the coefficient for the aggregate indirect effects of team identification on purchase intentions through the mediators was statistically significant.These results indicated that the relationship between team identification and purchase intentions was partially mediated by sponsor-stadium fit and gratitude towards the sponsor.The authors suggest, based on these findings, that team identification is sequentially related to purchase intentions, first through sponsorstadium fit and then through gratitude towards the sponsor.From a practical standpoint, a company needs to create a strategic match between the sponsor and the sponsored stadium and only then can expect fans to feel appreciation and thankfulness towards the sponsor before implementing a commercial transaction.
From this research, practitioners can draw conclusions to better inform their marketing decisions and practice.The results of this study indicate that naming-rights sponsors of non-traditional stadiums can expect increased sponsor-stadium fit by strengthening the sense of team identification among team fans.Previous research has shown that team identification relies largely on the team's distinctiveness (Carlson et al., 2009;Mael and Ashforth, 1992), which is associated not only with fan experiences, traditions and rituals, but also with unique characteristics of home stadiums (Boyle and Magnusson, 2007;Underwood et al., 2001).Given this theoretical basis, sponsors that present a team-related brand identity and add distinctive features may be able to create a preference and image fit with sport stadiums.On the other hand, one study reported that corporate renaming of a traditional stadium can be perceived as a threat to the team's distinctiveness and to important elements such as the historical importance of the venue (Reysen et al., 2012).Therefore, significant consideration must be given to the assessment of fans' reactions, both emotional (e.g.fear and anger) and behavioural (e.g.negative word-of-mouth and consumer backlash), when changing historic stadiums' names.In contrast, as tested in this study, one can expect positive sequential relationships among team identification, sponsor-stadium fit and gratitude towards the sponsor when renaming non-historic stadiums (if sponsors succeed in creating a reasonable connection with these stadiums).

Limitations and directions for future research
Several limitations may influence the results of this study.First, the findings of this study might be considered contextual and cannot necessarily be extrapolated beyond this particular setting (the stadium sponsorship between a Division II football club and a non-historic stadium in Japan).The relationships among the proposed constructs may change across different sport settings.Particularly, it will be interesting to replicate this study in the contexts of the corporate renaming of a historic stadium and a facility sponsor providing funding that would keep the team from moving to another city.Additional efforts should be made to ascertain the nature and the directionality of the proposed relationships in these settings.A second limitation to consider is the omission 419 Sponsorstadium fit Figure 1.Theoretical framework and hypotheses

Table I .
Sponsorship context recommendation of 3.0 and All correlation coefficients are statistically significant at the 0.01 significance level ( p o0.01).The AVE value for each construct is shown in italic on the diagonal.a Correlations are taken from ϕ matrix using Mplus version 7.31 and are reported in the lower triangle of the ϕ matrix; squared correlations are depicted in the upper triangle of the ϕ matrix; b the mean scores and SDs (standard deviations) for the five constructs are calculated using IBM SPSS statistics 20.0 , the hypothesised model demonstrated acceptable fit to the data ( χ 2 /df ¼ 2.75; CFI ¼ 0.97; TLI ¼ 0.96; RMSEA ¼ 0.073; SRMR ¼ 0.051).With respect to hypothesis testing, a bootstrap TID, team identification; FIT, sponsor-stadium fit; GRAT, gratitude towards the sponsor; PI, purchase intentions through the sponsorship.*po0.05;**po0.01 2 /df ¼ 2.75; CFI ¼ 0.97; TLI ¼ 0.96; RMSEA ¼ 0.073; SRMR ¼ 0.051 Notes: