Reexamining event attributes, benefits, and values in a time of pandemic: A multi-faceted approach

Abstract

The COVID-19 pandemic increased attention to product/service-customer relationships in the context of events where social gatherings take place. Event attributes have been a topic of continued interest in research and practice. Due to COVID-19, however, event attendees' preferences and requirements related to personal values and satisfaction may be changing. Using a sequential mixed-method exploratory strategy with means-end chain theory and the Kano model as a research framework, this study identifies important event attributes, benefits, personal values, and specific satisfaction attributes. Key findings highlight pleasure and excitement and sense of safety as salient personal values. Respondents also found health/safety attributes particularly important, considering them must-be attributes. Providing health/safety services prior to the event would be effective in decreasing dissatisfaction. Implications for researchers and practitioners are discussed.

1. Introduction

The World Health Organization (WHO) officially declared COVID-19 a global pandemic on March 11, 2020, as the virus continued to spread (WHO, 2020). The United States federal government declared a national state of emergency in January 2020, followed by a March recommendation from the Centers for Disease Control and Prevention (CDC) to cancel events with 50 or more people (CDC, 2020). Echoing the impact of the 2003 SARS epidemic and the 2009 H1N1 (Swine Flu) pandemic, the COVID-19 pandemic has substantially affected the tourism industry, with event markets among the hardest hit tourism sub-sectors (Gössling, Scott, & Hall, 2021). Public health and safety concerns led to numerous domestic and international events being cancelled or rescheduled for 2021 (Event Marketer, 2020). Industry statistics indicate that event businesses across industry sectors - meetings (MPI, 2020), live events (Lampert-Greaux, 2020), and exhibitions (Shaw, 2020) - have suffered cancellations, economic hardship, and financial losses since March 2020.

A major source of local economic benefit also became a casualty of pandemic-related cancellations: the state fair. The U.S. state fair industry generates more than $4.7 billion in revenue each year and attracts as many as 150 million residents and tourists (International Association of Fairs and Expositions, 2019). State fairs contribute directly to urban and rural communities' economies. Their cultural and social significance are a celebration of a region's agricultural heritage and a driver of economic activity (Deliso, 2020). According to Smith Travel Research (STR)’s trend reports, hotel performance for the months when state fairs were held in 2019 (e.g., Arizona State Fair, Iowa State Fair, Minnesota State Fair, and Ohio State Fair) showed year-over-year growth in occupancy, average daily rate, and revenue per available room. The increasing role of state fairs, however, has been put on hold during the pandemic. In an effort to minimize disruption and continue providing event services, many leading industry organizations such as the Events Industry Council (2020), International Festivals and Events Association (2020), and International Association of Exhibitions and Events (2020) advised event organizers to protect the health and safety of attendees through relevant policies, procedures, plans, and practices.

In such an uncertain event environment, organizers face the challenging dynamics of individuals' risk perceptions toward COVID-19 infection. An event is a highly social function with inherent crowding. Customers' perception toward events may negatively affect their psychology and behavior more than ever before in public gathering spaces where physical/social distancing is taking place. While some studies show that young adults, novelty seekers, experienced travelers, and repeat visitors travel despite perceived risks (Elsrud, 2001; Lepp & Gibson, 2003; Rittichainuwat & Chakraborty, 2009), the spread of disease largely serves as a compelling deterrent to travel (Kozak, Crotts, & Law, 2007).

Amid the pandemic, consumers' expectations for event products/services are subject to change. As such, event organizers need to be able to examine event attributes in order to track and predict possible changes, find opportunities to improve services, and seek ways to offer services that can fulfill certain needs. Product and service attributes significantly influence customer perception and consumption experience (Mittal, Ross, & Baldasare, 1998; Xiang, Schwartz, Gerdes Jr., & Uysal, 2015; Xu, 2020). In particular, customer requirements change rapidly with uncertainty (Chong & Chen, 2010; Reeves & Bedner, 1994). From a business perspective, one way for firms to gather insight from customers in times of change or adjustment is to examine product/service attributes and customer benefits and values (Freeman & Perez, 1988; Kumar & Grisaffe, 2004), then identify requirements/attributes that can lead to customer satisfaction (Kano, Seraku, Takahashi, & Tsuji, 1984).

Scholars have been researching event attributes for decades. Numerous studies have shown that event attributes are associated with outcome variables such as event motivation and attendance (Crompton & McKay, 1997; Gursoy, Spangenberg, & Rutherfold, 2006; Nicholson & Pearce, 2001), emotions (Grappi & Montanari, 2011; Lee, Lee, Lee, & Babin, 2008), benefits (Chen, 2006; Fu, Zhang, Lehto, & Miao, 2018), values (Lee, Lee, & Choi, 2011; Lee, Lee, & Yoon, 2009; Yoon, Lee, & Lee, 2010), and satisfaction and behavioral intentions (Baker & Crompton, 2000; Gannon, Taheri, & Oly, 2019; Song, Lee, Kim, Bendle, & Shin, 2014; Tanford & Jung, 2017; Thrane, 2002; Velikova, Slevitch, & Mathe-Soulek, 2017; Yuan & Jang, 2008). Little is known, however, about risk-related event attributes, and the impact of an attribute in a pandemic context is even less known. Past studies examined event-related risks and safety/security issues, providing meaningful implications for practice (Boo & Gu, 2010; Sohn, Lee, & Yoon, 2016; Swart, George, Cassar, & Sneyd, 2018; Toohey & Taylor, 2008). Nonetheless, specific discussions regarding health/disease risks and relevant event products/services are scarce. Indeed, a dearth of empirical evidence in prior research leaves much to be learned about how people perceive or respond to a specific event attribute during a pandemic. Moreover, our review of recent and relevant research suggests that events as a product or service have not been adequately considered from the consumers' point of view. We argue that this knowledge is critical for event management approaches to potential attendees and event service/product development during and after the pandemic. It is the kind of empirical research we believe is now necessary.

Taken together, our research has two purposes. First, assuming current event products/services need to be adjusting to change, we will identify what event attributes customers consider more important than others during the pandemic. Grounded in the notion of the attribute-consequence-value sequence in means-end chain theory (Gutman, 1982), this study seeks to understand why customers prefer certain event attributes, what benefits they perceive, and what personal values they ultimately obtain. Second, we seek to determine which event attributes are more likely to increase customer satisfaction. As COVID-19-related risks emerge, we postulate that certain requirements or event attribute expectations exist. In turn, certain attributes contribute to event satisfaction more than others. Based on the traditional Kano model (Kano et al., 1984), we will identify potential satisfaction attributes by measuring how people perceive the functional and dysfunctional form of a quality attribute. Accordingly, this study presents a framework for integrating the means-end chain theory and Kano model. This contextual framework has been validated in exploring customer experience to better understand customers' inner thinking toward product/service attributes in the context of tourism and hospitality (Lin, Fu, & Fu, 2021; Lin, Fu, & Li, 2018; Lin, Shiu, & Wu, 2012). By integrating two theoretical models, we broaden our knowledge of links among event attributes, consequences/benefits, and personal values while gaining more insight on cognitive structure for achieving customer satisfaction through event attributes in a time of pandemic.

2. Literature review

2.1. Event attributes, benefits, and values

While no consistent definition of event attributes exists in the literature, scholars have widely used tangible and intangible event characteristics, physical features, and service aspects of an event to describe them. This widely held view may reflect different constructs or abstract terms such as utility, perceived quality, or festival environmental cues, depending on how each researcher viewed event-specific attributes (Grappi & Montanari, 2011; Gursoy et al., 2006; Nicholson & Pearce, 2001; Tanford & Jung, 2017; Yoon et al., 2010; Yuan & Jang, 2008).

Many studies have identified a certain attribute, or a dimension composed of multiple attributes, in addressing linkages between visitor experience, satisfaction, and/or behavioral intentions (Crotts, Pan, & Raschid, 2008; Girish & Chen, 2017; Tanford & Jung, 2017; Thrane, 2002; Velikova et al., 2017; Yuan & Jang, 2008). For example, in a wine/food festival, attributes such as wine variety/tastings (Crotts et al., 2008) and wine/personnel/entertainment (Velikova et al., 2017) have been found to drive visitor experience and satisfaction more than others. Additionally, design elements (e.g., atmosphere) were found to contribute significantly to visitors' perceived value in these festivals (Axelsen & Swan, 2010; Fu et al., 2018; Yuan & Jang, 2008), implying that certain attributes contribute to values more than others. Overall, the literature has firmly established that event attributes and customer values link to satisfaction. The impact of a health/safety-related attribute has been rarely considered.

Regarding the benefits or consequences of events, studies have examined these factors either from the perspective of the tourism industry or the host community (Getz, 2008; Getz, 2010; Thrane, 2002; Yolal, Gursoy, Uysal, Kim, & Karacaoğlu, 2016; Yürük, Akyol, & Şimşek, 2017). For example, Yolal et al. (2016) demonstrated that social benefits (i.e., community benefits and cultural/educational benefits) positively affect residents' subjective well-being. Ample discussion also exists on the personal benefits or positive personal impact of participating in events (Cole & Illum, 2006; Fu et al., 2017; Gitelson, Kerstetter, & Kiernan, 1995; Lee, Arcodia, & Lee, 2012; Meretse, Mykletun, & Einarsen, 2016). Lee et al. (2012) found that the benefit visitors gain by attending a multicultural festival is a transformational one, namely a change in their attitude toward migrants/the dominant population. In a food festival context, Mertse et al. (2016) found that customers' three most important benefit factors were food, enjoyment and atmosphere, buying and tasting, and tradition and celebration. From a marketing perspective, motivation-based event benefits often are used for market segmentation by examining what people seek at events (Chang, 2006; Li, Huang, & Cai, 2009). Many empirical studies have discussed event benefits in the broader context of event tourism or participation motivation, yet few have considered the benefits customers derive from observing or experiencing an individual attribute.

Regarding values, some scholars have measured personal values through many associated variables (Carlson, Rahman, Rosenberger III, & Holzmüller, 2016; Fu et al., 2018; Hede, Jago, & Deery, 2005; Lee, Lee, & Arcodia, 2014; Woosnam, McElroy, & Van Winkle, 2009). Woosnam et al. (2009) used Kahle's nine-item List of Values scale to find that entertainment was the dominant value among festival attendees. With a theater event, Hede et al. (2005) identified three personal value domains including extrinsic (e.g., to have a sense of belonging), hedonistic (to have fun and enjoyment), and achievement-based (e.g., to be self-fulfilled). They argue that personal values (i.e., end states) appear related to attribute-specific satisfaction but not to overall satisfaction (Hede et al., 2005). A recent study has found that event attributes (e.g., event design) are directly related to benefits (e.g., enjoyment) and value (e.g., emotional value) (Fu et al., 2018) in terms of causal relationships among event attributes, benefits, and personal values. The results showed benefits as a mediating variable that increases the influence of event attributes on personal values (Fu et al., 2018). Taken together, prior studies exploring multiple event attributes have contributed to a more comprehensive understanding of customer experience dynamics that includes benefits, perceived values, personal values, satisfaction, and behavioral intentions.

2.2. Means-End Chain theory

Means-end-chain theory (MEC) is “a model that seeks to explain how a product or service selection facilitates the achievement of desired end states” (Gutman, 1982, p.60). The theory assumes that consumer knowledge concerning a product or service is hierarchically organized into three levels of abstraction: attributes, consequences, and values (Gutman, 1982; Reynolds, Gengler, & Howard, 1995). MEC theory posits that consumers choose a product or service according to the attributes that provide them with the benefits (i.e., consequences) that will satisfy their desired end states (i.e., personal values) (Gutman, 1982). The attribute-consequence-value (A-C-V) relationship proposed by MEC theory postulates that positive consequence variables can be acquired in the form of benefits a consumer feels after consuming product/service attributes. These, in turn, fulfill values most personally relevant to consumers.

In a modified model of this theory, Olson and Reynolds (1983) further divided each level of abstraction. Attributes associated with experiencing the product or service can be concrete (e.g., physical features) or abstract (e.g., product style), while the consequences can be functional (e.g., convenience) or psychological benefits (e.g., accomplishment). Values can be instrumental (e.g., honest) or terminal (e.g., self-esteem), the latter of which is related to final states of existence (i.e., end states).

In a tourism and hospitality context, the MEC approach has been widely used to investigate various topics, including motivation, perception, attitude, and behavioral intention. This approach has helped scholars explore the hierarchical relationship among attributes, consequences/benefits, and values (Jiang, Scott, & Ding, 2015; Kim & Kim, 2019; Kim, Kim, & King, 2016; Klenosky, 2002; López-Mosquera & Sánchez, 2011). Prior research has shown the implications of achieving desired values/end states that are likely to influence behavior. Indeed, MEC theory has proven useful in understanding personal values as the basis for tourist behavior (McIntosh & Thyne, 2005).

Relatively little attention has been devoted so far to using the MEC approach in an event-specific context. One of the few empirical studies to do so applied this approach to a community-based heritage festival, identifying four event attribute factors (i.e., event design, historical re-enactment, physical facets, and social interactions), two perceived benefit factors (i.e., symbolic meanings, personal benefits), and four perceived value factors (i.e., emotional value, social value, functional value for quality, functional value for price) (Fu et al., 2018). In another recent study, Yeh and Lin (2017) asked tourists to recall their nostalgic experiences at a cultural festival and identified 10 attributes (e.g., enjoy the battle array performance), eight consequences (e.g., learn and understand history and culture), and seven values (e.g., sense of beauty). In the realm of sporting event tourism, Chen (2006) interviewed members of a university sports fan club and explored MEC theory's A-C-V continuum using a laddering technique. Overall, while method variance exists in previous studies, MEC-based methods make it possible to reveal relationships among attributes, consequences/benefits, and personal values.

2.3. Kano model

The Kano model (Kano et al., 1984) distinguishes among different quality attributes to identify customer needs and examines how product attributes or features affect satisfaction. Because of its ability to assess attribute-level performance, this model has been widely used in a variety of contexts and industries (Gregory & Parsa, 2013) when developing a new product during the product's/service's design stage or when improving the product's life cycle (Mikulić & Prebežac, 2016; Witell, Löfgren, & Dahlgaard, 2013). The Kano model considers the non-linear nature of the customer satisfaction function, meaning the relationship between attribute performance and customer satisfaction is asymmetric and non-linear. The model typically classifies product or service attributes into five categories. The first, must-be attributes, represents attributes that are expected to be implicitly present. These attributes cause dissatisfaction when not present because the customer takes them for granted but do not result in satisfaction when fulfilled, creating the non-linear impact on satisfaction. One-dimensional attributes, the second category, result in satisfaction when fulfilled and in dissatisfaction when not fulfilled, thus they have a linear impact on satisfaction. Attractive attributes, the third category, provide satisfaction when fulfilled but do not cause dissatisfaction when unfulfilled because the customer does not expect them, resulting in a non-linear impact on satisfaction. The fourth and fifth categories are for indifferent attributes, which result in neither satisfaction nor dissatisfaction regardless of fulfillment, and reverse attributes, which result in dissatisfaction when present (Kano et al., 1984).

The tourism and hospitality scholars have used the Kano model in various contexts, including bed & breakfast lodging (Lin et al., 2012), in-flight services (Go & Kim, 2018), restaurants (Chen, 2014; Mathe-Soulek, Slevitch, & Dallinger, 2015), and hotels (Albayrak, 2019; Chiang, Chen, & Hsu, 2019; Yang, Cheng, Sung, & Withiam, 2009). Studies also have modified the model through various conceptual approaches, such as using a three-factor satisfaction model (Albayrak, 2019; Füller & Matzler, 2008; Matzler & Renzl, 2007) and integrating it with other tools to clarify customer requirements. Examples of this include importance-performance, asymmetric impact-performance, and penalty-reward-contrast analyses (Albayrak & Caber, 2015; Mikulić & Prebežac, 2011; Velikova et al., 2017).

While researchers increasingly have applied the Kano model to a variety of tourism and hospitality products/services by refining and integrating it with other approaches (Pandey et al., 2013), relatively little research has focused on event-related products/services. Several empirical studies, however, do support the application of the Kano concept to event attendees, including those at wine festivals (Velikova et al., 2017) and an Oktoberfest beer tent (Harrington, Von Freyberg, Ottenbacher, & Schmidt, 2017), as well as residents who attended a temple festival (Yeh, Lin, & Lu, 2019). Using a penalty-reward-contrast analysis grounded in the Kano model, Velikova et al. (2017) identified five wine festival attributes – facilities, personnel, food, wine, and entertainment – attendees consider important. Specifically, personnel and entertainment were considered basic/must-be factors, while wine was considered a performance/one-dimensional factor (Velikova et al., 2017). Harrington et al. (2017) classified Oktoberfest beer tent attributes into three delighters/attractive attributes (e.g., shopping, special souvenirs available in the tent), 11 satisfiers/one-dimensional attributes (e.g., staff friendliness and competence), and two dis-satisfiers/must-be attributes (e.g., party environment in the tent). Overall, some available empirical studies have shown the Kano approach is applicable in identifying event attributes that influence satisfaction.

2.4. Overview of research process

This study focuses on identifying event attributes, consequences/benefits, personal values, and satisfaction attributes by using MEC theory and the Kano model to achieve its purpose. We employ a sequential exploratory strategy, meaning we begin by collecting and analyzing qualitative data. We then analyze quantitative data used primarily to explore a phenomenon by applying qualitative findings to different samples and developing instrumentation for collecting quantitative data (Creswell & Creswell, 2018). We use this combined approach for Study 1 and Study 2. The research process is shown in Fig. 1 .

Fig. 1.

Research process.

3. Study 1: event attribute-consequence-value linkages

We explore relationships among event attributes, consequences/benefits, and personal values by assuming the current pandemic changes the importance of event attributes for people. In doing so, we answer the following questions: 1) What event attribute is more important than others and why?, 2) What do people obtain from experiencing event attributes and why?, and 3) What effects do benefits have on customers and how do they value them?. Because MEC theory has been suggested as a suitable theoretical lens for understanding how consumers think about a product/service in a hierarchical cognitive structure (Phillips & Reynolds, 2009; Reynolds & Gutman, 2001), we adopted an MEC analytical framework as the basis for answering these questions.

3.1. Methods and approach

Unlike previous studies on events (Chen, 2006; Fu et al., 2018; Yeh & Lin, 2017), we used a four-level means-end model, or a standard MEC (i.e., attributes-functional consequences-psychological consequences-values) (Olson & Reynolds, 2001). For the MEC analysis, we followed the three steps typical of the process: laddering interviews, content analysis, and construction of a hierarchical value map (HVM) based on an implication matrix (Reynolds & Gutman, 1988; Reynolds & Gutman, 2001). The laddering interview is a semi-structured qualitative method in which respondents answer questions in their own words. To build a means-ends ladder, consequences and values are gathered using the typical question: “Why is this important to you?” (Reynolds & Gutman, 1988). Responses are coded and analyzed, then frequencies of linkages are summarized in an implication matrix. Based on the implication matrix, we construct the HVM, an aggregate map of cognitive structure with abstraction (Reynolds & Gutman, 1988). Given the challenges of in-depth, one-on-one interviews during the pandemic, we employed a mixed-methods approach using the qualitative data and semi-structured online surveys. In the study's first phase, we gathered and analyzed text data to understand what specific attributes people talk about and to extract all concrete/abstract attributes. In addition to a literature review on event attributes, we sought to add new attributes to the list by analyzing several industry content sources with features regarding event planning under pandemic conditions.

In the first online survey, we asked participants to rate the importance of attributes to understand needs and preferences. Typically, an MEC is ranked according to attributes' relative importance (Phillips & Reynolds, 2009). In the second survey, we asked participants to choose the five most important attributes among those selected in the first survey. Participants also were asked to respond to questions on benefits and personal values in their own words so we could better understand why they perceive certain event attributes as relevant to their needs. To improve participants' understanding of the meanings of consequences and values, we provided several examples of both terms. Participants who had not engaged in such prior thinking would not have been able to answer otherwise (Reynolds & Gutman, 1988). Based on respondents' descriptions in the boxes, we first analyzed the content by classifying all responses into the three basic A-C-V levels then by organizing all responses under individual codes (Phillips & Reynolds, 2009). Among several possible event types, we chose state fairs for this research with pandemic considerations in mind. The state fair has a local legacy, a socioeconomic impact, and needs to be sustainable. Moreover, during the pandemic, “staycations” have become an attractive alternative for people who prefer a short trip within their state's borders. As a result, a community-based event like a state fair may be more attractive than out-of-state travel options.

3.2. Data collection

To collect rich data that can provide valuable insight into state fair experiences and identify event attributes experienced at a state fair, we gathered qualitative data from the TripAdvisor review site. TripAdvisor is the single source available that provides state fair attendees' comments and opinions within one platform. Using data from this site, four student assistants created an entire data set, including ratings, year posted, residency, and accompaniment type. A total of 3781 reviews of 12 state fairs hosted before the 2020 pandemic occurs comprises 100% of our sample.

In addition, we extracted text data from several industry sources, including the CDC (2020) and WHO (2020), that provided health and safety protocols, safety guidelines, and considerations for event planning and preparedness regarding COVID-19. These documents were published between March and June of 2020, concurrent with this research. We also asked a small group of event professionals to identify health/safety-related event attributes. Because studies on pandemic-related event attributes are rare in the literature, we also considered experts' opinions and practices regarding prevention and mitigation of health risks at events.

The data analysis process began by organizing all text data and reading it to make sense of the whole data set. During this process, we began to identify key attributes (Braun & Clarke, 2008). We adopted a mixed approach to analyze the data, initially applying a deductive approach. An initial coding started with attributes from the existing literature, from which we identified key attributes as initial coding categories of analysis. Subsequently, we used an inductive approach to extract attributes from the data, engaging in a repetitive process of sorting and grouping attributes into categories by revising the categories and formulating new ones. Due to redundancy in the major categories, we merged some similar categories to create broader, more comprehensive categories. We also eliminated some categories with too few attributes. We discussed and reached consensus through a deliberative process, finally deriving 121 attributes.

Because quantitative data collection should occur concurrently or sequentially in a mixed research approach (Tashakkori & Teddlie, 2003), we administered the first survey in early July 2020 to 110 students who were taking hospitality courses at a large public university, using the initially identified 112 attributes. We prepared the online questionnaire using the Qualtrics web-survey platform, posting the survey link on each course website with an announcement during a two-week period. Students were asked to rate the importance of attributes if attending a state fair using a seven-point Likert scale (1 = the least important, 7 = the most important). A total of 103 students participated in the survey. The results revealed that 48 attributes showed relative importance above the mean of 6.0. We then used these 48 attributes for the second-round survey, in which we invited students who had experienced a state fair within the last five years to participate. After posting this survey link on the course website over a two-week period in July 2020, we received 52 valid responses, later used for MEC analysis.

3.3. Results

3.3.1. Event attributes

The TripAdvisor review dataset on the 12 state fairs comprise a total of 3781 reviews and 318,534 words in aggregate, which we used for content analysis. All reviews were posted between January 2008 and February 2020, the most popular years for reviews being 2016 (17.9%), 2017 (15.3%), 2015 (14.9%), 2018 (14.5%), 2014 (10.1%), and 2019 (8.7%). Reviewers experienced state fairs with family (32.5%), as a couple (24.7%), with friends (16.4%), solo (4.1%), for a business purpose (2.6%), or with another form of accompaniment (19.4%). Reviewers were in-state (58.0%), out-of-state (30.7%), or neither (n/a) (10.9%). Using the rating scale (5 = excellent, 1 = terrible), a majority of reviewers reported a positive event experience. A rating of 5 (62.8%) was most popular, followed sequentially by 4 (25.21%), 3 (7.9%), 2(2.2%), and 1(1.6%). We also content analyzed a total of 24,668 words from industry sources, identifying 433 general attributes and 57 features/services specific to COVID-19. After accounting for relevance, redundancy, and omissions, we combined and reduced these identified attributes to 112, representing nine before-event attributes, 95 during-event attributes, and eight after-event attributes (refer to Appendix A).

In the first survey, respondents (n = 103) were 67.8% female and 32.2% male and primarily in upper grades with seniors (70.3%) being followed by juniors (21.4%), then sophomores (8.3%). As for ethnicity, respondents identified as Caucasian (32.8%), African-American (32.1%), Asian (24.7%), other (6.2%), or Hispanic/Latino (4.2%). The average age of participants was 22.56 (SD = 3.94). Most students (75.2%) previously had a state fair experience, with 88.7% of those respondents reporting satisfaction regarding overall fair performance. As shown in Table 1 , a total of 48 attributes gained a score over 6.0 out of 7.0. In the before- during-, and after-event categories, the highest-ranked attributes were event information (B1), handwashing options (D19), and safe controlled crowd (A2), respectively.

Table 1.

The 48 important attributes.

Attributes	M	Rank
Before the event
B1. Event information (e.g., time/date, location, fair schedule, ticketing information, parking, vendors, sponsors, contact information)	6.68 (0.70)	1
B2. Online ticketing system (e.g., fast and simple, confirmation via email or text, printed tickets and wristbands)	6.31 (0.85)	4
B3. Hygiene and sanitation information regarding COVID-19 (e.g., hygiene rules, guidelines, protocols, policies, procedures)	6.64 (0.79)	3
B4. Event safety information regarding COVID-19 (e.g., social distancing rules, face covering rules, indoor event venue capacity, crowd/traffic/attendee flow management plans, food handling)	6.65 (0.76)	2
B5. Contact information of local health authorities and/or urgent care/ER nearby	6.09 (1.26)	5
During the event
D1. Location (e.g., easy to find, not far drive, safe neighborhood, public transportation options)	6.32 (1.02)	15
D2. Parking (e.g., easy to find parking spot, sufficient parking, reasonable price, effective parking attendant)	6.33 (0.95)	14
D3. Admission fee (e.g., not pricey, price for single/group, daily specials)	6.24 (0.96)	19
D4. On-site ticket sales (e.g., easy ticket purchase/procedure system, short lines, multiple booths, self-service ticketing kiosks)	6.00 (1.09)	33
D5. Effective access control system (e.g., less lineage and queuing, reduced waiting time, passive scanning, easy and quick check-in)	6.08 (0.89)	30
D6. Security screening (e.g., metal detector)	6.13 (1.06)	27
D7. Health and safety services (e.g., distribution of face masks and hand sanitizers, decals on floors and seats)	6.44 (0.91)	10
D8. Information desk (e.g., maps, flyer of events)	6.11 (1.06)	29
D9. Way finding (e.g., easy to navigate)	6.28 (0.74)	17
D10. Security personnel (e.g., friendly security)	6.18 (1.13)	22
D11. Uniqueness (e.g., unique programs reflected by local culture, locally themed event)	6.11 (0.91)	29
D12. Entertainment (e.g., concerts, parades, carnivals, performing arts, dance performances, musical acts)	6.00 (0.94)	33
D13. Amusement (e.g., ride variety, overhead condoler ride, skylift)	6.16 (1.13)	21
D14. Food (e.g., variety of food selection, local food, drink, good taste)	6.49 (0.91)	7
D15. Healthy food options (e.g., meeting all dietary restrictions, smart calories)	6.01 (1.20)	32
D16. Food prices (e.g., reasonable, too high, too low)	6.50 (0.80)	6
D17. Kiosks (e.g., water, beer, wine, alcohol)	6.17 (0.97)	26
D18. Restrooms (e.g., availability, cleanliness, family option)	6.48 (0.95)	8
D19. Handwashing options (e.g., facilities with soap and water and/or hand rub dispensers)	6.67 (0.78)	1
D20. Atmosphere/ambiance (e.g., high, expected, low, needs work)	6.04 (0.93)	31
D21. Fair grounds (cleanliness, space allotment, staffing)	6.37 (0.95)	12
D22. Rest areas (e.g., lots of shade, shelter, cushioned seating, air conditioning)	6.31 (0.86)	16
D23. Design (e.g., landscape, street layout, midway, excellent arrangement)	6.20 (0.84)	20
D24. Crowd control (e.g., well managed, needed additional, too much)	6.01 (1.06)	32
D25. Signage (e.g., guidance, effectiveness, directional)	6.00 (1.35)	33
D26. Organization (e.g., smooth flow of the events, organized, not organized)	6.35 (0.77)	13
D27. Event implementation (e.g., event production, stage management)	6.12 (0.89)	28
D28. Event staff (e.g., availability, friendliness and helpful)	6.56 (0.71)	2
D29. Security (e.g., sheriff's/police presence, cameras, gate checks)	6.45 (0.94)	9
D30. First aid, EMT (e.g., onsite, able to locate, availability)	6.18 (1.11)	21
D31. Water or hydration stations (e.g., free or low cost, availability, cleanliness)	6.54 (0.73)	3
D32. Facility maintenance (e.g., event venue/building/ground, clean atmosphere)	6.26 (1.10)	18
D33. Cleaning and disinfection of the venue (e.g., amount, thorough)	6.52 (0.85)	4
D34. Physical distancing (e.g., regulating the flow of the crowd, minimal physical interaction with people)	6.34 (0.93)	13
D35. Daily event information (e.g., attendee size, news/stories, special event info., COVID-19 updates)	6.00 (1.40)	33
D36. Waste management (e.g., trash can and recycling stations, regular cleaning)	6.43 (0.91)	11
D37. Safety (e.g., clean, spacious, well-lite, pathways clear, exit signs)	6.50 (1.16)	5
After the event
A1. Multiple exits	6.35 (1.19)	4
A2. Safe controlled crowd	6.55 (0.78)	1
A3. Security personnel everywhere	6.03 (1.24)	5
A4. Well-lit parking lot	6.54 (0.84)	2
A5. Shuttle service to public transportation	6.09 (1.08)	6
A6. Updated information of the fair regarding COVID-19	6.40 (1.079)	3

3.3.2. Implication matrix (IM) and hierarchical value map (HVM)

In the subsequent survey, a total of 52 students participated (71.1% female, 28.9% male, average age 23.42, SD = 3.75). The sample size is acceptable for MEC framework application with the laddering technique, which suggests at least 20 respondents (Reynolds & Gutman, 2001). A majority of students (68.8%) reported experiencing a state fair within the past three years. The data were content analyzed to generate associated categories for elements in each A-C-V category. We conducted the subsequent coding process by classifying each response into four levels (i.e., attributes, functional consequences, psychological consequences, and values). Through the repetitive process of content analysis, the two independent coders aggregated all direct (e.g., A-C-V) or indirect (e.g. A-V) links based on respondent-provided content. The laddering process generated a chain with one or more links. Additional or missing linkages regarding coding categories were discussed and adjusted. Two coders performed the final coding of the ladders, with inter-coder reliability showing 72.1% of the Cohen's kappa values exceeded 0.61, placing it in the substantial agreement range (Cohen, 1960). Finally, the coding resulted in a total of 191 ladder linkages with 20 attributes, 13 consequences, and 6 values, for an average of 2.72 ladders per respondent.

Among the 48 attributes in Table 1, the 20 that more than half of the respondents considered as important were as follows: event information (B1), hygiene and sanitation information (B3), event safety information (B4), parking (D2), admission fee (D3), health and safety services (D7), uniqueness (D11), entertainment (D12), amusement (D13), food (D14), food prices (D16), restrooms (D18), handwashing options (D19), atmosphere/ambiance (D20), design (D23), event staff (D28), security (D29), cleaning and disinfection of the venue (D33), safety (D37), and safe controlled crowd (A2). The 13 consequences were: learning (C1), prevention (C2), feeling safe (C3), care/protection (C4), convenience (C5), decreased stress (C6), good service (C7), enjoyment (C8), revisit potential (C9), experience of culture (C10), socialization (C11), mood enhancement (C12), and feeling secure (C13). The six personal values were: trust (V1), sense of safety (V2), comfortable (V3), pleasure and excitement (V4), worthy to spend time (V5), and sense of security (V6).

Using these MECs, we built an implication matrix based on the number of direct associations. In Table 2 , the element code indicates the order the elements appear in the text; this is irrelevant to the number of times the element is mentioned. The higher the frequency, the closer the relationship between the two elements. As seen in the matrix, “pleasure and excitement” is the value with the most elements leading to it, thus it is seen as the core value in terms of importance to state fairs, followed by “sense of safety.” “Pleasure and excitement” is connected with one psychological consequence (“enjoyment”) and three functional consequences (“good service,” “socialization,” and “experience of culture”), the strongest association existing with “enjoyment,” which itself is connected with the “entertainment” attribute. The second core value, “sense of safety,” shows strong connections with one psychological consequence (“feeling safe”) and two functional consequences (“prevention,” “care/protection”). Additionally, the results imply that a single consequence could serve multiple ends. For instance, the “prevention” consequence led to different personal values (“trust” and “sense of safety”). Likewise, a certain attribute led to multiple consequences in the MECs. For example, “parking” led to two different sets of means-end relationships (“parking” ➔ “convenience” or “decreased stress” ➔ “comfortable”).

Table 2.

Implication matrix.

	C1	C2	C3	C4	C5	C6	C7	C8	C9	C10	C11	C12	C13	V1	V2	V3	V4	V5	V6	Out
B1	5																			5
B3		9	2																	11
B4	1	2	14	3									1							21
D2					9	9	2		1											21
D3									5											5
D7		1	1	5			1													8
D11								1	2	11										14
D12						2		38	3		1									44
D13							1	3	1		9									14
D14				1				16	2	3										22
D16							2	1	7											10
D18					2	5	1													8
D19	1	3	9	1																14
D20								2	1		1	9								13
D23						1	3	8												12
D28							5	1												6
D29		1	3	1						1			10							16
D33		2	1	5			1													9
D37	1		17	1						1										20
A2		1	2	2		1							11							17
C1														6						6
C2	1		1	1										7	7				1	18
C3		3		1										3	29	3	1			40
C4	1	2	2											2	18				3	28
C5						1	2									5				8
C6					1			2							1	7	2	3		16
C7																	6			6
C8			1			2		3	1							3	37	3		50
C9							1										2	15		18
C10	8							3									18	4		33
C11						1		2									10			13
C12						2	1										2	7		12
C13		1	3	1										1	2				14	22
V1															2				1	3
V2																	2		3	5
V3																				0
V4															1	2		1		4
V5																	1			1
V6															3					3
In	18	25	56	22	12	24	20	80	23	16	11	9	22	19	63	20	81	33	22	576

Based on the 134 non-zero active cells in the implication matrix, we constructed the HVM for each of the identified categories by connecting all elements and selecting the linkages (Fig. 2 ). This reflects the salient MECs in a diagram (Reynolds and Gutman, 1988). Because researchers can decide whether to consider only direct connections between concepts or both direct and indirect connections (Valette-Florence & Rapacchi, 1991), we consider only direct connections in the HVM (i.e., A-C-V sequence). To avoid complexity of interpretability, we chose 36 linkages within the HVM that covered 18.8% of all direct linkages after applying the cutoff level of 5, common with a sample size of 50–60 (Reynolds & Gutman, 1988). In this study, the HVM gives an overview on the most salient hierarchical relations.

Fig. 2.

Hierarchical value map (HVM).

3.4. Discussion of Study 1

With the four levels of the MEC framework, we confirm the A-C-V sequences. This demonstrates that an individual event attribute results in consequences that link to a personal value. “Pleasure and excitement” emerged as the most prominent core value, followed by “sense of safety,” “worthy to spend time,” “sense of security,” comfortable,” and “trust.” The top two core values were similar to personal values found in a tourism, leisure, and event context such as “pleasure and enjoyment” and “security” (Hede et al., 2005; Ho, Lin, & Huang, 2014; Jiang et al., 2015; Klenosky, 2002; López-Mosquera & Sánchez, 2011). The three emerging personal values - “worthy to spend time,” “comfortable,” and “trust” - have received scant attention in prior studies, particularly in MEC studies on events (Fu et al., 2018; Yeh & Lin, 2017).

Study 1 produced interesting results. For example, we found that price (i.e., admission fee and food price) is closely connected to the consequence of revisitation potential, which leads to the “worthy to spend time” value. This causal relationship needs to be empirically tested in future works to validate this finding. Another interesting finding is that pre-event COVID-19-related health/safety information is more important than relevant services during the event, suggesting proper pre-event communication may reduce concerns and increase trust in events. In summary, our findings enable us to understand how potential event attendees link event attributes with particular consequences/benefits, as well as how these consequences can satisfy their personal values. Testing of causality to validate the MEC relationships discussed in Study 1 remains for subsequent research. Empirical evidence is necessary to better understand potential event attendees' pandemic-related attribute needs, particularly how these relevant event products/services influence their benefits and personal values.

4. Study 2: Identification of satisfaction attributes

In Study 2, we incorporated the 48 event attributes into the Kano model in order to identify what attributes have the potential to drive customer satisfaction during a pandemic. Study 2 will answer the following questions: 1) What will be customers' response regarding event attributes that the organizer offers with the presence/fulfillment of attributes?, 2) What will be customers' response regarding event attributes the organizer offers but are absent/unfulfilled?, and 3) What event attribute can increase satisfaction?

4.1. Methods and analysis approach

In an event context, we applied the Kano model, which considers both customer assessments of an attribute's functional characteristics and their assessments of the attribute's dysfunctional characteristics. One question was positively worded: “How would you feel if each attribute were present or fulfilled by a state fair?” (functional). The second question was negatively worded: “How would you feel if each attribute were not present or fulfilled by a state fair?” (dysfunctional). As shown in Table 3 , participants had five possible responses: “Satisfied,” “It must be that way,” “I am indifferent/neutral,” “I can live with it that way,” and “Dissatisfied.” To determine the Kano's classification of attributes, we used an evaluation table and calculated the frequency of responses to the paired functional and dysfunctional questions. Based on these assessments, we classified event attributes into five categories that meet different kinds of customer requirements, thus influencing customer satisfaction differently: must-be attributes (M), one-dimensional attributes (O), attractive attributes (A), indifferent attributes (I), and reverse attributes (R).

Table 3.

Kano classification table.

Customer requirements	Dysfunctional
Functional	1. Satisfied	2. Must be	3. Neural	4. Live with	5. Dissatisfied
1. Satisfied	Q	A	A	A	O
2. Must be	R	I	I	I	M
3. Neutral	R	I	I	I	M
4. Live with	R	I	I	I	M
5. Dissatisfied	R	R	R	R	Q

A = Attractive, I = Indifferent, M = Must-Be, O = One-Dimensional, Q = Questionable results, R = Reverse.

Next, to better classify individual attributes in each category, we used the customer satisfaction (CS) coefficient and customer dissatisfaction (CD) coefficient, each of which is based on the calculation of indices according to Berger et al. (1993). Below are the formulas related to the computation of the average impacts of satisfaction and dissatisfaction, respectively:

Extent of satisfaction:

$$\mathrm{CS}={\sum }_{i=1}^n\frac{\left(\text{Attractive},\mathrm{One}\text{dimensional}\right)}{\text{Attractive},\mathrm{One}\text{dimensional},\text{Must}\mathrm{be},\text{Indifferent}}$$

Extent of dissatisfaction:

$$\mathrm{CD}={\sum }_{i=1}^n\frac{\left(\mathrm{One}\text{dimensional},\text{Must}\mathrm{be}\right)\ast -1}{\text{Attractive},\mathrm{One}\text{dimensional},\text{Must}\mathrm{be},\text{Indifferent}}$$

4.2. Data collection and participants

Using Qualtrics, we created the online questionnaire comprising functional and dysfunctional questions for each attribute. Pretesting was conducted with several student assistants to evaluate clarity of instructions and identify question defects. Participants for the main survey were recruited through Amazon's Mechanical Turk (MTurk) crowdsourcing marketplace in September 2020 and invited to participate voluntarily. After reading a detailed description of the study and clicking a university-approved consent form, participants were asked to complete the 20-min survey. To strengthen data validity, individuals were eligible to participate if they were aged 18 years and over and had attended a state fair.

A total of 346 participants completed the survey for a nominal payment. As shown in Table 4 , the majority of respondents were males (59.0%) employed at a company (59.0%). A plurality of respondents were in their 30s (41.6%) with university degrees (38.4%), earning $50,000 to $74,999 in annual household income (34.1%). These respondents were most likely to have visited a state fair with family/relatives (44.5%) and were in-state residents at the time of the visit (90.2%). Most respondents (79.8%) reported having visited the state fair regularly before the pandemic, with most visits (69.7%) occurring between 2017 and 2019. The majority of respondents (80.9%) reported overall satisfaction with the fair experience.

Table 4.

Profile of participants.

	Categories	n	%		Categories	n	%
Gender	Male	204	59.0	Age	20–29	79	22.8
	Female	141	40.7		30–39	144	41.6
	No response	1	0.3		40–49	73	21.1
Education	High school	22	6.4		50–59	33	9.5
	Community college	71	20.5		Above 60	17	5.0
	Vocational training	8	2.3	Purpose	Leisure	190	54.9
	University	133	38.4		Business	154	44.5
	Masters	108	31.2		No response	2	0.6
	Doctorate/PhD	3	0.9	Companion	Family/Relatives	154	44.5
	Other	1	0.3		Friends/Partner	109	31.5
Occupation	Company employee	204	59.0		Social group	41	11.8
	Owner of a business	19	5.5		Colleagues	32	9.2
	Professional	77	22.3		By myself	7	2.0
	Public official	13	3.8		Other	2	0.6
	Freelance	17	4.9	Status	In-state resident	312	90.2
	Student	3	0.9		Out-of-town tourists	30	8.7
	Housewife/Husband	2	0.6		No response	4	1.2
	Retired	2	0.6	Recent visit	Last year	114	32.9
	Unemployed	9	2.6		2–3 years ago	127	36.7
Income	Under $35,000	45	13.0		4–5 years ago	65	18.8
	$35,000 to $49,999	62	17.9		6+ years ago	24
	$50,000 to $74,999	118	34.1		Can't remember	16	4.6
	$75,000 to $99,999	70	20.2	Residency	New England	16	4.6
	$100,000 to $149,999	34	9.8		Middle Atlantic	48	13.9
	$150,000 to $199,999	12	3.5		East North Central	36	10.4
	$200,000 and above	4	1.2		West North Central	19	5.5
	No response	1	0.3		South Atlantic	85	24.6
Regular visit	Yes	276	79.8		East South Central	18	5.2
	No	70	19.9		West South Central	43	12.4
Overall event experience	Satisfied	280	80.9		Mountain	23	6.6
	Dissatisfied	35	10.1		Pacific	55	15.9
	Neutral	31	9.0		No response	3	0.9

n = 346.

4.3. Results

Categorization of state fair event attributes, satisfaction and dissatisfaction coefficients, and their sums are shown in Table 5 . Must-be attributes (M) consisted of 16 items, including event information (B1), hygiene and sanitation information regarding COVID-19 (B3), event safety information regarding COVID-19 (B4), parking (D2), and entertainment (D12). One-dimensional attributes (O) consisted of 10 items, including online ticketing system (B2), admission fee (D3), food (D14), and food prices (D16). Attractive attributes (A) consisted of 9 items (A), including amusement (D13), uniqueness (D11), healthy food options (D15), and shuttle service (A4). Indifferent attributes (I) consisted of 13 items, including contact information of local health authorities (B5), location (D1), security screening (D6), and information desk (D8).

Table 5.

Kano model analysis results.

Attribute	M	O	A	I	R	Q	Kano	CS	CD	Sum
Before the event
B1. Event information	48.84	15.03	22.25	9.83	2.31	1.73	M	0.39	−0.67	−0.28
B2. Online ticketing system	26.59	32.37	23.99	13.01	2.60	1.45	O	0.59	−0.61	−0.03
B3. Hygiene/sanitation information	34.68	9.83	23.99	28.32	2.02	1.16	M	0.35	−0.46	−0.11
B4. Event safety information	38.73	8.38	17.63	32.37	2.31	0.58	M	0.27	−0.49	−0.22
B5. Contact information	25.43	7.80	16.18	45.95	2.89	1.73	I	0.25	−0.35	−0.10
During the event
D1. Location	31.21	8.09	13.58	42.20	3.47	1.45	I	0.23	−0.41	−0.19
D2. Parking	47.11	21.39	12.14	15.61	2.60	1.16	M	0.35	−0.71	−0.36
D3. Admission fee	29.77	39.60	15.32	11.27	3.18	0.87	O	0.57	−0.72	−0.15
D4. On-site ticket sales	29.19	39.88	11.27	15.90	2.02	1.73	O	0.53	−0.72	−0.19
D5. Access control system	28.03	37.86	12.72	16.18	3.76	1.45	O	0.53	−0.70	−0.16
D6. Security screening	12.14	12.43	17.92	53.47	2.89	1.16	I	0.32	−0.26	0.06
D7. Health/safety services	41.33	8.09	15.61	32.66	1.73	0.58	M	0.24	−0.42	−0.17
D8. Information desk	26.01	10.40	21.97	36.42	3.47	1.73	I	0.34	−0.38	−0.04
D9. Way finding	18.79	8.09	10.69	58.09	3.18	1.16	I	0.20	−0.28	−0.08
D10. Security personnel	36.42	9.54	19.94	31.21	2.31	0.58	M	0.30	−0.47	−0.17
D11. Uniqueness	20.23	25.14	45.38	8.09	0.87	0.29	A	0.71	−0.46	0.25
D12. Entertainment	31.79	28.03	25.14	13.01	0.58	1.45	M	0.54	−0.61	−0.07
D13. Amusement	16.18	22.54	47.40	11.27	1.73	0.87	A	0.72	−0.40	0.32
D14. Food	13.87	47.98	18.21	16.47	3.18	0.29	O	0.69	−0.64	0.04
D15. Healthy food options	9.54	11.56	52.02	19.08	6.65	1.16	A	0.69	−0.23	0.46
D16. Food prices	21.68	46.24	17.63	9.25	4.34	0.87	O	0.67	−0.72	−0.04
D17. Kiosks	16.47	9.25	52.60	15.90	4.34	1.45	A	0.66	−0.27	0.38
D18. Restrooms	58.09	9.25	10.40	16.47	4.91	0.87	M	0.21	−0.71	−0.51
D19. Handwashing options	47.40	7.23	12.43	26.59	5.20	1.16	M	0.21	−0.58	−0.37
D20. Atmosphere/ambiance	19.65	23.70	40.46	13.01	2.31	0.87	A	0.66	−0.45	0.21
D21. Fair grounds	29.77	12.14	19.36	34.10	3.18	1.45	I	0.33	−0.44	−0.11
D22. Rest areas	15.90	8.09	50.29	20.52	2.89	2.31	A	0.62	−0.25	0.36
D23. Design	28.32	8.38	32.37	27.75	2.31	0.87	A	0.42	−0.37	0.05
D24. Signage	20.81	45.95	8.96	21.68	2.02	0.58	O	0.56	−0.68	−0.12
D25. Crowd control	26.88	8.67	17.05	41.04	3.76	2.60	I	0.27	−0.38	−0.10
D26. Organization	22.54	39.02	21.39	12.72	2.89	1.45	O	0.63	−0.64	−0.01
D27. Event implementation	35.26	36.13	10.40	16.47	1.45	0.29	O	0.47	−0.73	−0.25
D28. Event staff	48.55	11.56	29.48	8.09	1.73	0.58	M	0.42	−0.62	−0.20
D29. Security	41.62	11.85	15.03	27.17	2.89	1.45	M	0.28	−0.56	−0.28
D30. First aid/EMT	30.92	8.09	19.08	38.73	2.60	0.58	I	0.28	−0.40	−0.12
D31. Hydration stations	28.32	11.27	34.68	21.68	3.18	0.87	A	0.48	−0.41	0.07
D32. Facility maintenance	29.77	8.09	21.39	35.26	4.05	1.45	I	0.31	−0.40	−0.09
D33. Cleaning/disinfection	35.55	4.62	24.57	30.92	3.18	1.16	M	0.31	−0.42	−0.11
D34. Physical distancing	30.64	7.80	18.50	39.60	2.60	0.87	I	0.27	−0.40	−0.13
D35. Daily event information	28.03	8.96	17.63	43.06	2.02	0.29	I	0.27	−0.38	−0.11
D36. Waste management	26.59	44.80	13.58	9.54	3.76	1.73	O	0.62	−0.76	−0.14
D37. Safety	58.67	7.23	19.08	12.14	2.31	0.58	M	0.27	−0.68	−0.41
After the event
A1. Multiple exits	8.67	14.16	27.17	45.66	3.47	0.87	I	0.43	−0.24	0.19
A2. Safe controlled crowd	34.39	8.09	23.41	29.19	3.76	1.16	M	0.33	−0.45	−0.12
A3. Security personnel	38.73	9.54	16.18	30.06	4.05	1.45	M	0.27	−0.51	−0.24
A4. Well-lit parking lot	36.43	11.27	22.54	26.88	2.60	0.29	M	0.34	−0.49	−0.14
A5. Shuttle service	13.01	9.25	52.60	20.52	2.89	1.73	A	0.64	−0.23	0.41
A6. Updated information	29.19	3.18	23.12	43.06	0.87	0.58	I	0.27	−0.33	−0.06

Proportions of respondents who fall into each category are shown in the table: M = Must-be, O=One-dimensional, A = Attractive, I=Indifferent, R = Reverse, Q = Questionable results.

Next, we calculated customer satisfaction and dissatisfaction coefficients (i.e., CS and CD). A satisfaction coefficient close to 1 indicates the attribute is highly likely to impact customer satisfaction, whereas a dissatisfaction coefficient approaching −1 indicates dissatisfaction will increase if the attribute is not fulfilled (Berger et al., 1993). Results show that event attributes with high satisfaction coefficients — including amusement (D13), uniqueness (D11), food (D14), healthy food options (D15), and food prices (D16) — can greatly increase satisfaction. By contrast, event attributes with high dissatisfaction coefficients — including waste management (D13), event implementation (D27), admission fee (D3), on-site ticket sales (D4), parking (D2), and restrooms (D18) — can strongly influence dissatisfaction. The sum of the coefficients implies indifferent attributes with positive-sum coefficients such as security screening (D6) and multiple exits (A1) are closer to attractive attributes. Other indifferent attributes with negative-sum coefficients — including contact information of local health authorities (B5), location (D1), information desk (D8), and way finding (D9) — are closer to must-be attributes.

Fig. 3 shows the Kano classification in the satisfaction and dissatisfaction indexes plot. This plot sheds light on what attributes are associated with potential satisfaction in the case of being present/fulfilled, as well as potential dissatisfaction in the case of being absent/unfulfilled. As seen in the figure, entertainment (D12) in the must-be attribute quadrant can be considered a one-dimensional attribute, the fulfillment of which helps increase satisfaction. On the other hand, the one-dimensional attribute, event implementation (D27), can be interpreted as a must-be attribute whose non-fulfillment can increase attendee dissatisfaction.

Fig. 3.

Plot of satisfaction and dissatisfaction indexes.

4.4. Discussions of study 2

According to the results of the Kano categories, most COVID-19-related attributes are must-be attributes, along with parking, entertainment, event staff, and security personnel. People take these attributes for granted; thus, state fair attendees will be very dissatisfied if they are unfulfilled. With one-dimensional attributes such as admission fee, food price, food, and online/onsite ticketing systems, the higher/lower the level of fulfillment, the higher/lower the customer's satisfaction, respectively. Attractive attributes such as amusement, uniqueness, atmosphere, and kiosks have the most influence on customer satisfaction. Meanwhile, attendees may not have much interest in indifferent attributes such as facility maintenance, fair grounds, and way finding.

Several findings are somewhat consistent with previous studies showing that staff and entertainment are must-be attributes and facilities are indifferent attributes (Velikova et al., 2017), while food and cultural attributes contribute as satisfiers (Harrington et al., 2017). Inconsistent findings also exist. Some must-be attributes (e.g., admission fee) and one-dimensional attributes (e.g., restrooms) in our study were classified as indifferent attributes in a previous study (Velikova et al., 2017). While atmosphere and design are attractive attributes that greatly influenced satisfaction in our study, event environment was a dissatisfier/must-be attribute in prior work (Harrington et al., 2017). In other respects, our findings support previous studies that showed satisfaction influenced by attributes such as programs (Lee et al., 2009; Lee et al., 2011; Tanford & Jung, 2017) and organization (Yuan & Jang, 2008). Overall, our results suggest that state fair organizers should pay attention to present/fulfill one-dimensional attributes and attractive attributes that drive increased influence satisfaction, allocating more resources for them during decision making. We further suggest that event organizers pay attention to must-be attributes and indifferent attributes that are close to one-dimensional and attractive attributes, which can help increase attendee satisfaction.

5. General discussion and conclusions

The COVID-19 pandemic has brought renewed attention to health and safety concerns at events where social gatherings take place. This timely research was motived by the necessity of examining the product/service-consumer relationship in an event context because of potential changes in customers' needs and preferences. Employing a sequential exploratory strategy of the mixed-method approach with an integration of MEC theory and the Kano model, we sought to identify opportunities to improve existing event products/services. A key implication of Study 1 is that hedonic value (i.e., pleasure and excitement) and self-protective value (i.e., sense of security) are core attendee values. In particular, event attendees may desire to achieve hedonic value more than other values, regardless of pandemic conditions. Findings in Study 2 imply that hedonic attributes either can greatly increase satisfaction (e.g., amusement) or greatly decrease it (e.g., entertainment). Must-be health/safety attributes (e.g., health/safety services), meanwhile, do not increase satisfaction, but they do increase dissatisfaction if not presented/fulfilled. To increase event attendees' satisfaction, delivering high-performance, attractive attributes and one-dimensional attributes is a strategic way to induce a pleasurable customer experience during and after a pandemic.

Our results include some other important pandemic-specific implications. For example, relevant important attributes yield personal benefits of prevention, care/protection, and trust in events, satisfying the personal value of sense of safety. Besides, people tend to give more importance to health/safety attributes as basic requirements, thus these attributes can cause great dissatisfaction if unfulfilled. Providing relevant information/services before the event starts would be more effective in decreasing dissatisfaction than implementing services during the event.

5.1. Theoretical and practical implications

This study extends existing knowledge of event attributes by using an MEC-Kano theoretical framework to understand customer-preferred attributes in terms of benefits, values, and satisfaction. This study can bridge a gap in tourism and hospitality literature, where very few empirical studies integrating the MEC and Kano model exist (Lin et al., 2012; Lin et al., 2018; Yeh & Lin, 2017). In the context of event products/services, this remains a relatively unexplored research area (Pandey, Sahu, & Joshi, 2020). We expand the research lens to reveal previously unknown or unobserved event attributes, especially pandemic-related ones. By using sequential qualitative-quantitative design research, this study extends the line of event attribute research and demonstrates their direct linkages to satisfaction using a qualitative approach (Tanford & Jung, 2017) and quantitative approach (Akhoondnejad, 2016; Crotts et al., 2008; Harrington et al., 2017; Lee et al., 2008; Robinson & Clifford, 2012; Song et al., 2014; Velikova et al., 2017; Yuan & Jang, 2008).

This study also contributes to an increased understanding of how an individual event attribute links directly or indirectly to personal benefits and personal values (Fu et al., 2018; Hede et al., 2005). Unlike previous studies, we identified event attributes by analyzing text data and gleaned satisfaction attributes from Kano's functional requirements perspective. By using a four-level means-end model with a laddering technique, we contributed to prior work that has examined personal values in a three-level A-C-V sequence (Chen, 2006; Yeh & Lin, 2017). Because epidemic/pandemic-related event services were not well documented in the literature, the results of this study can be used to characterize future events during/after an epidemic or pandemic. Additionally, this study can serve as a foundation for future, larger studies that could produce empirical evidence. One possible extension of this work would involve incorporating perceived risks regarding epidemic/pandemic disease and behavioral intentions to be able to characterize segments by personal values and Kano attributes in a staycation or event tourism context. The effect of hospitality consumers' adoptive belief regarding COVID-19 on their positive/negative emotion is not just direct but is also moderated by perceived health risks (Foroudi, Tabaghdehi, & Marvi, 2021). In that sense, perceived health risks may play a key role in lowering or enhancing satisfaction among attendees who experience health/safety-related event attributes.

In addition to the practical implications discussed in studies 1 and 2, our overall findings offer additional marketing and managerial implications for event managers in their recovery preparedness efforts. Event mangers can benefit using the MEC and Kano approaches in examining existent and non-existent event attributes to expand or adjust their pandemic-related products/services. Managers should revisit their event products/services and prioritize attributes by importance and by what leads to achieving personal values and improving potential satisfaction.

From a product/service perspective, event attribute-based marketing may help increase attendee satisfaction. From a customer perspective, by contrast, value-based marketing may motivate their participation. Events are a phenomenon of human culture and have various impacts, including personal health and well-being (Getz, 2010; Getz & Page, 2016). Because people increasingly value both, leisure activity providers have a significant business growth opportunity (Pyke, Hartwell, Blake, & Hemingway, 2016). At the same time, providing appropriate programs/services is essential to creating a pleasurable and safe event environment. When customers perceive crowding, they are likely to prefer safety-oriented products (Maeng, Tanner, & Soman, 2013). Thus, delivering a variety of health/safety-relevant programming in the form of entertainment may be a strategic way to stand out from competing events.

5.2. Limitations and future directions

This study has several limitations that should be addressed. First, there may be various reasons causing attendees' attribute preferences regarding importance, including predispositions toward certain physical features or health. Future work needs to investigate whether associations exist between these characteristics and attendees' opinions regarding attributes, consequences, and values. Second, there is a need for some caution in assuming that means-end models are organized in a hierarchical A-C-V sequence. A future study could apply various forms of analysis, including MEC network analysis, to delineate asymmetric relationships among event attributes, consequences, and personal values. Regarding the Kano model, other asymmetric methods also could provide more specific information for the classification of event product/service attributes, depending on their asymmetric influences on customer satisfaction. Third, our quantitative data do not rely upon on-site event experience, meaning findings are not free from hypothetical bias. Also, this study focused on one event type (i.e., state fair) and had mostly young respondents; our methods should be applied in a variety of event contexts with diverse groups. Further, perceived health/safety risks of an event in a time of pandemic may be different for different age groups and health statuses. Future empirical research can investigate the relationship between event attributes and perceived risk among attendees with varying profiles. Finally, due to the scope of this research, we did not attempt to segment event markets based on results. While an integrated MEC-Kano model approach enables us to understand potential event attendees' state fair-related needs, segmentation is necessary in subsequent research to provide an effective targeting strategy in the new, post-pandemic normal.

Biographies

Soyoung Boo is an Assistant Professor in the Cecil B. Day School of Hospitality Administration of J. Mack Robinson College of Business at Georgia State University. Her research topics include event management, data mining, online distribution channels, business tourism, and destination marketing. Her publications appear in internationally recognized peer-reviewed journals including Tourism Management, Journal of Travel Research, Cornell Hospitality Quality, Journal of Tourism and Hospitality Research, and Journal of Business Ethics.

Tony Kim is an Assistant Professor at Hart School of Hospitality, Sport, and Recreation Management, James Madison University. His research interests include food safety/sanitation, hotel & restaurant cleanliness, event management and consumer behavior in restaurants.

Appendix A. Appendix

Table 1.

Descriptive statistics of 112 event attributes.

Event Attributes	M	SD
Before the event
1. General event information (e.g., time/date, location, fair schedule, ticketing information, parking, vendors, sponsors, contact information)	6.68	0.70
2. Online ticketing system (e.g., fast and simple, confirmation via email or text, printed tickets, and wristbands)	6.31	0.85
3. Discounts or promo code (e.g., coupons for reduced admission ticket price, special day price	5.80	1.13
4. Mobile event apps (e.g., ticketing, e-brochure, map view, GPS, virtual platforms, and gamification)	5.82	1.20
5. Social media connections (e.g., connecting with event staffs/vendors/sponsors/other attendee)	5.33	1.36
6. Event ride-sharing apps (e.g., Uber, Lyft)	5.14	1.78
7. Hygiene and sanitation information regarding COVID-19 (e.g., hygiene rules, guidelines, protocols, policies, procedures)	6.64	0.79
8. Event safety information regarding COVID-19 (e.g., social distancing rules, face-covering rules, indoor event venue capacity, crowd/traffic/attendee flow management plans, food handling)	6.65	0.76
9. Contact information of local health authorities and/or urgent care/ER nearby	6.09	1.26
During the event
1. Location (e.g., easy to find, not far drive, safe neighborhood, public transportation options)	6.32	1.02
2. Parking (e.g., easy to find parking spot, sufficient parking, reasonable price, effective parking attendant)	6.33	0.95
3. Shuttle service (e.g., to and from the parking lot and the event gate, to and from public transportation and fair site)	5.49	1.13
4. Accessible drop off/pick up area	5.71	1.20
5. Reasonable admission fee (e.g., not pricey, price for single/group, daily specials)	6.24	0.96
6. Onsite ticket sales (e.g., easy ticket purchase/procedure system, short lines, multiple booths, self-service ticketing kiosks)	6.00	1.09
7. Multiple entrances	5.67	1.31
8. Effective access control system (e.g., less lineage and queuing, reduced waiting time, passive scanning, easy and quick check-in)	6.08	0.89
9. Security screening (e.g., metal detector)	6.13	1.06
10. Health and safety services (e.g., distribution of face masks and hand sanitizers, decals on floors and seats)	6.44	0.91
11. Information desk (e.g., maps, flyer of events)	6.11	1.06
12. Lost & Found	5.67	1.42
13. Mobility rental service (e.g., stroller, scooter, and wheelchair rentals)	5.10	1.68
14. Free Wi-Fi	5.04	1.64
15. Good crowd control	5.98	1.07
16. Wayfinding (e.g., easy to navigate).	6.28	0.74
17. Security personnel (e.g., friendly security)	6.18	1.13
18. Community booths (e.g., nonprofit organizations)	5.02	1.28
19. Vendors (e.g., many vendors, various merchandize, reasonable price, friendly vendors, sell real products, artisan products)	5.76	1.25
20. Entertaining events (e.g., concerts, parades, carnivals, performing arts, dance performances, musical acts)	6.00	0.94
21. Lots of free entertainment stages	5.49	1.12
22. Cultural events (e.g., cultural stages, interesting cultural activities)	5.73	0.97
23. Educational events (e.g., fair history booth, historical information, demonstration, agricultural education, cooking demonstration, and gardening)	5.14	1.33
24. Participating activities/programs (e.g., sports events, talent competition, wood carving, amateur performing competition, bungee jumping)	5.47	1.20
25. Family-orientated programs (e.g., fun programs for all ages, interactive areas)	5.39	1.27
26. Agricultural programs (e.g., agricultural show - 4H, crop displays, trailers and wheelers exhibit, antique farm equipment exhibit)	4.73	1.46
27. Learning/training programs (e.g., American production, machinery, livestock learning, pet training)	4.76	1.60
28. Eco center (e.g., ecological tent, environmental center, clean energy exhibit)	5.00	1.52
29. Contemporary, non-agricultural exhibits (e.g., science displays, technology)	4.67	1.46
30. Novelty (e.g., new programs, new foods, new drinks, new vendors, new exhibits)	5.63	1.03
31. Uniqueness (e.g., unique programs reflected by local culture, locally themed event)	6.11	0.91
32. Nostalgia (e.g., costume contents, music, exhibits, decor)	5.59	1.18
33. International bazaar (e.g., marketplace, exhibitions)	5.29	1.32
34. Lots of music (e.g., free open-air concerts, music bands on stages, local bands)	5.78	1.22
35. Lots of things to see (e.g., butter cow, celebrity crop art, sand sculpture, marching band, RV show)	5.37	1.61
36. Amusement (e.g., ride variety, overhead condoler ride, skylift)	6.16	1.13
37. Kids events (e.g., kid play zone, mini bumper cars, kids' artwork)	5.22	1.92
38. Food (e.g., variety of food selection, local food, drink, good taste)	6.49	0.91
39. Concession stands	5.92	0.95
40. Healthy food options (e.g., meeting all dietary restrictions, smart calories)	6.01	1.20
41. Food prices (e.g., reasonable, too high, too low)	6.50	0.80
42. Many kiosks (e.g., water, beer, wine, alcohol)	6.17	0.97
43. Exhibits (e.g., arts and crafts, local art and miscellaneous displays, photography exhibits)	5.49	1.04
44. Animals (e.g., livestock barn with shows, petting zoo, farm animal exhibits, pig racing)	4.84	1.74
45. Restrooms (e.g., availability, cleanliness, family option)	6.48	0.95
46. Handwashing options (e.g., facilities with soap and water and/or hand rub dispensers)	6.67	0.78
47. Fabulous atmosphere/ambiance (e.g., high, expected, low, needs work)	6.04	0.93
48. Shopping (e.g., vendor merchandize, flea market)	5.27	1.36
49. Event souvenirs (e.g., variety, good quality, affordable price)	5.31	1.27
50. A lot of giveaways/free samples (e.g., giveaways from vendors, sponsors, and other event stakeholders)	5.27	1.30
51. Fair grounds (e.g., cleanliness, space allotment, staffing)	6.37	0.95
52. Rest areas (e.g., lots of shade, shelter, cushioned seating, air conditioning)	6.31	0.86
53. Well laid-out design (e.g., landscape, street layout, midway, excellent arrangement)	6.20	0.84
54. Crowd control (e.g., well managed, needed additional, too much)	6.01	1.06
55. Signage (e.g., guidance, effectiveness, directional)	6.00	1.32
56. Well-organized event (e.g., smooth flow of the events, organized, not organized)	6.35	0.77
57. Good event implementation (e.g., event production, good stage management)	6.12	0.89
58. A real feel of local culture overall	5.86	1.19
59. Event staff (e.g., event staff availability, friendly and helpful staff)	6.56	0.71
60. Good security (e.g., sheriff's/police presence, cameras, gate checks)	6.45	0.94
61. Convenient camping ground/motor home/RV parking site (e.g., full hookups, electric, water, and sewer hookups, showers, laundry, Wi-Fi)	4.27	2.06
62. Amenities in the RV parking area (e.g., laundry room, bathhouse)	4.29	2.01
63. Surroundings (e.g., close to local restaurant, museums)	5.06	1.76
64. Nighttime show (e.g., fireworks, laser show)	5.51	1.45
65. Capacity management (e.g., less crowded at event venues)	5.82	1.18
66. Influence of other people/fairgoers (e.g., good manners, etiquette)	5.80	1.09
67. Socializing with other attendees (e.g., meeting/talking with new people)	5.20	1.48
68. Media participation (e.g., live TV shows, media booth/local TV and radio stations)	4.71	1.73
69. State government booths (e.g., DMV, license booth for hunting/fishing. Dept. of Natural Resources)	4.14	1.76
70. Enough sprayers to cool down visitors	5.18	1.53
71. Sidewalks and roads for people in a wheelchair	5.78	1.41
72. ATMs	5.67	1.54
73. First aid, EMT onsite	6.18	1.11
74. Water or hydration stations (free or low cost, availability, cleanliness)	6.54	0.73
75. A tent sponsored by political organizations	3.61	2.19
76. Tent for non-profit organizations (e.g., 4-H, Girl/Boy Scouts)	4.67	1.74
77. Tent for colleges and universities	4.59	1.63
78. Health-related programs (e.g., yoga and wellness programs)	4.98	1.52
79. Commercial booths	4.76	1.64
80. Facility maintenance (e.g., event venue/building/ground, clean atmosphere)	6.26	1.10
81. Feel for history (e.g., frontier village)	4.78	1.72
82. Photo zone	5.45	1.47
83. Place to nurse an infant	5.41	1.91
84. Cleaning and disinfection of the venue (e.g., amount, thorough)	6.52	0.85
85. Physical distancing (e.g., regulating the flow of the crowd, minimal physical interaction with people)	6.34	0.93
86. Mobile partitions in an event venue	5.41	1.47
87. Hand washing and sanitization stations	5.90	0.816
88. Cleaning staffs everywhere	5.93	1.24
89. Installation of a dispenser with disinfectant in public areas	5.92	1.16
90. Virtual event platform (e.g., virtual exhibit hall, attendee networks)	5.04	1.54
91. Available layout/seating charts/current capacity outside of the event venue	5.73	1.39
92. Daily event information (e.g., attendee size, news/stories, special event info., COVID-19 updates)	6.00	1.40
93. Good waste management (e.g., trash can and recycling stations, regular cleaning)	6.43	0.91
94. Regular downtown trolley service during the fair	5.39	1.39
95. Safety (e.g., clean, spacious, well-lite, pathways clear, exit signs)	6.50	1.16
After the event
1. Multiple exits	6.35	1.19
2. Farewell signs (e.g., “thank you” and “see you next year” message)	5.71	1.25
3. Safe, controlled crowd	6.55	0.78
4. Security personnel everywhere	6.03	1.24
5. Well-lit parking lot	6.54	0.84
6. Shuttle service to public transportation	6.09	1.08
7. Social media connection (e.g., post-event activities, photo contest, sharing event experience with others, connecting with attendees/vendors/sponsors, satisfaction survey)	5.37	1.49
8. Updated information of the fair regarding COVID-19	6.40	1.07