Factors Influencing Exhibitor Satisfaction and Loyalty: A Meta-Analysis on the Chinese Exhibition Market

Abstract

With the exhibition sector becoming increasingly important for the hospitality and tourism industry, it is critical to understand what drives the success of an exhibition. Through a comprehensive literature review, we first identified two broad categories of factors, namely exhibition attributes and exhibitor perceptions, that influence exhibitor satisfaction and loyalty. Next, a meta-analysis was conducted based on 26 empirical papers studying the Chinese exhibition market to quantitatively evaluate these relationships. The results show that among the exhibition attributes, booth management, service personnel, and exhibition environment are the most important factors affecting exhibitor satisfaction, while exhibition brand is the most important factor affecting loyalty. Among exhibitor perceptions, service quality is more important for satisfaction, while perceived value is more meaningful to loyalty. This study offers insights into strategies for exhibition organizers to cultivate long-term relationships, and to better cope with the challenges of emerging forces such as the Internet.

1. Introduction

An exhibition is a major form of MICE (meetings, incentives, conferences, and exhibitions) event and one of the main components of the tourism industry [1]. The global exhibition market generates about 32,000 exhibitions per year, involving 5 million exhibitors and 303 million visitors. The United States, China, and Japan are ranked the top three countries in that order in terms of the exhibition sector’s contribution to GDP [2]. Europe, Asia-Pacific, and North America offer the largest venues with 45.2%, 23.7% and 23.5% of the total indoor exhibition space worldwide, respectively [3]. Within this context, exhibition organizers aim to create a satisfactory experience for exhibitors in order to increase the chance for cooperation and future business opportunities. As such, it is important to understand what drives the success of an exhibition, particularly what leads to exhibitor satisfaction and loyalty.

Research literature on exhibitions encompasses several areas, including exhibition selection [4,5,6], attendance motivation [7,8,9], performance [10,11], effectiveness [6,12], exhibitor and attendee behavior [13,14], and service quality and satisfaction [15,16]. Many studies have investigated the relationship between exhibitor satisfaction and loyalty. Some measure the relationship between exhibitor motivation and satisfaction based on the expectation-performance model [17,18], while others focus on the influencing factors of satisfaction [18,19,20] or loyalty [4,21]. Indeed, the long-term sustainable relationship between business partners has become a strategic asset for firms, and high satisfaction and high loyalty undoubtedly point to repeated consumption, positive word-of-mouth recommendation, and long-term sustainable relationship [21,22]. In the Internet era, especially under the current circumstances caused by the COVID-19 pandemic, how to maintain or, in case of a lockdown, re-establish the long-term and stable relationship between the organizers and their users (exhibitors and visitors) is the most important topic, and satisfaction and loyalty are the starting points to answer this question.

However, few studies have conducted an integrated evaluation of how the exhibition experience affects exhibitor satisfaction and subsequent behavior. To fill this research gap, this study involved the development of an exhibitor–satisfaction–loyalty model based on a comprehensive literature review, which yielded a set of research hypotheses. Then, a meta-analysis was conducted using studies focusing on the Chinese exhibition market to test these hypotheses in order to systematically quantify the effects of exhibition attributes and exhibitor perceptions on satisfaction and loyalty. The analysis produced a single effect size for each relationship, which shows the relative importance of each factor in driving satisfaction and repeat participation in the exhibition. Finally, implications for research and practice as well as limitations and future research directions are discussed.

2. Conceptual Framework and Literature Review

A conceptual framework was developed based on a number of literature streams related to the evaluation of exhibitor satisfaction and loyalty. Some literature measured satisfaction, including the SERVQUAL scale [23,24,25,26], the Kano model [27,28,29], and important-performance analysis [30,31,32,33,34], while research on exhibitor loyalty is often associated with satisfaction, based on SCSB(Sweden Customer Satisfaction Barometer), ACSI(American Customer Satisfaction Index), and ECSI(Europe Customer Satisfaction Index) [35,36,37]. Primarily, two types of factors were identified, exhibition attributes and exhibitor perceptions, which are expected to influence exhibitor satisfaction and loyalty (see Figure 1 for a graphical representation of the research framework).

Exhibition attributes include six factors: exhibition environment, exhibition brand, service personnel, booth management, exhibition design, and service information. Exhibition environment represents the industry environment and facilities at the venue. Exhibition brand refers to the marketing promotion and brand image of the exhibition. Service personnel refers to the professionalism of the field staff and the speed of service response, etc. Booth management represents the on-site management order and logistics transportation service of the exhibition. Exhibition design represents the visual identification system in the exhibition, including the brochure, logo, booth design, etc. Service information represents whether the information related to the exhibition is updated accurately and in a timely manner. Exhibitor perceptions include two elements: service quality and perceived value. Perceptions are evaluations of the overall experience that are not tied to a particular exhibition characteristic. Service quality is the perception of personnel service, customer service, and customer relationship, while perceived value represents the extent to which the exhibition is considered worthwhile. It should be noted that the quality of service focuses on the perception of services, while the service personnel are a measure of the qualities of personnel such as professionalism and affinity. Satisfaction is the exhibitor’s overall satisfaction with the exhibition while loyalty is defined as the willingness to participate in the exhibition again and to recommend it to others.

Table 1 lists the study variables, common labels used in the literature, and a representative source that applied to each label. In meta-analysis, it is essential to ascertain the operational definitions of variables to classify them under the correct attribute [38]. Researchers use a variety of labels to identify constructs that are conceptually similar. In most cases, the measurement scales were provided, and when they were not the description of the variable in the text was used to classify it. The literature review describes illustrative sources for each variable but does not cite every article. We describe the overall pattern of relationships for each variable with satisfaction and loyalty, but not the detailed statistics. Details of the meta-analysis are given following the literature review, and the complete list of articles used in the analysis and the relationships they reported are shown in the analysis section.

2.1. Exhibition Attributes

Exhibition attributes are various objective factors that affect the exhibitor’s experience, which mainly come from the organizers and exhibition venues. The following paragraphs describe these variables.

2.1.1. Exhibition Environment

The success of an exhibition benefits from a good exhibition environment, including the industry environment, the exhibition facilities, and the surrounding environment. The industry environment includes the development environment of the exhibition venue, the market status of the industry, and the development prospects of the industry [45]. Kim et al. [46] studied the physical environment, including urban accessibility and access costs, convenience, cleanliness and comfort, entertainment facilities, and varieties of food services. In addition, Siu et al. [40] also suggest that the quality of exhibition services depends on location, exhibition time, and surrounding environment (including convenient transportation, downtown distance, catering services, and cleanliness). Furthermore, exhibition services also depend on three different environmental aspects: environmental conditions, spatial functions and signs, and symbols and artificial landscapes [47]. Many studies have shown that a good external environment creates favorable conditions for the exhibition. Therefore, the following hypotheses were proposed:

Hypothesis 1a.

There is a positive relationship between exhibition environment and exhibitor satisfaction.

Hypothesis 1b.

There is a positive relationship between exhibition environment and exhibitor loyalty.

2.1.2. Exhibition Brand

Exhibition brand includes two aspects: brand image and exhibition marketing. Exhibition brand image is an overall understanding of the exhibition brand by the exhibitors, not only based on the overall situation of the exhibition but also prior information through various channels (such as the brand’s publicity, previous exhibitions, and exhibition organizer’s publicity) and with the past exhibition experience. Brand recognition not only directly affects attendees’ satisfaction but can also influence the satisfaction of the exhibitors and the intention to participate in the future by changing attendees’ expectations. Exhibition marketing refers to the marketing strategies pursued by the exhibition organizer or planner to attract exhibitors and visitors and to establish new contacts. Friedman [48] pointed out that organizers should conduct appropriate marketing programs and channels including direct mail, fax, advertising, sponsorship, the Internet, inviting specific buyers, public relations, and various products and services. Online platforms such as websites can continue to provide exhibitors with up-to-date information to attract buyers’ attention. Based on the above analysis, the following hypotheses were proposed:

Hypothesis 2a.

There is a positive relationship between exhibition brand and exhibitor satisfaction.

Hypothesis 2b.

There is a positive relationship between exhibition brand and exhibitor loyalty.

2.1.3. Service Personnel

Service personnel have direct contact with exhibitors and serve exhibitors. The professional performance of service personnel is reflected in the hospitality registration service process [26,49]. Organizers often carry out staff training before the exhibition, so that service personnel can meet the job requirements. High-quality service personnel can improve the work efficiency, reduce the obstacles in communication between organizer and exhibitors, and better serve exhibitors [43]. The requirements for service personnel also include whether the issues and complaints of exhibitors can be resolved in a timely and effective manner [45], and their ability to handle complaints is highly correlated with the satisfaction of exhibitors and their willingness to exhibit again. Therefore, the following hypotheses were proposed:

Hypothesis 3a.

There is a positive relationship between service personnel and exhibitor satisfaction.

Hypothesis 3b.

There is a positive relationship between service personnel and exhibitor loyalty.

2.1.4. Booth Management

Booth management means that exhibitors plan ahead and arrange booth location in the exhibition. Booth management includes exhibition area planning, booth logistics, booth safety, booth distribution, and booth control [43]. Exhibitors may not be able to get the booth location they desire, especially exhibitors who are late to register and may get less than ideal locations. The quality of a booth location is directly related to the flow of people at the booth and the exposure of the company. Therefore, booth management is of paramount importance for exhibitors, and improper placement of booths may cause exhibitors to feel dissatisfied and not want to participate again [28,44,50]. In addition, exhibition area planning, booth logistics, booth safety, and booth control also affect the exhibition plan, including the display of exhibitors’ products and the safety of exhibits, which all contribute to exhibitors’ satisfaction [18]. Therefore, the following hypotheses were proposed:

Hypothesis 4a.

There is a positive relationship between booth management and exhibitor satisfaction.

Hypothesis 4b.

There is a positive relationship between booth management and exhibitor loyalty.

2.1.5. Booth Design

Exhibition design is a creative activity that affects psychology, thoughts, and behavior of the audience. In the exhibition activities, the organizers use the space environment, presenting the information and content to be disseminated in front of exhibitors and audiences with the help of exhibition facilities and various advertisements. Exhibition design includes exhibition logo design, theme design, brochure design, on-site signage, and booth decoration [19]. Organizers value and pay attention to how the exhibitors are attracted and participate in the experience. An exhibition design often impresses the exhibitors and enhances the experience. Exhibitors use logos and other designs to identify exhibition organizer or exhibition themes as well [20]. In addition, exhibition design is often used to develop effective marketing strategies and improve user satisfaction [18]. Therefore, the following hypothesis was proposed:

Hypothesis 5.

There is a positive relationship between booth design and exhibitor satisfaction.

2.1.6. Service Information

Important features of service information include timeliness, reliability, and integrity of the information. Exhibition organizers provide important service information (such as exhibition location map, exhibition activities, marketing, and websites with the latest information) to exhibitors. Before the exhibition, exhibitors are interested in booth- and budget-related information resources, including booth size, location, and design. During the exhibition, organizers publish promotional information and other events through various channels. Chen [37] found that service information is positively correlated with exhibitor satisfaction based on his empirical research. Therefore, the following hypothesis was proposed:

Hypothesis 6.

There is a positive relationship between service information and exhibitor satisfaction.

2.2. Exhibitor Perceptions

Exhibitor perceptions consist of a variety of factors that exhibitors directly or indirectly perceive of the exhibition. They are more “subjective” compared to exhibition attributes. This section explains exhibition perceptions as primarily perceived service quality and perceived value.

2.2.1. Service Quality

Oliver [51] argues that perceived service quality is an overall evaluation based on customer attitude. Perceived service quality and exhibitor satisfaction are subjective assessments of exhibitors’ services. The range of perceived service quality is more specific than exhibitor satisfaction; hence, perceived quality is more likely to have an impact on exhibitor satisfaction. In this study, service quality is defined as the comprehensive perception and evaluation of exhibitors’ various services, including service attitude, accessibility, and ability to solve customer problems in a timely and effective manner. Lee and Min [52] pointed out that service quality is an important factor explaining exhibitor satisfaction, re-exhibition, and word-of-mouth promotion. Similarly, Wei and Lin [19] stated that “service” is a main driver affecting the satisfaction of a trade show. Therefore, the following hypotheses were proposed:

Hypothesis 7a.

There is a positive relationship between service quality and exhibitor satisfaction.

Hypothesis 7b.

There is a positive relationship between service quality and exhibitor loyalty.

2.2.2. Perceived Value

Perceived value is an intrinsic value perception produced by a service compared to cost [53]. In the exhibition context some studies regard perceived value as the full price paid by the exhibitor, representing the actual sum of all economic and other costs of the exhibition, namely monetary cost or economic value [54]. Some studies divide perceived value into cognitive value and emotional value [55,56,57], whereby cognitive value is related to understanding and explaining the psychological process of exhibitors, while emotional value is related to the personal, emotional, and social impact of the exhibitor. Fang’s [29] study shows that perceived value affects exhibitor satisfaction and behavioral intention. Similar studies concluded that perceived value has a significant impact on exhibitor satisfaction and loyalty [21,57]. Therefore, the following hypotheses were formulated:

Hypothesis 8a.

There is a positive relationship between perceived value and exhibitor satisfaction.

Hypothesis 8b.

There is a positive relationship between perceived value and exhibitor loyalty.

2.3. Satisfaction and Loyalty

Satisfaction is an emotional attitude that is generated when the actual feelings of the exhibition are compared with expectations. Loyalty means that customers repeat purchases when necessary, recommend positively via word-of-mouth, and reject competitors [58]. Cardozo [59] was the first scholar to propose the concept of consumer loyalty and pointed out that it drives consumers to buy again. In the exhibition sector, exhibitor loyalty is often reflected in its willingness to exhibit again and repeated exhibiting behavior as well as word-of-mouth publicity. Further, Oliver [60] argues that consumer perceptions of service quality directly affect the assessment of loyalty, while satisfaction with services affects consumers’ behavioral intentions. In the exhibition context, Severt [61] found a significant correlation between satisfaction and word-of-mouth and repeated participation intentions. Furthermore, Fang [39] suggested that there is a correlation between exhibitor satisfaction and exhibitor loyalty. Therefore, the following hypothesis was proposed:

Hypothesis 9.

There is a positive relationship between exhibitor satisfaction and exhibitor loyalty.

3. Methodology

Meta-analysis is a critical research method developed in pedagogy, psychology, and other scientific fields and has been widely used in tourism and event research [62]. As proposed by Glass [63], meta-analysis is a systematic and comprehensive quantitative research method to synthesize the results of theoretical relationships in independent studies. It allows the researchers to use statistical methods to evaluate individual studies in the context of all existing empirical literature. Meta-analysis often is a comprehensive analysis and evaluation of a large number of research results of importance, but lacks consistency. It analyzes the relationship between variables using correlation and significance, so as to summarize the common effect reflected by the research results [64]. The meta-analysis in this study was used to analyze the relationship between satisfaction and loyalty of exhibitors in the Chinese context. The following sections outline the key aspects of the methodology.

3.1. Selection of Literature

Asia’s largest exhibition market, China, was considered a suitable setting to explore this relationship for several reasons. First of all, the UFI (Union of International Fairs) [65] “Word map of exhibition venues” report shows that in the past five years, the growth rate of indoor exhibition areas in the Asia-Pacific region was the highest in the world at 24.6%, and its market share is the only region that has grown in the past five years (3.2%). While China’s exhibition areas account for 69.9% of the entire Asia-Pacific region, the number of China’s exhibition centers with more than 5000 square meters is the second largest in the world and grew by 26.5% in the past five years. Secondly, the “UFI & Explori Global Exhibitor Insights” report [65] shows that the satisfaction of exhibitors in the Asia-Pacific region ranks second in the world. Furthermore, the willingness of exhibitors in the Asia-Pacific region to participate again is ranked first in the world. Finally, among the 317 academic publications in Chinese and English, we found that more Chinese exhibition sites were assessed—64.3% (9/14) and 83.9% (26/31) for English studies and for Chinese and English studies combined, respectively.

In order to obtain the literature on the relationships between satisfaction and loyalty of exhibitors as comprehensively as possible, we searched for peer-reviewed journal articles in both the English and Chinese languages. We searched for all relevant articles published up to September 2019 using the following key terms and term combinations. The search terms for satisfaction and loyalty of exhibitors were adopted from prior literature, including: exhibition, exhibitor, trade show, satisfaction, loyalty, perceived quality, perceived value, behavioral intention, motivation, service quality, service cognition, post-exhibition behavior, exhibitor performance. Then, the search was conducted in databases and relevant journals. The databases were CNKI(China National Knowledge Infrastructure), Wanfang Data, VIP(VIP Chinese Journals Database), Web of Science, Business Source Complete (EBSCO), Science Direct, Scopus, and Emerald. For a specific article, if no full text was available, we made sure it met the requirements in the title or abstract. Relevant management, tourism, and exhibition journals were searched to ensure full coverage. These included the International Journal of Hospitality Management, Tourism Management, Industrial Marketing Management, Tourism Management Perspectives, Event Management, Journal of Convention & Event Tourism, Management World (Chinese), Management Science (Chinese), Nankai Business Review (Chinese), Tourism Tribune (Chinese), and Tourism and Hospitality Prospects (Chinese).

This effort resulted in 317 articles with the first article published in 2000. Then, five selection criteria were used to identify a useful sample: First, considering that meta-analysis is a method of integrating and comparing different quantitative study results into an independent effect size, only quantitative articles were selected [64]. Second, the study must report the correlation between variables used in our study. Third, effect statistics required for meta-analysis must be reported and must include one of the following three groups of statistical data: sample size and Pearson correlation coefficient r, p-value, or t-value; correlation coefficient (r) and t-value, standard error, or variance; or path coefficient. Fourth, in the case of the same study being published in multiple stages, repeat publications, or different studies published on the same and cross samples, only the study with the most detailed content and larger sample sizes was included. Lastly, studied exhibition sites should be located in China. This resulted in 26 independent empirical academic articles. Among them, there were 9 articles in English and 17 in Chinese outlets; 22 articles (research data) were from Mainland China, and the remaining four articles were from Hong Kong, Macao, and Taiwan.

3.2. Description of Data

Although there were only 26 articles, the total number of quantitative samples in the selected literature was 14,303. Since meta-analysis is based on articles of quantitative nature, it is equivalent to indirectly using 14,303 observations for research and analysis. Meta-analysis has few requirements on the number of articles [66], and it is not uncommon to see relevant studies with a relatively small number of articles [53,67]. As such, our collection of literature gave us sufficient data to understand the focal relationships.

The articles covered a variety of analytical methods: regression, structural equation modeling, and so on. In terms of study variables, we examined the distribution of satisfaction and loyalty. For satisfaction, we found 12 articles on exhibition environment, 13 on exhibition brand, 10 on service personnel, 8 on booth management, 3 on exhibition design, 2 on service information, 6 on service quality, and 14 on perceived value. For exhibitor loyalty research, there were three articles about exhibition environment, five on exhibition brand, five about service personnel, seven on perceived value, and one each about booth management and service quality.

3.3. Effect Coding

The coding procedure was strictly in accordance with the literature. In this study, the coding results were composed of the primary information and effect size in the collected articles, which often refers to the correlation coefficients among variables. Primary information included authors, title, journal, publication date, sample size, region, relationship measurement or its related dimensions, research model, etc. Information on effect size included variable correlation coefficients, such as Pearson correlation coefficient, t-value, p-value, path coefficient, or β coefficient, which all could be transformed to corresponding correlations [68]. Since most of sampled studies used regression-based techniques or correlational data, we chose correlation coefficient as the effect size measure to compare or analyze across articles.

Effect size was coded by taking the simple average of correlation coefficients of different dimensions of variables in the same sample population as the effect value [69]. For example, Gu [49] reported two measures of loyalty (return intention and willing to pay); Lee [18] measured WOM(Word of Mouth) and return intention.

4. Analysis

We used Excel 2019 and Comprehensive Meta-Analysis (CMA) version 2.0 to complete the processing of data. After coding, sample size, correlation r, and the coding of each variable for our study were input into CMA 2.0 [67]. Then, the summary effect size was obtained through the calculation followed by the tests of heterogeneity and publication bias [70,71]. Hypotheses were then tested. The coding results and relationships are presented in Table 2.

4.1. Effect Sizes

Due to the different methods of the analyzed articles, it is necessary to combine the effect size of multiple studies into a general effect size. Our study adopted Hunter and Schmidt’s [66,72] meta-analytic approach. First, we modified effect size r’ = r/(${\alpha }_{xx}{\alpha }_{yy})$, where r’ is the correlation coefficient of each study after reliability correction), and r represents the Pearson correlation coefficient of each independent study; ${\alpha }_{xx}$ represents the reliability coefficients of exhibition attributes; and ${\alpha }_{yy}$ represents the reliability coefficients of exhibitor perceptions. In order to reduce the attenuation deviation of correlation coefficient caused by the reliability defect of the scale, and a small study of missing reliability values of variables, we used the weighted average reliability of other similar research samples instead. Second, Fisher’s Z-value transformation was carried out for the effect size r’: FZ = 0.5ln(1+r’)/ ln(1−r’). In this process, the distribution of correlation coefficients was transformed into a gradual normal distribution, which simplified the calculation steps of the normal estimation of the subsequent confidence interval as the size of the sample size N can no longer be considered. Then, the standard error of each effect size and the reciprocal of variance weight w_i were calculated: w_i = n_i − 3. After that, we weighted average Fisher’s Z value: Z_i = ${\sum }_{i=0}^g{\mathrm{w}}_i{\mathrm{z}}_i$/${\sum }_{i=0}^g{\mathrm{w}}_i$, where n_i is the sample size of the effect size; w_i the weight; and g the number of independent samples of an effect size, which is the number of documents containing the effect value. Finally, the final effect size r_z = (e^2Zr – 1)/(e^2Zr + 1) was obtained by reconversion. Such effect size is considered to be accurate and reliable [73]. Through the above conversion steps, 292 effect sizes were finally obtained.

4.2. Bias Analysis

Rothstein et al. [73,74] considered that publication bias may bring about a threat to validity. Thus, as a robustness test it was necessary to estimate whether there was a risk of publication bias regarding the effect sizes in sampled studies before further testing our hypotheses. Publication bias refers to the research in the published literature not systematically representing the population [74]. For our study, a funnel plot and fail-safe N were applied to estimate the potential publication bias in the sampled literature [74].

The fail-safe N refers to the number of unpublished studies that would be required to increase the p-value to an insignificant level [74]. As suggested by Rosenthal [75], when the fail-safe N is less than 5K + 10, the issue of publication bias needs to be addressed. The larger the fail-safe N, the smaller the publication bias, and the more stable the meta-analysis results.

An additional commonly used test for publication bias is the funnel plot. It suggests that the accuracy of each effect size included in the study should increase with the increase of sample size. Taking the magnitude of the combined effect as the abscissa and the standard error as the ordinate, if there is no publication bias, it should be an inverted funnel-shape; specifically, the points on the funnel plot are symmetrically scattered around the real value estimated by the study effect points. In addition, the standard error of small samples is large, which is represented by points scattered at the bottom of the funnel plot [71]. Funnel plots can be used to detect potential biases due to the small representativeness of studies with small sample size. If a set of effect sizes is unbiased and extracted from a single population, the variation between effect size based on small samples is larger than that based on large samples. Therefore, the scatter diagram between sample size and effect value should be funnel-shaped.

In order to establish the validity of our research, we first conducted a meta-analysis on the influencing factors of satisfaction and loyalty and the overall effect of satisfaction and loyalty. To do this, we used a fail-safe N and drew a funnel plot to ascertain the potential publication bias in our sample. Table 3 is the fail-safe N table of this study. The fail-safe N of this study is 21,231 (p = 0.000, α = 0.05), which significantly exceeds the critical value (5 × 26 + 10 = 140) and suggests that 21,231 unpublished research reports are needed to reverse the conclusion of this study. This implies that there was no significant publication bias in the selected literature.

Figure 2 is a funnel plot of the selected literature in this study, which is a scatter diagram of the effect value relative to the sample size. As can be seen, the funnel chart is, in general, funnel-shaped and symmetrically distributed, and the scattered points are mostly concentrated in the top part of the funnel with small dispersion, indicating that there is no serious publication bias.

4.3. Heterogeneity Analysis

In order to ascertain whether there is a significant difference among the sampled studies, we need to determine the heterogeneity of the sample. The heterogeneity test analyzes the significant differences between multiple independent studies. In meta-analysis, confidence interval and heterogeneity analysis are often used to estimate the accuracy of mean effect value. If there is heterogeneity between effect values, the range of the confidence interval is larger. We identified the heterogeneity in our sample by adopting two common methods: Q-value and I² value [70]. The Q-value reflects systematic difference. If it is significant, the systematic differences may affect the results. On the other hand, the I² value represents the percentage of total variation due to inter-study rather than sampling error. A higher I² value represents larger heterogeneity. Hence, we can choose the heterogeneity of the sample based on the Q-value and I² value, and the selection criteria are as follows. If there exists high heterogeneity, the random-effects model is more reliable instead of the fixed-effects model, and vice versa [76].

The critical value of Q is k-1 (df(Q)). If Q ≤ k-1, both models can be selected; when Q > k-1, the random-effects model is used. The standard I² values of 25%, 50%, and 75% correspond to three diverse states of heterogeneity, namely low, moderate, and high, respectively [70,76]. We used the random-effects model when I² ≥ 50%, and the fixed-effects model when I² ≤ 50%. The heterogeneity testing results are described in Table 3. In meta-analysis, homogeneous data can be combined. If heterogeneity occurred, we used the random-effects model to correct it. For example, for the test of the overall effect, the Q-value of 705.257 significantly exceeds the df(Q) (k-1 = 25), indicating the systematic difference among the 26 sampled studies that may significantly affect meta-analysis results [69]. In addition, an I² value of 96.455% indicated high heterogeneity in our sample [70]. Therefore, the random-effects model was used in our study.

5. Results

5.1. Relationships with Satisfaction

The meta-analytic results on the correlations of exhibitor satisfaction with exhibition attributes and exhibitor perceptions are shown in Table 4. There are several studies reporting the relationship of exhibitors’ satisfaction with exhibition environment, exhibition brand, service personnel, booth management, exhibition design, service information, service quality, and value perception. The results clearly show that the relationships between exhibition attributes and satisfaction are statistically significant (p < 0.001). Booth management appears to be the top driver of satisfaction with a summary effect size of 0.425, followed by service personnel (r = 0.403), exhibition environment (r = 0.392), booth design (r = 0.369), exhibition brand (r = 0.363), and service information (r = 0.319). As for the relationship between exhibitor perceptions and satisfaction, the summary effect size of service quality is 0.688 > 0.6, with the 95% confidence interval at (0.531, 0.800), indicating that there is a strong correlation between service quality and satisfaction. Therefore, H1a, H2a, H3a, H4a, H5, H6, H7a, H8a, and H9 are supported.

Cohen [77] proposed that the correlation of effect values of 0.10, 0.30, and 0.50 are considered low, moderate, and high. From the results, we see that among exhibition attributes such as exhibition environment, exhibition brand, exhibition design, service personnel, booth management, and service information, are all moderately correlated with satisfaction (0.3 < r < 0.5). Comparatively speaking, booth management, service personnel, and exhibition environment are more important factors affecting the exhibitors’ satisfaction. Among exhibitor perceptions, service quality and perceived value have a significant correlation with satisfaction, and service quality has a greater impact on satisfaction.

5.2. Relationships with Loyalty

The meta-analytic results on the correlations between exhibitor loyalty with exhibition attributes and exhibitor perceptions are shown in Table 5. There are multiple studies reporting the relationship of exhibitors’ loyalty with exhibition environment, exhibition brand, service personnel, booth management, service quality, perceived value, and satisfaction. The results show that the relationship between exhibition attributes and loyalty is significant for two measurements: exhibition brand (z = 3.292, r = 0.562, p = 0.001) and service personnel (z = 10.378, r = 0.451, p = 0.029), whereby the exhibition brand effect is stronger. In addition, the correlation coefficients between other two exhibition attributes and exhibitor loyalty are not significant; that is, there were no significant effects for exhibition environment (r = 0.491, Q = 1.457, p = 0.173) and booth management (r = 0.481, Q = 0.000, p = 1.000). This implies that exhibition environment and booth management have no direct effects on exhibitor loyalty. Regarding the relationship between exhibitor perceptions and loyalty, the results show that there is a strong correlation between perceived value and loyalty (p < 0.000), while service quality is not significant. As for the relationship between satisfaction and loyalty, the effect size is 0.497 and the 95% confidence interval is (0.376, 0.602), which implies that satisfaction significantly affects loyalty (p < 0.000). Therefore, these results support H2b, H3b, H7b, and H9, while H1b, H4b, and H8b are rejected.

Among the exhibition attributes, exhibition brand is highly related to loyalty (r > 0.5) and service personnel is moderately related to loyalty (0.3 < r < 0.5). Comparatively, exhibition brand is a more important factor affecting exhibitor loyalty. In addition, effects of exhibition environment, booth management, and service personnel on loyalty were not shown. Among exhibitor perceptions, perceived value has a significant correlation with loyalty, while service quality does not show such a relationship.

6. Discussion and Conclusions

Based on a meta-analysis of 26 useable correlation matrices published in journals up to 2019, this paper confirms that both exhibition attributes and exhibitor perceptions have impacts on exhibitors’ satisfaction and loyalty. In particular, booth management, service personnel, and exhibition environment are more important factors affecting the exhibitors’ satisfaction, while among exhibitor perceptions, service quality has a greater impact on satisfaction. In addition, exhibition brand is a more important factor affecting exhibitor loyalty; among exhibitor perceptions, perceived value has a significant correlation with loyalty, while service quality does not show such a relationship. Satisfaction is strongly related to loyalty; therefore, exhibitors who are satisfied with the exhibition will be motivated to return in the future. The findings have significant theoretical and practical implications for scholars and managerial professionals in hospitality and tourism industry.

6.1. Theoretical Implications

The present study contributes to the literature in hospitality and tourism management. First, this study shows that tangible exhibition attributes (exhibition environment, service personnel, and booth management) are more strongly related to satisfaction, whereas intangible exhibition attributes (exhibition brand) are more important for loyalty. Other studies have similar findings, such as Tanford et al. [58] who pointed out that tangible attributes may influence loyalty through satisfaction, whereas intangible attributes may influence loyalty directly.

Furthermore, our study also finds that the exhibitors are loyal (long-term effect) and take most core and key elements seriously, for example, exhibition brand and service personnel. For exhibitor perceptions, both service quality and perceived value have a significant correlation with satisfaction, while service quality has a greater impact on satisfaction, and perceived value has a significant correlation with loyalty. Service quality did not show such a relationship. This means that although the exhibitor may be dissatisfied with certain exhibition factors or service quality time, considering the price/performance ratio, they are willing to participate again.

Our research thus contributes to the literature by establishing a strong empirical foundation for event and exhibition research grounded in findings from non-service and non-hospitality frameworks and concepts. Hence, our work allows scholars to improve the existing frameworks and to look for more dynamic theoretical frameworks applicable to events and exhibitions.

6.2. Practical Implications

This study also offers implications for practitioners. First, this study suggests that exhibition organizers should pay attention to the construction and maintenance of the exhibition brand. We found that among exhibition attributes, exhibition brand is the most important factor affecting loyalty. Exhibition brand includes brand image and exhibition marketing [24], which are often mixed together. In other words, a combination of marketing and publicity are helpful to establish a good brand image, to attract the attention of a wide range of professional media and trade visitors, and to improve the visibility of the exhibition. A renowned exhibition, deeply rooted in its history with a strong public reputation in the industry, can help establish a good relationship and long-term partnership with customers.

Furthermore, exhibition organizations should strengthen their training program. The results show that service personnel could not only make exhibitors feel satisfied, but also attract them to return and participate repeatedly. High quality service personnel are not only reflected in the reception registration, but also in all aspects of the exhibition. To improve the satisfaction and loyalty of exhibitors, it is necessary for exhibition organizers to improve the professionalism of service personnel, which is conducive to reducing obstacles in communication and enhancing customer relationship between the two parties.

Finally, our findings show that the exhibition organizers should be cautious with the overall visual identification system of the exhibition and should also update relevant information in real time. We found a positive relationship between exhibition design, service information, and exhibitor satisfaction. Exhibition design is of great significance in conveying the image of exhibitors. The design of exhibition manuals, advertisements, and on-site exhibition stands can attract visitors to the exhibition and encourage participation, thus increasing the attention rate of the exhibition. For exhibition information, detailed and real-time exhibition websites are likely to create a lasting impression on exhibitors, which is the most important basis for participation decision-making compared with other exhibitions. All of these are important ways of giving the participating exhibitors a better experience.

6.3. Limitations and Directions for Future Research

This study, like all meta-analysis articles, has a selection bias. Although we used various means to include the relevant literature, the research scope may not have included all studies. Furthermore, this paper covers all effective research in the context of the Chinese exhibition market. With more empirical research in the future, it is possible to expand the sample size and retest the robustness of the hypothesis beyond the Chinese exhibition context. International comparison data for exhibitor satisfaction and loyalty from scientific studies would also allow us to put the findings from the Chinese literature analyzed in this study into context. UFI [3] provides a rough overview through the post-show research of more than 1040 trade exhibitions over 40 countries. Regarding the likelihood to return, compared with the Americas, Asia-Pacific, Europe, and Middle East/Africa, the UFI report [3] shows that exhibitors in the Americas have a more polarized view, with more respondents falling into both the “definitely will/already booked” and “definitely not” categories than other regions. Although satisfaction varies between regions, loyalty is almost similar. So, what are other factors determining whether exhibitors will support an event in future? These are the topics we need to investigate in future research. In addition, this study only analyzed the factors affecting the satisfaction and loyalty of exhibitors, and it did not take into consideration the influence of regulatory variables. In future research, situational factors such as exhibition hosting area, exhibition scale, and exhibitor position can be introduced to understand a variety of contextual factors. Lastly, this study defined satisfaction as the exhibitor’s overall satisfaction with the exhibition. However, exhibitor satisfaction can also be measured by more objective measures such as number of contracts and contacts during the exhibition. Hence, it is important to also study drivers of success in addition to satisfaction and loyalty.

Author Contributions

Conceptualization and Design, L.L., Z.X., Y.L.; Meta-analysis, Y.L., L.L.; Writing—Original Draft, L.L., Y.L.; Writing—Review and Editing, Z.X., F.J.Z., and N.M. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Social Science Fund of China “Research on Business Model Innovation of Deep Integration of Internet and Exhibition industry” (18CGL021).

Conflicts of Interest

The authors declare no conflict of interest.

Figures and Tables

Figure 1. Research framework.

Figure 2. Funnel plot.

Table 1

Variables and sources.

Variables	Other Commonly Used Variable Names	Illustrative Sources EmployingSome or All of the Variables
Exhibition environment	Market state, policy environmentIndustrial environment/agglomerationFacilities, venue equipmentSurrounding environment, catering, hotel	Fang K. et al. (2016) [39]Lee (2014) [18]Siu et al. (2012) [40]
Exhibition brand	Exhibition marketing, publicity, promotionBrand image, popularity, reputation	Zhang H. and Zhang Q. (2016) [35]Zhang (2007) [41]
Service personnel	Reception, registrationFriendly service attitudeTimely, professional, and effective communication	Breiter and Milman (2016) [42]
Booth management	Booth safetyBooth distributionSite management, construction, logisticsExhibition area planning	Lee (2014) [18]
Booth design	Exhibitors manualBooth designExhibition logo	Lin (2013) [43]Wei (2015) [19]
Service information	Update information in timeAccurate informationInformation integrity	Lin (2013) [43]
Service quality	Personnel service, customer service, customer relationship	Chien (2019) [44]Chen (2012) [26]Lee (2014) [18]
Perceived value	Functional valueCost	Breiter and Milman (2006) [42]
Satisfaction	Overall satisfactionBrand satisfactionPurpose achievement, perception, and expectation	Zhang H. and Zhang Q. (2016) [35]Zhang Tao (2018) [45]
Loyalty	Willingness to payWord-of-mouth publicityExhibiting again	Fang et al. (2016) [39]Lee (2014) [18]

Table 2

Summary of the relationship between exhibitor satisfaction and loyalty.

Citation	Location	n	Relationship Measured	Summary Effect Size	Correlation r
Wang and He (2009)	Changsha	200	1, 2, 3, 4, 7, 8, 9, 10, 14, 17	0.525	0.481
Gu, Zeng, and Yang (2010)	Shanghai	519	1, 2, 3, 4, 17	0.327	0.316
Cai and Chen (2013)	Qingdao	144	1, 2, 4, 14,	0.379	0.362
Chen (2013)	Yiwu	432	1, 2, 3, 5, 6	0.283	0.276
Yang (2016)	Shanghai	193	1, 2, 4, 14	0.431	0.406
Fang, Zhou, and Liu (2016)	Shanghai	294	1, 2, 3, 4, 14, 17	0.062	0.062
Zhang H. and Zhang Q. (2016)	Guangzhou	106	2, 3, 8, 9, 17	0.283	0.276
Wang, Hu, and Ling (2017)	Shenzhen	1030	1, 6, 14	0.361	0.346
Zhang et al. (2018)	Xinjiang	202	1, 3, 13, 14, 17	0.870	0.701
Zhang and Xu (2019)	Tianjin	202	9, 16	0.554	0.504
Wang, Zhang, and Guo (2019)	Nanning	908	9	0.432	0.407
Chen et al. (2012)	Taipei	120	3, 13	0.810	0.670
Chien and Chi (2019)	Taiwan	113	2, 4, 13, 14, 17	0.974	0.750
Fu and Yi (2019)	Shenzhen	307	1, 3, 14	0.277	0.271
Jin and Weber (2013)	Shanghai, Hangzhou, Wuhan, Nanjing	616	13, 8	1.214	0.834
Lai and Wai (2015)	Macao	412	13, 17	0.449	0.421
Lee et al. (2014)	Hong Kong	350	1, 4, 5, 17	0.611	0.545
Wei and Lin (2015)	Taipei	128	1, 2, 3, 4, 5	0.254	0.249
Wu et al. (2016)	Macao	596	13, 14, 15, 16, 17	0.757	0.639
Yi et al. (2018)	Shenzhen	304	7, 16	0.497	0.460
Jiao (2014)	Tianjin	220	14, 17	1.115	0.806
Wang (2018)	Tianjin	223	1, 2, 3, 7, 8, 9, 17	0.602	0.539
Chen (2013)	Guangxi	342	2, 14, 16	0.229	0.225
Zhang (2010)	Beijing	266	14, 16, 17	0.425	0.401
Gao (2012)	Xiamen	280	2, 14, 16, 17	0.124	0.123
Miao (2013)	Shanghai	212	2, 8, 14, 16, 17	0.537	0.491

Note: relationship measured in the study: 1. exhibition environment–satisfaction; 2. exhibition brand–satisfaction; 3. service personnel–satisfaction; 4. booth management–satisfaction; 5. booth design–satisfaction; 6. service information–satisfaction; 7. exhibition environment–loyalty; 8. exhibition brand–loyalty; 9. service personnel–loyalty; 10. booth management–loyalty; 11. booth design–loyalty; 12. service information–loyalty; 13. service quality–satisfaction; 14. perceived value–satisfaction; 15. service quality–loyalty; 16. perceived value–loyalty; 17. satisfaction–loyalty.

Table 3

Meta-analysis of global effects.

	Number of Effects	Total n	Summary Effect Size	Z	P	LowerBound	UpperBound	Q	I2	Fail-Safe N	5K + 10
Fixed effects	26	8719	0.468	47.121	0.000	0.451	0.484	705.257	96.455	13734	140
Random effects	26	8719	0.473	8.841	0.000	0.380	0.556

Table 4

Satisfaction meta-analysis results.

	Number of Effects (Studies)	Total n	Summary Effect Size	Two-Tail	95% Interval			Homogeneity Test
				Z	P	LowerBound	UpperBound	Q	Df(Q)	P	I2
Exhibition environment	12	4022	0.392	7.570	0.000	0.297	0.478	121.735 ***	11	0.000	90.964
Exhibition brand	13	3186	0.363	6.652	0.000	0.262	0.456	118.173 ***	12	0.000	89.845
Service personnel	10	2531	0.403	5.586	0.000	0.270	0.520	125.301 ***	9	0.000	92.817
Booth management	8	1941	0.425	5.136	0.000	0.266	0.544	93.415 ***	7	0.000	92.507
Booth design	3	910	0.369	3.177	0.001	0.147	0.555	23.933 ***	2	0.000	91.643
Service information	2	1462	0.319	8.666	0.000	0.250	0.385	1.854 ***	1	0.000	46.060
Service quality	6	2059	0.688	6.556	0.000	0.531	0.800	154.173 ***	5	0.000	96.757
Perceived value	14	4399	0.464	6.434	0.000	0.335	0.575	326.756 ***	13	0.000	96.021

Note: *** p < 0.001.

Table 5

Loyalty meta-analysis results.

	Number of Effects	Total n	Summary Effect Size	Two-Tail		95% Interval		Heterogeneity Test
				Z	P	LowerBound	UpperBound	Q	Df(Q)	P	I2
Exhibition environment	3	727	0.491	14.390	0.000	0.433	0.544	1.457	2	0.173	46.060
Exhibition brand	5	1357	0.562	3.292	0.001	0.261	0.764	168.990 ***	4	0.000	97.633
Service personnel	5	1639	0.451	10.378	0.000	0.375	0.521	10.772 *	5	0.029	62.866
Booth management	1	200	0.481	7.359	0.000	0.367	0.581	0.000	0	1.000	0.000
Booth design	-	-	-	-	-	-	-	-		-	-
Service information	-	-	-	-	-	-	-	-		-	-
Service quality	1	596	0.639	18.422	0.000	0.589	0.684	0.000	0	1.000	0.000
Perceived value	7	2202	0.419	4.920	0.000	0.262	0.554	104.566	6	0.000	94.262
Satisfaction	14	3993	0.497	7.096	0.000	0.376	0.602	295.237	13	0.000	95.597

Note: *** p < 0.001, * p < 0.1.