The Effect of Sport Involvement on Support for Mega Sport Events: Why Does It Matter

For residents of a city that hosts a mega sport event, sport involvement can be associated with their perceptions of the impacts and quality of life (QoL) gained from that event. The attributes of mega sport events, with multiple sports in one competition, are linked with the level of residents’ sport involvement, specifically their interest in and identification with sports, which can foster more positive perceptions and enhance the anticipated QoL from the games. Despite the importance of sport involvement on the support for a mega sport event, most studies have mainly focused on how perceptions of the impact from the event influence support based on social exchange theory (SET). Hence, this study examined how sport involvement affected the relationships among impact perceptions, QoL, and support for the 2018 PyeongChang Winter Olympic Games. Structural equation modeling (SEM) was employed to examine the hypotheses in the proposed model, using a sample of 301 Korean residents. The results revealed that sport involvement had a positive effect on event impact perceptions and QoL, which, in turn, significantly influenced support for the Olympic Games in the pre-stage. The study suggests that sport involvement can leverage support for a mega sport event through the creation of positive perceptions of the impacts of the event.


Introduction
As a mega sport event, the Olympic Games can have a significant impact on sustainable community development in the host country, specifically economic, social, and cultural development [1,2]. The success of the Games relies heavily on the local community recognizing the benefits of hosting the Games [3]. In recent years, cities bidding to host the Games have faced opposition from residents despite the potential for sustainable community development.
Residents' opposition is mainly based on fears about the potential high cost for the host city and nation [4]. Residents' support for the event is affected by their impact perceptions, which are an important factor in increasing local support for the Olympic Games [3]. Thus, exploring the process of how residents' support is affected by impact perceptions is more important and relevant than ever before. Residents are an important resource for the Games, providing volunteer services and contributing to the event's atmosphere through their interactions with spectators, athletes, and other stakeholders [5]. Quality of life (QoL) is affected by these interactions and can serve as a platform for support for the Games [6]. Therefore, scholars have paid attention to the role of QoL in the hosting of a mega sport event. Previous studies have examined the relationships among impact perceptions, QoL, and community support [7][8][9], with most of the research utilizing social exchange theory (SET). Although SET is a useful theory to explore the relationships in terms of perceived benefits and costs for

Social Exchange Theory and Sport Involvement
SET has been applied as a useful theoretical framework in a number of studies that sought to understand and explain why residents support mega events and how residents react to mega events [16]. SET is predicated on the notion that if individuals receive benefits without any unexpected costs, they will participate in the exchange process and support relevant efforts [14]. If residents perceive benefits, then the degree of support will increase [3]. Because SET can simultaneously explain the positive and negative impacts gained by a mega event, it is helpful toward providing a better grasp of residents' support of mega events which is a critical component in destination development [15].
In general, studies that utilized SET focused on dimensions such as economic, environmental, and sociocultural factors [16], when examining and explaining the perceptions of mega sport events among residents and the relationship of these dimensions to residents' support. Most of the studies using SET as their theoretical framework found that the positive and negative impact perceptions of mega events affect in a corresponding manner residents' support for mega events [8,[17][18][19][20]. However, even though SET is widely applied in various research studies [3], SET as applied to research problems related to residents' perceptions about sport event impacts or other tourism aspects has some limitations. SET is based on the assumption that individuals participate in exchange if they receive more benefits than costs in the exchange; however, the nature of the benefits and costs from the exchange can be ambiguous. For example, residents may prioritize environmental protection over economic gain or may prioritize living standards over environmental concerns [7]. In addition, the type of interaction during the exchange process can be unintentional, involuntary, or even intangible [13]. McGehee and Andereck [21] criticized the application of SET from two perspectives: a) that individuals will always choose what they believe is best for themselves, and b) that residents may change their perceptions of what is most beneficial or what is not beneficial at all. In other words, depending on the timing of the choice, what an individual considers a benefit or cost can change. Látková and Vogt [10] suggested utilizing SET along with another theory to provide a deeper understanding and perspective regarding Sustainability 2019, 11, 5687 3 of 16 the attitudes and reactions of residents, and to overcome the limitations of SET. When investigating reasons and factors that can influence the impact perceptions of residents, many researchers have implemented SET with another theory such as the theory of reasoned action [22], social identity theory [23], or Max Weber's theory of formal and substantive rationality [12]. These theories were used to explain which factors affect residents' impact perceptions. Based on these theories, many studies used extrinsic factors (e.g., seasonality, density) and intrinsic factors (e.g., demographics, community attachment) to determine the factors that have an effect on residents' impact perceptions.
In addition, residents' values and beliefs can affect their perceptions [24]. Involvement theory demonstrates that a person's involvement with an object of interest is a determinant of how the person's expectations are formed and the object is evaluated [25]. Beliefs and expectations for objects of interest can be affected by level of involvement with the object [26]. As noted by Zaichkowsky [27], involvement is "a person's perceived relevance of the object based on inherent needs, values, and interest" (p. 342). The level of involvement is revealed by individual identification towards the object of interest. In addition, the level of involvement affects a series of behavioral decisions and can act as an important moderator that explains the relationships between the variables within the behavioral decisions of individuals [28]. McDaniel and Chalip [29] asserted that a viewer's enjoyment of the television coverage of the Olympics is affected by that viewer's level of interest in the sports game that is being played during the Olympics. In addition, visitors who are very involved with the event's programs and activities can be more satisfied and intent to return to the destination in the future [30]. Involvement can be referred to as the degree of commitment regarding an object, action, or experience [31], perceived personal relevance [32], and level of psychological connection [33]. Therefore, the sports competitions can be the object of involvement in a multi-sport event such as the Olympics and can be used as antecedent to the image, perception, and attitude toward mega sport events [34].
As noted by Beaton et al. [35], "sport involvement is present when individuals evaluate their participation in a sport activity as a central component of their life and provides both hedonic and symbolic value" (p. 128). In their study, sport involvement was separated into multi-faceted constructs, including symbolic value, hedonic value, and centrality. Sport involvement can be regarded as the assessment of how much sport activity is at the center of a person's life, including watching and participating in various sport events. Therefore, sport involvement can affect the inherent expectation formation and evaluation process of the event's outcomes (e.g., economic, social, environmental) among residents with different levels of involvement. In addition, the level of sport involvement can influence the perceptions of an individual's QoL [36]. Much of the existing research has proven that sports activity can play a critical role in improving QoL [37]. Furthermore, the level of sport involvement is closely linked to an individual's amount of sport activity [38], and sport involvement can affect QoL [36]. If involvement is the level of psychological connection [33] and the degree of a person's involvement is relative to an experience or object [31], the level of an individual's involvement developed through a sports activity and experience can be seen as a preliminary predictor of QoL.

Categories and Classification of Mega Sport Event
Event impacts feature tangible and intangible dimensions, including economic (e.g., increase in employment, increase in tourist attraction), environmental (e.g., promotion of environmental issues, improvement of the local environment), and sociocultural (e.g., improvement in cultural identity, community cohesion) components [39,40]. In addition, knowledge development and sport development have been discussed within the area of sociocultural impacts [41][42][43].
Specifically, studies about economic impacts have examined the interrelationship between economic outcomes and tourism development, which could lead to economic development. From the perspective of traditional sport event studies, economic impacts include an increase in employment, a business investment, or other change that could bring economic growth such as tourism development [44,45]. Environmental aspects include problems that can change the environmental profile of the area, including recycling, improvement in air quality, awareness of the importance of the environment, and the formation of a new environmental policy [22]. According to Prayag et al.'s research [22], positive environmental impacts perception of the 2012 London Olympic Games strengthened residents' support for the Games. Furthermore, knowledge development can be brought about by community service, as well as engagement in new programs, which can make human capital development possible [1,46]. Recent studies have suggested an increased interest in the effects of intangible benefits from hosting mega events [47], and in the socio-cultural factors that stem from event hosting [39]. Key studies on impact perceptions in sport event studies are summarized in Table 1.

The Relationship between Sport Involvement, Impact Perceptions, QoL, and Support
A range of factors can affect the event impact perceptions of residents [55]. According to SET, demographic variables can affect residents' impact perceptions because these characteristics cause individuals to have different experiences regarding the impacts of mega events [21]. Furthermore, individuals may oppose development through mega events that go against their beliefs and values, considering the attitudes of residents based on SET [56]. For example, residents who think highly of cultural heritage and environmental protection tend to have a negative perception of development through mega events [24]. Therefore, individual beliefs and values have an important effect on the perception of sport event impacts. From a psychological perspective, individual beliefs and values can influence an individual's interest and relevance with that object alluding therefore to the importance of those factors in any evaluation of that object [57]. Since sports is a primary component of the Olympics, an individual's sport involvement may influence that individual's perception of the Olympics. Furthermore, the Olympic Games is not a single sport event but a combination of several sport events, leading it to be classified as a multi-sport event. Therefore, general sport involvement can be related to an individual's level of psychological connection to a multi-sport event like the Olympic Games [33] and can influence one's expectations toward the Olympic Games as an object of interest.
Previous studies also indicate that people with high level of sport involvement are likely to participate in habitual sport activities and thus enhance QoL from physical, mental benefits through activity in sports [58]. Additionally, individual's involvement with leisure activity participation can also lead to enhance QoL as the activity can be an important factor in the meaning of the individual's life [59]. Some previous research into the association of impact perceptions with support, makes the general assumption that support is a direct outcome of positive and negative impacts [60]. Positive impact perceptions can increase support among residents. Even though there has been much research within the sport event domains into residents' perception of impacts and support for events [22,60], the studies using QoL as an outcome of event impact perceptions and a predictor of event support have been limited [6]. Hiller and Wanner [42] found that the potential effect of QoL perception on event support. In addition, Ma and Kaplanidou [6] noted the influence of resident evaluations of event impacts on QoL and the positively influence of QoL on event support. Looking at the latter studies and examining the findings from an SET perspective, QoL can act as an exchange platform between impact perceptions and support [2,39]. For example, if residents perceive that event impacts influence their QoL, then residents will recognize improved QoL as a benefit, and residents' support for the event will increase as a result [6]. Using empirical and theoretical evidence, a research model with the following hypotheses was formulated ( Figure 1): Previous studies also indicate that people with high level of sport involvement are likely to participate in habitual sport activities and thus enhance QoL from physical, mental benefits through activity in sports [58]. Additionally, individual's involvement with leisure activity participation can also lead to enhance QoL as the activity can be an important factor in the meaning of the individual's life [59]. Some previous research into the association of impact perceptions with support, makes the general assumption that support is a direct outcome of positive and negative impacts [60]. Positive impact perceptions can increase support among residents. Even though there has been much research within the sport event domains into residents' perception of impacts and support for events [22,60], the studies using QoL as an outcome of event impact perceptions and a predictor of event support have been limited [6]. Hiller and Wanner [42] found that the potential effect of QoL perception on event support. In addition, Ma and Kaplanidou [6] noted the influence of resident evaluations of event impacts on QoL and the positively influence of QoL on event support. Looking at the latter studies and examining the findings from an SET perspective, QoL can act as an exchange platform between impact perceptions and support [2,39]. For example, if residents perceive that event impacts influence their QoL, then residents will recognize improved QoL as a benefit, and residents' support for the event will increase as a result [6]. Using empirical and theoretical evidence, a research model with the following hypotheses was formulated ( Figure 1): Hypothesis 1. Sport involvement will be positively associated with event impact perceptions. Hypothesis 2. Sport involvement will be positively associated with QoL perceptions. Hypothesis 3. Impact perceptions will be positively associated with QoL perceptions. Hypothesis 4. QoL perceptions will be positively associated with resident support for the Olympic Games.
Hypothesis 5. Impact perceptions will mediate the relationship between Sport Involvement and QoL. Hypothesis 6. QoL will mediate the relationship between sport involvement and support for the Olympic Games.
Hypothesis 7. QoL will mediate the relationship between event impact perceptions and support for the Olympic Games.   Hypothesis 5. Impact perceptions will mediate the relationship between Sport Involvement and QoL. Hypothesis 6. QoL will mediate the relationship between sport involvement and support for the Olympic Games.
Hypothesis 7. QoL will mediate the relationship between event impact perceptions and support for the Olympic Games.

Data Collection
This study used Survey Korea, an online market research company, to e-mail the questionnaire to potential participants/residents in South Korea. The survey company has a panel membership of more than 18,000 members in South Korea. The members are a diverse group intended to reflect the South Korean population. The initial English-language questionnaire was translated into Korean. Accuracy of the translations was checked by Korean professors with both English and Korean language proficiency. After confirming the correspondence of meaning between the two versions, the survey company emailed it to a randomly selected sample of 1000 panel members. All data was collected in March, 2017, and the research procedures and surveys received Institutional Review Board (IRB) approval. There were 301 respondents to the e-mail survey. Of the 301 respondents, 181 were male (60.1%) while 120 were female (39.9%). The average age of respondents was about 41 years of age (SD = 9.83). Most of the respondents (78%, n = 235) were between 21 and 40 years of age. The subjects of this study were mainly residents of South Korea's six big cities: Seoul, Incheon, Daejeon, Daegu, Gwangju and Busan. The subjects of past studies on the relationships among impact perceptions, QoL and support for the Olympic Games were residents of the host city. The reason, however, that subjects of this study were not limited to residents of the host city, but rather reflected the host country, was that South Korea is a very small country, and the support, desire to host, and decision-making of the central government toward the host city have greatly helped the small city of PyeongChang in Gangwon-do province. Furthermore, the opinions and support of the entire South Korean population were an important deciding factor in the decision to host the Olympics.

Measurement Scales
This study used the existing literature's scales to measure sport involvement, impact perceptions, QoL, and support, for the Olympic Games. A five-point Likert scale was employed, as recommended by Beaton et al. [35], to measure nine sport involvement items. Those nine items were categorized into three aspects of involvement (e.g., hedonic value, symbolic value and centrality). In order to measure the impact perceptions of the Olympic Games, five sub-domains (e.g., economic impacts, sociocultural impacts, environmental impacts, tourism impacts, sport development impacts) were used, and 15 items were used based on the existing literature [6,8]. All items were presented on a five-point Likert scale that ranges from 1 = "strongly disagree" to 5 = "strongly agree." QoL was also measured with one statement from Kaplanidou et al. [6] that measured overall QoL perceptions. The reason for using a single item to measure overall QoL is that the main interest was in the overall quality of life evaluation. In addition, the multi-dimensional scales of QoL feature high correlation among the sub-dimensions of items and the holistic evaluation item used in this study [6]. In addition, an overall QoL rating using a single item that reflects individuals' diverse value and preference is useful when the purpose of measuring QoL is to assess in a broad perspective rather than in the detailed description of the constructs [61]. Support for the Olympic Games was also measured on a five-point Likert scale using one item adapted from Prayag et al. [22] and Kaplanidou et al. [6].

Data Analysis
To examine the hypothesized model, the collected data were analyzed using SPSS 25.0 and AMOS 25.0. Confirmatory factor analysis (CFA) using AMOS 25.0 was utilized to test the reliability and validity of the constructs. The hypothesized relationships in the model were analyzed using structural equation modeling (SEM). Descriptive statistics, internal consistency, and Pearson bivariate correlation coefficients were computed before testing the proposed model. The average variance extracted (AVE) and the significant factor loadings were used. When the AVE value for the construct was higher than the squared correlations between the respective constructs, the discriminant validity was achieved. The multiple fit indices were used to assess the goodness of fit and parsimony of the structural model, including a chi-square (χ2) value with related degrees of freedom (df), mean square error of approximation (RMSEA), comparative fit index (CFI) [62]. CFI should exceed 0.90 and the RMSEA value in the range of 0.05 to 0.08 indicates a reasonable error of approximation although a closer fit is smaller than 0.05. [62,63].

Results
The internal consistency was assessed by Cronbach's alpha coefficients, indicating and all variables had a Cronbach's alpha level over 0.70, which indicates high internal reliability. Descriptive statistics for the sport involvement, impact perceptions, QoL, and support for the Olympic Games are presented in Table 2.

Measurement Model
As noted by Hair [64], the hierarchical component model allows the researcher to utilize a more parsimonious model with better goodness of fit indices. Thus, the hierarchical component model was employed to achieve the purpose of this study: identify the structural relationships among sport involvement, impact perception, QoL, and resident support. The data analysis was conducted on the basis of Rindskopf and Rose [65].
A hierarchical approach using a CFA model fit comparison for the second order factor structures was followed for the test procedure. Results are presented in Table 3. First, the analysis of sport involvement indicated that both model 3 and 4 had slightly better fit indices than model 1, although all were acceptable except for model 2. In selecting the best model, the second order factor was a better fit to consider than other models if it had acceptable fit indices compared to a first order factor model's correlated structure that could present problems with discriminant validity [66]. In choosing the best model from an equivalent model we took into account theoretical backgrounds [67]. Considering the Akaike Information Criterion (AIC), the best model overall was the second order structure since the lower the AIC, the better the model [68].
Second, the analyses of the impact perceptions revealed that models 1, 3, and 4 were acceptable. The AIC of model 4 was 233.116, which indicated the best model. Given the above evidence and criteria for the best model, the second order structures of the impact perceptions from conceptual and empirical evidence could be considered a better model than others. Thus, we considered the second order structure as the best model to test the structural model. Table 4 shows that discriminant validity is appropriate between constructs because the square root of the AVE for each construct in the pair was larger than the correlations between the pairs of constructs [65]. The convergent validity was assessed using standardized factor loadings, AVE values, and construct reliabilities. Table 5 shows standardized factor loadings, standard error, AVE values, and composite reliability. All standardized factor loadings greater than 0.50, and all AVEs exceeded 0.50. Composite reliability was greater than 0.70, which confirmed convergent validity of measurement scale [65].

Construct Relationships
One of main purposes of the study was to explore the relationship between sport involvement, event impact perceptions, QoL and support for the Olympic Games. Table 6 exhibits the goodness of fit measures of the model with χ2, CFI, and RMSEA, and indicates acceptable values for the structural model validity [65,69]. H1 and H2 postulated that sport involvement would be positively correlated with impact perceptions and QoL. The results indicated that sport involvement showed a significant and positive effect on impact perceptions (β = 0.51, t = 7.38, p < 0.01) and QoL (β = 0.30, t = 4.42, p < 0.01). Thus, H1 and H2 were supported. H3 postulated the positive effect of impact perceptions on QoL. The results showed a statistically significant relationship of event impact perceptions with QoL (β = 0.29, t = 4.41, p < 0.01). H4 predicted that QoL would positively affect support for the Olympic Games. The results showed a significant relationship of QoL with support for the Olympic Games (β = 0.26, t = 4.45, p < 0.01). Hence, H3 and H4 were supported.

The Mediating Effects
The mediation anlysis including BC boostrapping and the Sobel test was used to to investigate the mediating role of impact perceptions and QoL between sport involvement and support. The both methods for tesing mediation effects are more suitable than mediation analysis using a single method [70,71]. The results indicated that sport involvement had significant indirect effect toward support (sport involvement → QoL → support; β = 0.117; 95% boostrap CI (0.193 to 0.560); Z = 2.642, p < 0.01) and QoL (sport involvement → impact perceptions → QoL; β = 0.152; 95% boostrap CI (0.262 to 0.710); Z = 3.245, p < 0.01) via QoL and impact perceptions. QoL significantly mediated impact perceptions toward support (Impact perceptions → QoL→ Support; β = 0.077; 95% boostrap CI (0.077 to 0.110); Z = 2.780, p < 0.01) (see Table 7). Sport involvement and impact perceptions had no significant direct effects on support, indicating full mediating role of QoL. These results supported H5, H6 and H7. Thus, QoL can play a key role in predicting support.

Discussion
This study investigated the effect of sport involvement on support for the Olympic Games in the pre-event stage. Specifically, we explored the structural model of support for the Olympic Games, between sport involvement, and event support and featuring impact perceptions, and QoL as mediators in the model relationships. Our exploration was based on two theories: SET and involvement.
First, although both the first-order model and the second-order model showed acceptable model fit indices, the second-order model was employed to examine the structural relationships among constructs. Our results indicated that the constructs in the first-order model are co-varying, which is inconsistent with Kaplanidou et al.'s previous study [6], which identified the differential effects of impacts on QoL. The high correlation among the constructs results in multicollinearity [8]. As noted by Ma and Kaplanidou [6], the higher-order model allows to distinguish the first-order constructs, identifying the effect of different impacts on QoL and support without facing multicollinearity issues.
Second, we found that as sport involvement increased, impact perceptions became more positive. Such a result is in line with the findings of previous studies that have identified people who were highly involved in the sport are likely to be more satisfied with a sport event [34,55]. In particular, the findings imply that a high level of sport involvement affects positively event impact perceptions. One possible explanation for this result is that prior interest and experience with sports can cause residents to positively associate with attributes of the Olympic Games, given that the Olympic Games combine various sporting events. In other words, one's interest in sports and the meaning and importance of sports in one's life can be linked to the positive perception of the potential benefits of hosting the Olympic Games; specifically, residents link specific attributes of the Games (as multiple sport events) to the personal value of sports. The findings of this study extend prior findings about impact perceptions of hosting a mega sport event by suggesting the effect of sport involvement on impact perceptions of the Olympic Games and QoL [8,22].
Third, we examined the structural model, including sport involvement, impact perceptions, QoL, and support. Sport involvement positively influences impact perceptions and QoL, and then QoL has an impact on support. Event impact perceptions mediated the relationship between sport involvement and QoL, which can be explained by social exchange theory. The residents who were into sports, evaluated the event impacts positively before they observed any QoL benefits. Based on social exchange theory, positive QoL expectations functioned as a platform of positive exchange. Our findings also indicated that QoL can be simultaneously influenced by sport involvement and impact perceptions. Such findings can also be interpreted through expectations regarding the benefits of the Olympic Games and past positive sports experiences [6,8]. As sport involvement grows within the individual, the events impacts are viewed more positively and evaluated to have a more positive impact on the residents' QoL. According to SET, if residents understand that positive benefits can be gained from hosting a mega sport event, then they link this understanding with their QoL and as a result are more supportive of hosting such an event [54]. Consistent with previous studies [6,8], findings of this study indicated that QoL can be an important factor in the relationships between sport involvement, impact perceptions, and support for mega sport events. Additionally, QoL perceptions can be directly affected by sport involvement, resulting in increased support. A high level of sport involvement means that there is strong interest and some degree of participation in sports activities, which can also lead to a positive effect on QoL [36]. Because sport involvement reveals how important one considers sports and how interested one is in sports, it can affect people's perceptions of the impact the Olympic Games may have on QoL and, in turn, generate more support.

Theoretical Implications
Within the structural model of support for the Olympic Games, the association of sport involvement with QoL positively influences support for the Olympic Games. Residents who have both a high QoL and high sport involvement are more likely to support the Games as a multi-sport event with various points of attachment (e.g., interest in sports and identification with sports). This means that if participation or interest in sports is strong, an individual's QoL will be high and he or she will have a positive perception of the impact of the Games. Although the benefits of sports activity, such as improvement in health, social connections, and psychological happiness [72] might not be directly linked to the perception of impact of the Olympic Games, such positive benefits may indirectly affect the relationships among impact perceptions, QoL, and support. The results of this study add to previous findings on these relationships [34,56] by suggesting that sport involvement plays an influential role in generating support for the Olympic Games because of its impact on both impact perception and QoL in the pre-event stage.

Managerial Implications
Given the current challenge of resident resistance in cities bidding to host mega sport events (e.g., Calgary 2026 just had a referendum that rejected its bid for the 2026 Olympic Games), this study can serve as an advising tool to target or nurture societies where sport involvement is high. This approach can create support for hosting large events through appreciation for the event's impacts. Practically, experts serving on Olympic committees must be aware that sport involvement has an effect on support for hosting a mega sport event. Generally, organizers carry out various events and promotions prior to hosting a mega sport event, such as the Olympic Games, in order to garner support and interest. During these events, using promotions and strategies to increase sport involvement, while also raising interest in the specific events within the larger event, will help create focus and support toward the mega sport event.

Limitations and Future Research
This study has a few limitations. First, this study's data was collected using panels maintained by an online survey company. Thus, there may be issues regarding how the population's general characteristics can be reflected. Specifically, the panels held by the online survey company are composed of those who have access to the internet. Future studies should conduct surveys of those who actually reside in the city or country hosting the mega sport event. Furthermore, future studies could utilize secondary data to reflect actual changes in economic and sociocultural numbers in conjunction with psychological factors, such as perceptions of the Olympic Games, which would improve the study's accuracy by combining subjective factors, such as impact perceptions and QoL, with objective figures (e.g., economic, sociocultural, and QoL values).
Second, this study measured sport involvement as general sport involvement. Because the subject of this study was the Winter Olympic Games, measuring involvement with winter sports activities may reveal a different perspective. Furthermore, future studies on the support for mega sport events should reflect the effects of past experiences of hosting mega sport events on the impact perceptions, QoL, and support of residents. In addition, consideration must also be given to the differences between the social and cultural characteristics of nations and the extent of development within each nation. For example, soccer is not simply considered a sport but a sociocultural phenomenon, and even a religion, in Brazil [73]. Thus, the level of involvement with soccer in Brazil will be higher than in other countries, which has an effect on the support for sport events, such as the 2014 World Cup [73].
Third, this study considered residents' impact perceptions based on five event impacts because the purpose of this study was to analyze the structural relationship in the model instead of the association of sport involvement with each impact perception. Thus, we employed second-order constructs to model a generic concept of impact perceptions. Although second-order modeling in this study was effectively used to examine the relationships among constructs, the impact of sport involvement on each impact perception may vary. For example, the aspects of sport development, such as improved sport infrastructures and development of local sports teams, can be more closely associated with residents who are highly involved in sports [42]. Thus, additional research should be conducted to demonstrate the impact of sport involvement on each impact construct in the model of support for mega sport events, providing other empirical evidence of sport involvement. Furthermore, future studies examining the moderating role of sport involvement in the associations among impact perceptions, QoL, and support can provide more comprehensive outcomes in the context of mega sport events studies.
Finally, this study did not take into account the attitude of the residents toward the organizing committee or government associated with the Olympic Games. The favorable attitude of residents toward the organizing committee or government in the bidding and planning process may influence impact perceptions, arousing organizational citizenship behavior in regard to support. As explained by Zoghbi and Espino [74], the attitudinal environments associated with partners and tasks influence organizational citizenship behavior. Thus, favorable attitudinal environments connected with the organizing committee can help to increase organizational citizenship behavior among residents, in addition to affecting positive perceptions and support for the event. Thus, future study should consider organizational citizenship behavior as a crucial factor, or construct, in the model of support for mega sport events.
Funding: Publication of this article was funded in part by the University of Florida Open Access Publishing Fund.

Conflicts of Interest:
The authors declare no conflict of interest.