Residents’ Perceptions and Behavioral Intentions towards Mega-Sports Events: A Case Study of Beijing 2022 Olympic Winter Games

Abstract

Despite growing research on the impacts of mega-sports events, comparative studies of the value perceptions of residents in host and non-host cities are rare. Residents’ perceptions are effective indicators of their behavioral intentions, which are crucial for the success of events and for the place marketing of hosting cities. To fill this gap, this study constructed a model linking residents’ expected impacts, perceived value and behavioral intentions for mega-sports events. Informed by Social Exchange Theory, this study employed Structural Equation Modeling (SEM) to analyze 1527 valid questionnaires collected in mainland China (412 in host cities, 1115 in non-host cities). The results reveal that non-host-city residents have more substantial expectations of impacts, perceptions of value, and behavioral intentions than host-city residents. Residents in Northeast China had the highest perceptions and behavioral intentions toward the Beijing 2022 Winter Olympics, implying that the geographical location of the non-host cities is an influencing factor. Researchers and practitioners should pay attention to those differences in research design and event planning to promote the sustainable development of mega-sports events.

1. Introduction

Mega-sports events have drawn growing attention from tourism and event management research in the past twenty years [1]. The effects of mega-sports events have been paid particular attention to in terms of strengthening the host city’s image, boosting inward foreign direct investment, promoting tourism development, facilitating cultural participation, and enhancing environmental protection awareness [2,3]. The Olympics, as one of the most influential international mega-events, has been widely recognized for its role in promoting tourism development, improving environmental quality, and branding the city’s and country’s image [3,4]. For example, the Atlanta 1996 Olympics exerted a positive effect on employment in event-hosting venues, which boosted Olympic-related employment by 17% in the counties of Georgia [5]. The Sydney 2000 Olympics brought about a 30% increase in visitors’ participation in cultural activities [6]. Beijing’s air pollution reduced by 19% in one year due to the Beijing 2008 Olympic-related policies [7]. Given that these benefits contribute directly to local residents’ urban life, it is necessary to investigate to what extent they perceive these impacts and how their perceptions affect their behavioral intentions in participating and volunteering.

As indicated by relevant studies, residents’ impact expectations for mega-sports events can be the basis for perceived value, particularly of the utilitarian variety [8]. The International Olympic Committee (IOC) considers residents’ attitudes as a crucial selection criterion when evaluating potential host cities [9]. Therefore, how residents portray the impacts of the mega-events is essential in the bidding process. On one hand, residents’ expectations can turn events into “urban festivals”, generating positive experiences [4]. On the other hand, residents’ expectations are closely associated with their participation intentions and volunteering motions [8,10]. As shown by the retailing and transportation industries, consumers’ value perceptions significantly affect their behavioral intentions to purchase products or services [11]. Moreover, perceived value can be influenced by impact perceptions and contributes to behavioral intentions [12]. However, assessment of residents’ value perceptions has been less studied in mega-sports event research. Additionally, studies of impact perceptions of mega-sports events mainly concentrate on host city residents [9]. As for non-host city residents, perceived positive impacts are significantly predictive of support for Beijing’s bidding on the 2022 Winter Olympic Games [13]. However, few scholars compare host residents and non-host residents in one study. Therefore, focusing on the XXIV Olympic Winter Games Beijing 2022 (commonly referred to as Beijing 2022), this article aims to test a model linking residents’ expected impacts, perceived value, and behavioral intentions for mega-events and compare the differences and similarities between host and non-host city residents. Beijing 2022 is the second Olympic event China will host following the 2008 Summer Olympics (Beijing 2008). Previous studies have shown that Chinese residents share a common perception that Beijing 2008 has generated profound social-cultural benefits, economic and environmental impacts such as consolidating the national identity, shaping the host city’s image, facilitating investment, increasing job opportunities, and elevating residents’ awareness of environmental protection [3,13].

Residents’ impact expectation, value perceptions, and behavioral intentions toward mega-sports events are dynamic [14,15]. Residents’ perceptions of previous events can positively affect their perceptions and behavioral intentions for future events [16]. Therefore, it is necessary to know whether residents in mainland China had high expectations for the second Olympic games in China after Beijing 2008. In addition, understanding the willingness of residents in non-host cities is also essential since the live broadcasting of mega sports events has broken the spatial-temporal limitation, particularly with the help of innovative information technology. Therefore, this study aims to compare the similarities and differences in residents’ perceptions and behavioral intentions between host and non-host cites. Moreover, considering the geographical broadness of China, the study explores the regional heterogeneity in non-host cities to understand how geographical location affects residents’ perceptions.

The remainder of this paper is structured as follows. The following section presents a literature review and research hypotheses, followed by explaining the methodology and data sources. Empirical results are discussed in section four. Finally, the paper ends with discussions, theoretical and practical implications, limitations and suggestions for future research.

2. Literature Review

2.1. Theoretical Framework

Social Exchange Theory (SET) is one of the most influential theoretical frameworks for understanding an individual’s attitudes, support, and intentions in hospitality and tourism [17]. SET is “concerned with understanding the exchange of resources between individuals and groups in an interaction. Interactions are treated as processes in which ‘actors’ supply one another with valued resources” [18]. Individuals are more willing to join in the interaction process when they perceive benefits rather than costs [19]. Such evaluation is under the preconditions of reciprocity, benefits, and rationality [20]. According to the social exchange theory, residents could have a higher evaluation and stronger participation intention if they can perceive positive impacts of an event [19]. Prayag et al. [4] confirmed the mediating effect of overall attitude on the relationship between residents’ perceived impacts and support for the 2012 Olympic Games. Similarly, Chi et al. [21] explained residents’ perceived impacts significantly affected their support for the 2014 World Cup. Zhang et al. [9] also found a significant relationship between residents’ perceived impacts, satisfaction and behavioral intentions toward the Nanjing Youth Olympic Games. In this study, the three main factors in the exchange process are residents’ expectations of social-cultural, economic and environmental impacts for Beijing 2022 [22]. Therefore, we apply SET as the theoretical framework to investigate residents’ expectations of social-cultural, economic, and environmental impacts and value perceptions and behavioral intentions for Beijing 2022.

2.2. Impacts of Mega-Sports Events

Residents’ perceptions of mega-event impacts have been widely studied in different national and cultural contexts for various events, such as the FIFA 2002 World Cup in Seoul [23], the ICC Cricket World Cup 2007 in the Caribbean [24], the Beijing 2008 Olympic Games [3], the 2012 London Olympic Games [4] and the 2022 Qatar World Cup [25].

Generally speaking, the contribution of a mega-sport event is mainly shown in its economic impact, social-cultural impact and environmental impact [23,24]. The economic impact is an essential indicator of the success of significant sports events in terms of job-generating, investment increase, and tourism promotion [9,23,26]. However, events are not always promoted or staged for economic benefits, and the impacts of events are not discrete [27]. In addition, the evaluation of mega-events needs to consider the social-cultural and environmental impacts [28]. Positive social-cultural impacts of sports events include fostering residents’ pride, improving cultural communication, enhancing the cities’ image, strengthening community bonds, and promoting sports participation [4,15,23]. Some studies show that intangible cultural impact such as improving community pride and enhancing the host city’s image is regarded as a more impressive outcome by respondents rather than a tangible impact [20]. Although environmental impact perceptions affect residents’ support for future mega-events [29], positive environmental impacts such as improving residents’ environmental awareness, preserving local heritage, and promoting waste recycling are under-studied in mega-events research [2].

2.3. Perceived Value

Residents’ perceptions of mega-event impacts could influence their perceived value of the events [18]. Perceived value is defined as the “consumer’s overall assessment of the utility of a product (or service) based on perceptions of what is received and what is given”, which is regarded as “a trade-off between perceived benefits and perceived costs” [12]. Hence, perceived value in this study refers to residents’ appraisal of the value of Beijing 2022 based on their assessment of what is received (positive expected impacts) and what is given (negative expected impacts) [30]. However, measuring unidimensional value lacks validity because it is not easy for consumers to agree on the meaning of value [10]. Batra and Ahtola [31] stated that consumers purchase goods and services for two primary reasons: affective (hedonic) gratification and instrumental (utilitarian) reasons. Mega-events can simultaneously offer hedonic and utilitarian values [27].

Hedonic value emanates from sensations, emotional arousal, fun, and playfulness, mainly focusing on meeting consumers’ entertainment needs [32]. On the other hand, utilitarian value refers to a functional, instrumental and rational process [31]. Compared with utilitarian value, hedonic value is “more subjective and personal” [32]. In addition, tourists’ expectations are closely associated with a perceived value [33]. The significant relationship between tourists’ expectations and perceived value was highlighted in previous studies [8]. Hence, this study considers both respondent’s perceived hedonic and utilitarian values and proposes the following hypotheses.

H1a. 

Expected social-cultural impacts positively influence hedonic value.

H1b. 

Expected social-cultural impacts positively influence utilitarian value.

H2a. 

Expected economic impacts positively influence hedonic value.

H2b. 

Expected economic impacts positively influence utilitarian value.

H3a. 

Expected environmental impacts positively influence hedonic value.

H3b. 

Expected environmental impacts positively influence utilitarian value.

2.4. Behavioral Intentions

Perceived value could directly affect behavioral intentions in mega-sports events [10]. Behavioral intentions are defined as stated plans to perform a specific behavior [34], reflecting the individuals’ wish to conduct certain behaviors to show their ownership and disposal of services or products [35]. Revisiting is one frequently used behavioral intention in hospitality and tourism [34]. However, plans to revisit a destination occur after tourists’ first visit, which is not the case with the current study. Alternatively, this article considers respondents’ behavioral intentions to attend Beijing 2022. The article also considers the intention to volunteer as an indicator of behavioral intentions. Volunteers, as “soft infrastructure” [36], make significant contributions to mega-sports events [37].

Both hedonic and utilitarian values could significantly affect behavioral intentions [38]. For example, Nicholson and Pearce [39] found that visitors’ attending intentions toward festival events are more likely to be affected by hedonic value. Similarly, the importance of hedonic value toward behavioral intentions was emphasized in Korean consumers’ fast-food experience. However, some other studies have shown a stronger influence of utilitarian value than hedonic value [40].

H4. 

Hedonic value positively influences residents’ behavioral intentions.

H5. 

Utilitarian value positively influences residents’ behavioral intentions (Figure 1).

3. Methodology

3.1. Questionnaire Design

The questionnaire was designed to be comprised of two sections. The first section consisted of 19 items adapted from previous research in the field of tourism, mega-events, and marketing to obtain information about residents’ expectations of impacts, perceived value, and behavioral intentions. Three items measured expected social-cultural impacts (ESCI) [3,24,41]. Expected economic impacts (EECOI) were captured by three items [4,15,23]. Expected environmental impacts (EENVI) were measured using four items [9,25,41]. Four items for hedonic value (HV) and three items for utilitarian value (UV) were adapted from the HED/UT scale [42]. Two statements measured behavioral intentions (BI) for Beijing 2022: “I am willing to be a volunteer for Beijing 2022” [3] and “I intend to attend Beijing 2022” [43]. The second section of the questionnaire has seven demographic questions, including gender, age, education level, occupation, monthly income, amateur winter sports enthusiasts, and tourism industry practitioners. All items were measured on 5-point scales, where 1 = strongly disagree while 5 = strongly agree. In addition, some questions were reverse-coded to improve the quality of the questionnaire’s results. The questionnaire was designed in English and then translated into Chinese by linguists fluent in English and Chinese. The Chinese version was then back-translated into English. After completing the translation, a pilot test (N_host = 20, N_non-host = 35) was carried out to ensure that all items could be quickly and correctly understood [3]. Moreover, the content validity was also assessed by hospitality and tourism professors.

3.2. Data Collection and Data Analysis

This study’s target population is the residents of mainland China. The Seventh National Census in 2020 showed that the population of China is approximately 1.4 billion. Given this large population, the present study used a convenience sampling method, which is more suitable for collecting data from various residents in different areas [4,44]. This method may result in selection bias in sampling [24]. To address this problem, we attempted to balance the numbers of female and male respondents in different age groups [41]. Data were collected from July to August 2021. Due to the restriction of the COVID-19 pandemic, self-reporting questionnaires were distributed online via social media to 4200 potential respondents, of whom 1602 participated. The questionnaire was posted repeatedly to encourage more respondents. A total of 1527 valid questionnaires (412 from host cities and 1115 from non-host cities) were used for data analysis after removing answers with systematic missing values. IBM SPSS Statistics 26 and EQS 6.3 were used to analyze the data. First, the frequency and percentage of respondents’ demographic profiles were calculated. Next, Cronbach’s α was tested to ensure the reliability of latent factors. Then, a Confirmatory Factor Analysis (CFA) was used to examine whether the measurement items were adequate. Discriminant validity and convergent validity were assessed. A structural model was then measured by EQS 6.3. Independent sample t-tests were carried out to understand the differences between residents’ views of host cities and non-host cities. One-way Analysis of Variance (ANOVA) and t-tests were conducted to examine the differences across demographic characteristics. Finally, the regional heterogeneity in non-host cities was assessed by the structural model.

4. Results

4.1. Demographic Profile of Respondents

The sample size of 1527 valid questionnaires (412 from host cities and 1115 from non-host cities) was qualified for factor analysis, as established by previous research. Comrey and Lee [45] offered the following standards of adequate sample size for factor analysis: 300 = good, 500 = very good, 1000 or more = excellent. Nunnally [46] recommended that in determining an adequate sample size (N), the number of variables ratio (p) should be at least 10. Based on the above recommendation, the sample size satisfied the requirement of factor analysis. The respondents were composed of 700 (45.8%) males and 827 (54.2%) females. Most of the respondents were 19 to 25 years old (28.3%) or 26 to 35 years old (26.4%). These young adults accounted for more than half of the total respondents. Most of them (911, or 59.7%) had received a bachelor’s degree, and 22% had postgraduate degrees. As for monthly income, 31.3% of respondents received CNY 2500 (USD 343) or less, 24.4% received CNY 2500–CNY 4999 (USD 343–USD 685), and 16.5% received CNY 5000–7999 (USD 686–USD 1097). 438 (28.7%) were amateur winter sports enthusiasts and 262 (17.2%) were tourism industry practitioners. Of 1115 non-host city residents, 403 (36.1%) were from Northeast China, 321 (28.8%) from North China and 391 (35.1%) from South China. The above demographic data show that samples are representative.

4.2. Measurement Model

First, reliability was tested by Cronbach’s α. Internal consistency is high when Cronbach’s α is close to 0.70 [46]. As presented in

Table 1. The reliability and validity of the constructs.

	Host-Cities				Non-Host-Cities
Constructs and Items	Loading	Cronbach’s α	AVE	CR	Loading	Cronbach’s α	AVE	CR
Expected Social-cultural Impacts		0.918	0.792	0.920		0.902	0.757	0.903
Enhance host city’s image worldwide	0.866				0.831
Reinforce residents’ pride	0.912				0.902
Increase cultural exchange between residents & tourists	0.892				0.876
Expected Economic Impacts		0.889	0.733	0.891		0.895	0.758	0.904
Increase employment opportunities	0.828				0.820
Attract more business investment	0.879				0.909
Promote the development of tourism industry	0.860				0.880
Expected Environmental Impacts		0.932	0.781	0.934		0.942	0.809	0.944
Raise residents’ environmental protection awareness	0.835				0.844
Promote a low-carbon lifestyle of residents	0.946				0.925
Promote recycling	0.885				0.932
Preserve cultural and historical sites	0.866				0.893
Hedonic Value		0.972	0.899	0.973		0.972	0.898	0.973
Thrilling	0.928				0.929
Happy	0.959				0.958
Exciting	0.970				0.969
Interesting	0.935				0.935
Utilitarian Value		0.940	0.847	0.943		0.949	0.866	0.951
Necessary	0.892				0.915
Practical	0.943				0.951
Effective	0.926				0.925
Behavioral Intentions		0.770	0.630	0.772		0.889	0.803	0.891
I am willing to be a volunteer for Beijing 2022.	0.749				0.911
I intend to attend Beijing 2022.	0.836				0.881

Note(s): AVE = average variance extracted; CR = composite reliability.

, all Cronbach’s α values are higher than 0.8 except BI of host city residents, which is still 0.77, confirming the factor’s high internal reliability. Second, the values of Kaiser–Meyer–Olkin (KMO) measures for the host (0.952) and non-host groups (0.955) were all greater than 0.9, which is considered suitable for factor analysis [3].

Next, the convergent validity was assessed through factor loadings, average variance extracted (AVE), and composite reliability (CR) [47]. As shown in Table 1, all item loadings exceeded the acceptable value of 0.7 and were significant (p < 0.05) [48]. AVE depicting the variance for indicators captured by the latent construct exceeded the recommended value of 0.5 [48]. CR values exceeded the recommended value of 0.7 [48]. Moreover, the discriminant validity for each pair of factors in the CFA model was assessed. All chi-square difference tests were significant, which offered evidence of discriminant validity (

Table 2. The discriminant validity.

	Host-Cities		Non-Host-Cities
	Δχ2	Δd.f.	Δχ2	Δd.f.
ESCI & EECOI	96.1	1	242.1	1
ESCI & EENVI	327.3	1	487.0	1
ESCI & HV	133.1	1	329.4	1
ESCI & UV	239.1	1	382.9	1
ESCI & BI	32.0	1	182.4	1
EECOI & EENVI	256.1	1	309.3	1
EECOI & HV	173.7	1	301.8	1
EECOI & UV	214.7	1	327.0	1
EECOI & BI	54.6	1	183.4	1
EENVI & HV	542.2	1	1054.0	1
EENVI & UV	399.3	1	737.4	1
EENVI & BI	58.0	1	203.0	1
HV & UV	203.6	1	313.2	1
HV & BI	22.6	1	154.0	1
UV & BI	35.8	1	170.6	1

Note(s): ESCI = expected social-cultural impacts; EECOI = expected economic impacts; EENVI = expected environmental impacts; HV = hedonic value; UV = utilitarian value; BI = behavioral intentions.

) [49,50]. Discriminant validity was assessed for the variables through the Fornell–Larcker criterion. Correlations between sets of variables were lower than all AVE square roots (

Table 3. A correlation table.

	Non-Host-Cities
		ESCI	EECOI	EENVI	HV	UV	BI
Host-cities	ESCI		0.695	0.563	0.691	0.652	0.536
	EECOI	0.740		0.744	0.725	0.713	0.580
	EENVI	0.548	0.610		0.634	0.675	0.562
	HV	0.786	0.654	0.552		0.854	0.690
	UV	0.704	0.650	0.596	0.808		0.692
	BI	0.649	0.577	0.521	0.727	0.702

Note(s): ESCI = expected social-cultural impacts; EECOI = expected economic impacts; EENVI = expected environmental impacts; HV = hedonic value; UV = utilitarian value; BI = behavioral intentions.

).

Confirmatory Factor Analysis specifically handles measurement models concerning the relationships between measured items and latent factors [51]. The Maximum Likelihood (ML) estimation was adopted to estimate the parameter values [51]. The chi-square test is less reliable when the sample size is larger than 200 [3]. Hence, the multiple fit indices including normal fit index (NFI) (≥0.90), comparative fit index (CFI) (≥0.90), root mean square residual (RMR) (<0.08), and root mean square error of approximation (RMSEA) (<0.08) were assessed [50]. Both CFA models of host and non-host city groups reported evidence of good fit: host cities (x²/d.f. = 1.757, NFI = 0.983, CFI = 0.993, RMR = 0.037, RMSEA = 0.043) and non-host cities (x²/d.f. = 2.205, NFI = 0.989, CFI = 0.994, RMR = 0.022, RMSEA = 0.033). All the above fit indexes indicate that the CFA models are reasonably acceptable [52].

4.3. Structural Model

Standardized path coefficients (β), t-test, and R² estimates were used to assess structural models. R² estimates assess the predictive power of this structural model. Standardized path coefficients, known as path loading, could measure the strength of the relationship between independent and dependent variables. The recommended R² value is 0.10 [53]. The results reported that no additional tests were needed. As shown in Figure 2 and

Table 4. A structural model.

Hypotheses	SPC (β)	t-Value	Decision	SPC (β)	t-Value	Decision
	Host-Cities			Non-Host-Cities
H1a. Expected social-cultural impacts → Hedonic value	0.84	8.72 ***	Supported	0.35	5.76 ***	Supported
H1b. Expected social-cultural impacts → Utilitarian value	0.62	6.20 ***	Supported	0.28	4.55 ***	Supported
H2a. Expected economic impacts → Hedonic value	0.07	−0.71	Unsupported	0.46	5.71 ***	Supported
H2b. Expected economic impacts → Utilitarian value	0.05	0.45	Unsupported	0.43	5.24 ***	Supported
H3a. Expected environmental impacts → Hedonic value	0.13	2.31 **	Supported	0.06	1.680	Unsupported
H3b. Expected environmental impacts → Utilitarian value	0.23	3.66 ***	Supported	0.20	3.48 ***	Supported
H4. Hedonic value → Behavioral intentions	0.60	7.17 ***	Supported	0.38	7.07 ***	Supported
H5. Utilitarian value → Behavioral intentions	0.33	4.18 ***	Supported	0.44	7.89 ***	Supported

** p < 0.01, *** p < 0.001.

, both host and non-host city groups indicated an adequate model fit with the data (host cities, x²/d.f. = 2.136, NFI = 0.979, CFI = 0.988, RMR = 0.048, RMSEA = 0.053; non-host cities, x²/d.f. = 3.43, NFI = 0.983, CFI = 0.988, RMR = 0.044, RMSEA = 0.047).

Structural models of host and non-host city groups showed different statistical results. In the host-city group, contrary to expectations, EECOI did not significantly influence HV (H2a) or UV (H2b). On the other hand, the positive effect of ESCI on HV (β = 0.84, p < 0.01) (H1a) and UV (β = 0.62, p < 0.01) (H1b) was confirmed. Host-city residents’ EENVI also positively influenced their value perception for Beijing 2022 (HV, β = 0.13, p< 0.05; UV, β = 0.23, p < 0.01) (H3). Both HV (β = 0.60, p< 0.01) and UV (β = 0.33, p< 0.01) positively contributed to BI (H4 and H5).

In the non-host-city group, ESCI positively affected HV (β = 0.35, p< 0.01) (H1a) and UV (β = 0.28, p < 0.01) (H1b). Unlike the analytical results of the host-city group, EECOI did affect value perception. Hypotheses 2a and 2b were supported. EENVI was found to be unrelated to HV (H3a) but to positively influence UV (β = 0.20, p< 0.01) (H3b). Corresponding to the host-city group, BI was positively affected by HV (β = 0.38, p< 0.01) (H4) and UV (β = 0.44, p < 0.01) (H5).

4.4. Mean Comparison of EI, PV, and BI

All means of expected impacts were higher than 3.80 (

Table 5. Mean scores and standard deviations of residents’ EI, PV and BI.

	Host-Cities		Non-Host-Cities		t-Test
	Mean	SD	Mean	SD	MD
Expected Impacts
Enhance host city’s image worldwide	4.54	0.82	4.54	0.87	0.008
Reinforce residents’ pride	4.49	0.90	4.64	0.78	−0.148 ***
Increase cultural exchange between residents & tourists	4.41	0.91	4.58	0.82	−0.175 **
Increase employment opportunities	4.24	0.87	4.46	0.84	−0.222 ***
Attract more business investment	4.29	0.84	4.56	0.76	−0.272 ***
Promote the development of tourism industry	4.50	0.73	4.64	0.67	−0.145 ***
Raise residents’ environmental protection awareness	4.00	0.99	4.45	0.85	−0.451 ***
Promote a low-carbon lifestyle of residents	3.89	1.02	4.43	0.89	−0.537 ***
Promote recycling	3.80	1.05	4.40	0.90	−0.604 ***
Preserve cultural and historical sites	3.93	1.07	4.42	0.91	−0.494 ***
Perceived Values
Thrilling	4.37	0.85	4.52	0.81	−0.147 **
Happy	4.36	0.85	4.54	0.76	−0.174 ***
Exciting	4.36	0.88	4.53	0.80	−0.173 **
Interesting	4.36	0.83	4.53	0.78	−0.175 ***
Necessary	4.02	0.94	4.40	0.90	−0.382 ***
Practical	4.16	0.86	4.45	0.84	−0.290 ***
Effective	4.12	0.87	4.44	0.83	−0.314 ***
Behavioral Intentions
I am willing to be a volunteer for Beijing 2022.	3.95	1.04	4.40	0.96	−0.452 ***
I intend to attend Beijing 2022.	4.25	1.03	4.32	1.04	−0.070

** p < 0.01, *** p < 0.001.

). The item “enhance host city’s image worldwide” in the host-city group attained the highest mean of 4.54. In the non-host-cities group, “reinforce residents’ pride” and “promote the development of the tourism industry” both achieved a mean score of 4.64. Among all samples, “promoting recycling” became the item with the lowest mean (M_host = 3.80, M_non-host = 4.40).

It is worth noting that the mean of HV is greater than UV for both host and non-host cities. “Practical” of utilitarian value attains the highest score, while “necessary” has the lowest. The means of all the items measuring value are higher for non-host cities than for host cities. The results of the t-test show that each value item has significant differences between host and non-host cities. Moreover, residents from non-host cities showed a higher intention to volunteer and attend. The mean value of “volunteering” is 3.95 for the host-city group, while that of the non-host-city residents is 4.40. However, no significant difference showed up in the intention of attending.

4.5. Demographic Differences in EI, PV, and BI

To understand the differences caused by demographics, we used ANOVA and the independent t-test. Female residents rated all items higher than males, and significant differences (p < 0.05) were found in HV, UV, and BI between genders in both host and non-host-city groups (

Table 6. The demographic differences in residents’ EI, PV, and BI.

	Host-Cities (N = 412)						Non-Host-Cities (N = 1115)
	ESCI	EECOI	EENVI	HV	UV	BI	ESCI	EECOI	EENVI	HV	UV	BI
Gender
Female	4.58	4.44	4.04	4.54	4.22	4.26	4.61	4.59	4.45	4.59	4.48	4.45
Male	4.36	4.22	3.75	4.15	3.96	3.91	4.55	4.52	4.40	4.46	4.38	4.26
F value	7.67 **	9.47 **	9.88 **	23.48 ***	10.61 **	15.42 ***	1.96	3.26	1.31	8.30 **	3.96 *	12.02 **
Age
Below 18 years	4.15	3.88	3.30	4.41	4.09	4.14	4.43	4.44	4.02	4.38	4.24	4.14
19–25 years	4.56	4.38	3.95	4.51	4.21	4.25	4.66	4.57	4.37	4.65	4.55	4.52
26–35 years	4.53	4.42	3.89	4.45	4.14	4.20	4.54	4.48	4.21	4.46	4.31	4.25
36–45 years	4.40	4.29	3.84	4.04	3.89	3.81	4.53	4.62	4.53	4.51	4.41	4.32
46–60 years	4.22	4.13	3.91	4.04	3.94	3.80	4.56	4.55	4.57	4.47	4.40	4.32
60 years and above	4.67	4.46	4.48	4.46	4.23	4.00	4.83	4.80	4.77	4.73	4.66	4.50
F value	1.92	2.17	2.03	4.58 ***	1.61	3.17 **	1.96	1.91	8.77 ***	3.01 *	3.10 **	3.09 **
Education Level
Secondary and below	4.07	3.93	3.53	4.23	4.03	3.65	4.49	4.48	4.54	4.37	4.32	4.15
Vocational education	4.64	4.44	4.43	4.53	4.35	4.58	4.64	4.68	4.68	4.60	4.54	4.44
Bachelor	4.51	4.39	4.01	4.42	4.13	4.16	4.63	4.59	4.48	4.56	4.48	4.42
Postgraduate above	4.44	4.29	3.73	4.27	4.04	3.99	4.37	4.34	3.89	4.39	4.16	4.13
F value	1.44	1.81	5.99 **	1.49	1.12	4.18 **	6.11 ***	8.32 ***	33.54 ***	4.00 **	8.75 ***	6.03 ***
Occupation
Officials	4.23	4.24	3.89	4.05	3.85	3.92	4.54	4.57	4.26	4.49	4.35	4.39
Professionals	4.53	4.36	3.86	4.31	4.17	3.97	4.54	4.51	4.36	4.48	4.39	4.27
Company employees	4.45	4.40	3.95	4.37	4.10	4.06	4.57	4.56	4.44	4.48	4.36	4.36
Business owners	4.38	4.22	4.27	4.15	4.24	4.27	4.51	4.55	4.44	4.43	4.29	4.25
Students	4.58	4.34	3.76	4.51	4.15	4.28	4.65	4.56	4.33	4.63	4.52	4.50
Retirees	4.53	4.41	4.44	4.41	4.14	4.03	4.66	4.67	4.67	4.61	4.52	4.34
Other	4.32	4.16	3.82	4.17	3.88	4.04	4.54	4.54	4.55	4.50	4.45	4.30
F value	1.01	0.59	1.87	1.76	0.89	1.21	0.80	0.56	3.32 **	1.64	1.48	1.59
Amateur Winter Sports Enthusiast
Yes	4.61	4.53	4.22	4.56	4.30	4.33	4.71	4.71	4.63	4.73	4.67	4.66
No	4.43	4.27	3.78	4.28	4.02	4.01	4.53	4.50	4.35	4.45	4.33	4.24
F value	4.06 *	11.25 **	19.24 ***	10.31 **	9.42 **	10.35 **	13.00 ***	22.58 ***	27.31 ***	31.31 ***	40.35 ***	46.95 ***
Tourism Industry Practitioner
Yes	4.41	4.37	4.05	4.28	4.19	4.17	4.66	4.61	4.50	4.62	4.57	4.52
No	4.49	4.34	3.88	4.38	4.09	4.09	4.57	4.54	4.41	4.51	4.40	4.33
F value	0.46	0.08	1.54	0.71	0.72	0.33	2.28	1.78	1.90	3.76	7.34 **	7.06 **
Monthly Income
< CNY 2500	4.60	4.34	3.88	4.52	4.15	4.30	4.60	4.58	4.44	4.58	4.52	4.41
CNY 2500–4999	4.53	4.56	3.99	4.34	4.19	4.23	4.65	4.63	4.59	4.61	4.50	4.48
CNY 5000–7999	4.41	4.34	4.23	4.41	4.19	4.25	4.56	4.55	4.35	4.49	4.38	4.36
CNY 8000–9999	4.37	4.25	3.92	4.32	4.10	4.14	4.57	4.57	4.39	4.45	4.37	4.31
CNY 10,000–14,999	4.48	4.40	3.93	4.37	4.13	3.97	4.44	4.30	3.96	4.24	4.08	3.84
≥ CNY 15,000	4.41	4.28	3.72	4.16	3.95	3.89	4.29	4.23	4.04	4.22	4.01	3.96
F value	0.86	0.79	1.69	1.95	0.83	2.63 *	2.52 *	4.64 ***	9.72 ***	4.91 ***	6.57 ***	6.98 ***

Note(s): ESCI = expected social-cultural impacts; EECOI = expected economic impacts; EENVI = expected environmental impacts; HV = hedonic value; UV = utilitarian value; BI = behavioral intentions. * p < 0.05, ** p < 0.01, *** p < 0.001.

). Both sample groups showed that the 19–25 group had the highest mean score for HV and BI. As for monthly income, the significant difference (p < 0.05) in the host-city sample group is only observed in BI. For host-city residents, the relatively high-income groups conversely show lower behavioral intentions. Moreover, amateur winter sports enthusiasts from both groups have a higher mean score for expectations, value perceptions, and behavioral intentions. A significant difference (p < 0.05) was observed in UV and BI only in the non-host-city group, depending on whether they were tourism industry practitioners.

4.6. Regional Heterogeneity in Non-Host Cities

A structural model for Northeast China, North China and South China was estimated. As shown in Figure 3, the models of the aforementioned regions provided a good fit to the data (Northeast China, x²/d.f. = 1.994, NFI = 0.973, CFI = 0.986, RMR = 0.031, RMSEA = 0.051; North China, x²/d.f. = 1.628, NFI = 0.972, CFI = 0.989, RMR = 0.054, RMSEA = 0.053; South China, x²/d.f. = 2.311, NFI = 0.976, CFI = 0.986, RMR = 0.048, RMSEA = 0.052).

The statistical results reflect regional heterogeneity in non-host cities. North China was the only region that is fully consistent with the master model of the non-host city group. Northeast China showed the strongest relationship between ESCI and value perceptions while South China revealed the weakest relationship. Moreover, EECOI had the greatest impact on HV (β = 0.48) and UV (β = 0.50) in South China. Different from North and South China, no significant relationship was found between EECOI and utilitarian value in Northeast China. Only Northeast China and North China reported a significant relationship between EENVI and UV (H3b). North and South China confirmed that HV (H4) and UV (H5) positively affected residents’ BI. However, residents in Northeast China only associated utilitarian value (H5) with their behavioral intentions.

The ANOVA results showed that significant differences (p < 0.05) exist in all six factors (

Table 7. The regional heterogeneity in non-host cities.

	Northeastern	Northern	Southern	F Value
ESCI	4.69	4.45	4.61	7.93 ***
EECOI	4.66	4.41	4.57	10.02 ***
EENVI	4.61	4.04	4.49	40.15 ***
HV	4.66	4.38	4.54	10.29 ***
UV	4.58	4.21	4.45	16.31 ***
BI	4.56	4.18	4.33	12.93 ***

Note(s): ESCI = expected social-cultural impacts; EECOI = expected economic impacts; EENVI = expected environmental impacts; HV = hedonic value; UV = utilitarian value; BI = behavioral intentions. *** p < 0.001.

) for three non-host-city regions, and EENVI is the most affected by regional location (F value = 40.15). Northeast China attained the highest mean score in all factors, which was followed by South and North China.

5. Discussion

This article investigates the relationships among residents’ expectations of impacts, perceptions of value, and behavioral intentions for Beijing 2022. It compares distinct responses from host city residents and non-host city residents. The findings indicate that non-host city residents have an average higher performance in impact expectations, value perceptions, and behavioral intentions. A possible explanation is that residents are enthusiastic about the event and willing to join in the process when they perceive positive impacts, which is consistent with the social exchange theory [4]. However, non-host cities in the sample have few opportunities to host the Olympic Games, and their residents tend to show stronger perceptions and behavioral intentions toward Beijing 2022.

The overall mean of ESCI is higher than that of EECOI and EENVI, indicating that residents are more likely to expect intangible social-cultural benefits from Beijing 2022, which is congruent with previous studies [3,20]. Social-cultural impacts of events are not as immediately evident as other impacts but are never less critical than intended economic returns [27]. Thus, the current study supports the importance of intangible impacts of mega-sport events that were found to exceed the tangible impacts, which corresponds to the previous studies conducted by the contingent valuation method [54,55]. Among intangible benefit items, “pride enhancement” gained the most attention from residents, which implies that Chinese residents are eager to show China’s increasing national pride and image to the world.

The findings adhere to the social exchange theory, which emphasizes the significant relationships among impact expectations, perceived value and behavioral intentions in mega-sports events [40]. The results show that expected social-cultural impacts influence residents’ perceived value. The positive relationship between perceived value and residents’ behavioral intentions is stable among both host and non-host city respondents. However, the results suggest some incongruent findings with previous studies [8,9]. First, host city residents’ expectations of economic impacts do not affect their value perceptions. That is to say, Beijing residents do not necessarily associate the economic impacts of Beijing 2022 with its value. Learning from the Beijing 2008 experiences, Beijing residents prioritize the Olympics’ social-cultural impact and hedonic value [3,56]. Although Beijing 2008 was estimated to foster economic growth and stimulate consumption demand [57], Beijing residents realized that a mega sport event requires a huge investment in terms of hard and soft infrastructure, for which it might not be easy to generate economic benefit in the short run [58]. Hence, host-city residents show doubts on how the Olympics could affect their economic life.

Second, expected environmental impacts affect both host and non-host city residents’ value perceptions, but mainly on the utilitarian value. Environmental impacts are closely associated with valuable and convenient functions that present utilitarian value [42]. So, residents from both host and non-host cities associate environmental impacts closely with the utilitarian value, as both the Beijing 2008 Summer Olympics and the Beijing 2022 Winter Olympics emphasize the idea of green Olympics, causing the environmental friendliness concept to become deeply rooted in the hearts of Chinese residents, particularly with the help of a series of promotion campaigns.

Third, the geographical location of non-host cities matters to residents’ perceptions and behavioral intentions toward the Beijing 2022 Winter Olympics. Residents in Northeast China have the highest impact expectations, value perceptions, and behavioral intentions and showed the strongest relationship between ESCI and value perception. Northeast China is rich in natural ice and snow resources because of its geographical location and climate conditions, which makes winter sports popular among local residents. Therefore, the familiarity with winter sports gives local residents the strongest perceptions and behavioral intentions for Beijing 2022. In contrast, residents in South China can hardly feel the importance of the green development concept of Beijing 2022 while they tend to expect more tangible economic impacts. Different from the other two non-host city groups, the southern residents’ behavioral intentions are driven more by hedonic rather than utilitarian values. The lack of snowy weather constitutes a fundamental condition for their eagerness to experience winter sports and the winter atmosphere.

6. Theoretical and Practical Implications

6.1. Theoretical Implications

The current study formulates a structural model of residents’ expectations of impacts, perceptions of value, and behavioral intentions regarding Beijing 2022. Our findings offer four theoretical implications. Firstly, this research contributes to the current literature by confirming that residents’ perceived value is positively affected by their expected impacts for Beijing 2022, contributing to behavioral intentions, which is in line with the SET. It implies that relevant studies in mega-sports events should consider the role of value perceptions beyond the traditional “impact perception-behavioral intention” model. More precisely, perceived value in mega-sports events is still not very widespread, compared to other evaluations such as attitude, support and satisfaction [4,9,13,22]. Additionally, studies in event’s impacts generally consider more about how a mega-event contributes to the society, economy and environment mainly from the other-oriented dimension of utilitarian value [3,14,23,24]. However, the current study notices the self-oriented value of mega-sports events by confirmed the significant relationship between expected impacts and hedonic value [59].

Second, the differences were found in residents’ perceptions of host and non-host cities toward the same mega-sports events. A positive relationship between expected economic impacts and value perceptions was found among non-host city residents rather than host city respondents. Expected environmental impacts influence host-city residents’ hedonic and utilitarian values but only affect non-host city residents’ utilitarian values. Acknowledging and facilitating these differences is essential for event planners to formulate measures toward the specificities of different cities, which could increase residents’ support for the events [3]. Additionally, the influence of hedonic value on behavioral intentions is higher than utilitarian value among host city residents. In contrast, the impact of utilitarian value on behavioral intentions is more substantial among non-host city respondents. These findings suggest that host and non-host city residents should be distinguished in value-based research.

Third, our findings contribute to the field in terms of comparing residents’ perceptions and behavioral intentions between the two hosting experiences of Olympic events in the same city. Residents have insignificant economic impact expectations for Beijing 2022 compared to previous findings, but express the same perceptions for social-cultural impacts [15,58], implying researchers to care about changes in residents’ perceptions of the homogenous events at different times.

Fourth, regional heterogeneity was observed in non-host cities. Northeast China, North China and South China had different path relationships and strength of relationships. The findings indicate that geographical location and the distance to the hosting city could affect residents’ perception and behavioral intentions for mega-events. Therefore, regional differences can be taken into consideration in the study of mega-sports events to better understand the role of geographical spread.

6.2. Practical Implications

Except for the aforementioned theoretical implications, the findings of this study also have practical implications. First, the findings indicate that residents of both host and non-host cities have strong expectations of social-cultural impacts from Beijing 2022. Therefore, event organizers should offer residents opportunities to experience the positive social-cultural impacts of the events [9]. For example, more cultural and artistic elements could be included in the travel to maximize the cultural influence of the event.

Given that event success cannot be predicted by a single dimension of impacts [27], event organizers should inform residents of economic and environmental rewards beyond social-cultural benefits by designing an effective communication strategy. In addition, the authorities and event organizers should consider targeted communication and management strategies to increase residents’ positive impact perception and the perceived value that could affect their behavioral intentions to attend and volunteer for the events. For example, the organizing committee of Beijing 2022 published the “Low-carbon Management Pre-Games Report of the Beijing 2022 Games”. As a result, it made valuable technological innovations and breakthroughs in venue construction, emphasizing the sustainability goal.

This study can also benefit the sustainable development of sports and tourism industry by revealing the regional differences in residents’ perceptions and behavioral intentions of Beijing 2022. Market strategies can be formulated based on geographical location to achieve region-specific strategies. In addition, demographic differences should also be associated with strategy formulation. The results show that younger residents and students have stronger behavioral intentions. The younger residents tend to show a stronger hedonic value perception and intentions to attend Beijing 2022. So how to provide a once-in-a-lifetime memory for them is another challenge for event organizers. Amateur winter sports enthusiasts have higher expectations and behavioral intentions towards Beijing 2022, according to Chen et al.’s [60] and Ritchie et al.’s [61] studies.

7. Limitations and Future Research

The current study has limitations to be acknowledged. First, the study adopted convenience sampling and online distribution of questionnaires due to the restriction of the COVID-19 pandemic, which is disproportionately representative of the larger population [24]. Future research might employ stratified random sampling and data collection techniques of telephone surveys to better reflect the population and to improve the effectiveness of results [3,15]. Second, this research focuses on the residents’ expectations of impacts, perceived value, and behavioral intentions six months before Beijing 2022. Perceived impacts and behavioral intentions are likely to change with time, especially just before, during, and after the event [58,62]. The effects of such a mega-event are usually long-lasting [58]. For future research, it would be necessary to carry out a longitudinal study to compare the changes before and after Beijing 2022. Additionally, the research only measured residents’ positive impact expectations for Beijing 2022, neglecting the opposing side. Although only investigating positive constructs is typical for value-related research [63], it is still necessary to evaluate residents’ perceptions of negative impacts in the future. It is especially true since previous studies have shown that mega-events attitudes, support, and behavioral intentions might be weakened by negative expectations [3,4]. Another limitation of this study is that residents’ behavioral intentions only include plans to volunteer for and attend the event. Other behavioral intentions, such as making purchases [11], revisiting [64], and showing loyalty [10], could be investigated in future research, perhaps improving the accuracy of the results. Finally, this study is carried out in the context of China and only concerns itself with Beijing 2022. Hence, its findings might not lend themselves to be generalized and applicable to other national and cultural contexts. To improve its value more broadly, future research could focus on significant events other than Beijing 2022.