1. Risk Perception

The concept of risk perception refers to the subjective feeling and understanding of various objective risks in the external environment. At present, there is no unified conclusion on the definition of risk perception. It is now thought that the earliest risk research was that done by O.K. Burell on the impact of scientific analysis psychology on investment behavior. Bauer proposed the application of risk in the management field and interpreted risk perception as consumer awareness and subjective perception [18]. Hugstad, Taylor, and Bruce considered risk perception to be the risk a consumer may feel when purchasing a product or service [19]. McCaffrey divided risk perception into probability perception and loss perception [20]. Wang Wei-quan suggested that a high level of risk perception may lead to over-amplification of the individual’s influence on negative events, resulting in risk aversion [21]. Yeung and Morris proposed that, at any stage of consumer purchase, when consumers perceive security risks, their purchasing behavior will change accordingly. Scully and Joanne argued that consumers’ perception of risk tends to be higher than the actual level of risk, and consumers’ reaction to risk tends to be excessive and irrational.

There are many methods to measure risk perception. One of the most important research methods is the psychological measurement paradigm, which was put forward by many scholars and mainly represented by Slovic [10]. These scholars classified the risk according to its characteristics to understand why people react differently to risk. The main tool of this method is the questionnaire, with some subjective multiple-choice questions, the results of which determine the participants’ risk perception of a particular event. Roselius classified risk perception into four categories: financial, physical, time, and hazard risk [22]. Jacoby and Kaplan [23] further subdivided risk perception into five dimensions: functional, physical, psychological, social, and financial risk. Mumel added time risk on this basis, and the study of risk perception focuses on these six aspects. Xie and others found that duration, control, severity of consequences, and familiarity also affect risk perception [24].Zhou divided the perception of earthquake risk into four factors: probability, fear, influence, and control [25]. Liu and Hu believe that the psychological attribute of risk perception is stronger, and that the subjective evaluation of events made by the subject reflects different values under different cultural and ideological backgrounds and it is both a psychological paradigm and a cognitive process [26].

The concept of risk perception is widely used in the field of consumption. Due to the asymmetry of information, consumers cannot accurately predict the outcome of purchase, and this kind of psychological pressure to the uncertain result is the consumer’s risk perception in the process of consumption. Under the influence of risk perception, consumers may change their original attitude towards products, reevaluate the purchasing suggestions made by the people around them, and, fifinally, change their purchasing intention and behavior. The COVID-19 outbreak will have an impact on the risk perception of consumers, consequently affecting their willingness to consume products from restaurants. In this study, risk perception was combined with an epidemic background, and epidemic risk perception was used as an independent variable to study its predictive effect on consumer’s willingness to consume products from restaurants.In previous studies on risk perception and willingness to consume products from restaurants, Wojciech Kolanowski studied young Turkish and Polish consumers’ perceptions of food safety risks when they ate out by using a questionnaire, and young Polish consumers were found to be less aware of food safety risks when eating out than Turkish consumers [8]. Marcel Levy de Andrade, through structured questionnaires, evaluated the knowledge, risk perception, and optimistic bias of food handlers and consumers of restaurants and the relationship of these variables with the FBD risk of these establishments, and the results showed that consumers may have incorporated a sense of affection and identity to a place, associating it with making their own meals at home. Therefore, the consumer may not differentiate restaurants with regard to food safety [27].

2. Willingness to Consume

2. Willingness to Consume

Scholars in the academic circle have put forward several opinions on the definition of consumer will. Consumers’ willingness to consume is the premise of consumer behavior. Ajzen first defined willingness to consume as the likelihood that a consumer will consume a product or service [28]. Schiffman, Kanuk, and Wisenbilt (2010) held that willingness to spend is the probability or subjective probability that a consumer is willing to spend on a product [29]. Kotler thought that the consumption intention is a subjective attitude which will influence a consumer’s future consumption behavior because of the consumption behavior of others or themself. Bi suggested that willingness to spend is the possibility that consumers will buy a product again after they have a better understanding of the product [30]. The academic circle’s view on consumption willingness is generally regarded as the enthusiasm and subjective attitude of consumers in producing consumption behavior and making a consumption decision, which is affected by a change in other factors.

In terms of the influencing factors of consumers’ willingness to spend, Spiggle and Sewall considered that the factors that have a direct impact on consumers’ purchasing behavior are mainly the characteristics of consumers [31]. At present, the main influencing factors for willingness to consume are as follows: (1) the influence of consumer experience on consumer willingness, where the consumer experience is when the consumer uses the goods or services and influences whether the consumer will make a second consumption;(2) the effect of consumer-perceived value on consuming willingness, and the fact that the evaluation of a product after use is user-perceived value, which has an important effect on consumers’ consuming willingness [32–34].In the aspect of willingness to consume products from restaurants and its influencing factors, Chung-Te Ting constructed a contingent valuation model to study the importance of pre-processed services in the context of a possible restaurant service crisis [35]; Lalwani studied young couples in Singapore, and found that a spousal relationships have a significant impact on whether or not they go to a fancy restaurant. In a value–attitude–behavior model study [36], Jinhyun Jun found that customers’ health values and attitudes toward low-fat or low-calorie foods had an effect on their willingness to consume products from restaurants [37]; David Marshall used a mixture of correspondence analysis, cluster analysis, and discriminant analysis to investigate the interaction between British and Australian students in eating situations, places, and food choices [38]; Joo Ahn and others used the method of linear regression to analyze the impact of brand value on the customer satisfaction of fast food restaurants, and investigated how customer satisfaction further affects consumers’ willingness to spend—the results showed that food quality, brand image, brand awareness, and brand association were important factors affecting customer satisfaction, and there was a positive correlation between customer satisfaction and customers’ willingness to consume [39].

In the aspect of research on the relationship between public health emergency and willingness to consume products from restaurants, Arif Billah used the theory of planned behavior, and investigated the factors that affected consumers’ willingness to consume halal food during the epidemic in southern Thailand. The results showed that consumers’ habits and knowledge of halal food significantly influenced their behavior [40]. Maria studied the impact of the 2019 new coronavirus epidemic on the eating habits of Italian consumers, and the results showed that the impact of the epidemic on consumer behavior can be divided into changes in food buying, eating habits, and other behaviors [41]. On the basis of the above research, this paper introduces the concept of epidemic risk perception and risk attitude to study its impact on consumers’ willingness to consume products from restaurants.

3. Risk Attitude

Risk does not refer to the likelihood of loss in everyday life, but rather to uncertainty, including loss and gain, and risk attitudes refer to the degree of aversion or preference that people have for the outcome of such uncertainty. At present, there is no consensus on the precise definition of risk attitude. Weber, Blais, and Betz considered risk attitudes to be the attitudes and preferences of individuals when making economic decisions in uncertain future situations [42]. Schroeder, Tonsor, Pennings, and Mintert believe that risk attitude is the individual’s willingness to accept risk consistently in different situations: risk seekers tend to choose higher risk decisions, risk-neutral people will choose the average rate of return in different situations, and risk-averse people tend to minimize future risks in  their decision-making [43]. Zhang and Wei defifined risk attitudes as different risk choices made by different individuals for the same situation, which are inflfluenced by individual preferences, social status, and economic conditions [44]. The risk attitude referred to in this paper is the willingness of consumers to bear the risk when facing uncertainty in the process of restaurant consumption in the context of an epidemic situation.

3. Risk Attitude

At present, there are many methods to measure risk attitude. Because the questionnaire survey method is simple and easy to operate, the cost of time and resources is low, and the requirements for the cognitive ability and understanding ability of the subjects are low. Combined with the specific situation of this study, this paper adopts the questionnaire survey method of the direct measurement method to measure the risk attitude.

The influencing factors of risk attitude of scholars mainly include individual difference, external environment, and risk perception.

From the perspective of individual difference, the main influencing factors of risk attitude are age, education level, and income level. Zhou and colleagues found that, among risk-neutral individuals, the male samples were more than the female samples, while, in risk aversion and risk preference, the male samples were less than the female sample [45]. Arrow found that, the higher the individual’s income level, the stronger the risk-averse attitude. However, in a study by Friend and Blume, they found that, the higher the individual’s income level, the higher the proportion of risky assets in their total assets [46].The influence of external environment on individual risk attitude is mainly reflected in scale reward, economic field, and risk-sharing. In terms of scale rewards, Binswanger found in a gambling experiment that, as the rewards increased, the participants became more risk-averse [47].

As far as the field of economic activity is concerned, Elke found a high correlation between the field of economic activity in which an individual participates and their risk attitude in the course of measuring the scale [42]. In terms of risk-sharing, Wang and Huang found that individuals are less risk-averse when sharing risk than when taking it alone, with this effect being particularly pronounced among vulnerable groups [48].

From the perspective of risk perception, Palich and Ray proposed that risk perception influences individual decision-making by influencing an individual’s risk attitude [49]. Lv found that risk preference has an indirect effect on an enterprise’s willingness to purchase insurance through a study on the impact of managers’ risk attitude and risk perception on their decision-making. This indirect effect is achieved through the mediating role of risk perception [50].