1.1. Background
Middle-aged and elderly groups, defined as people over 46 years old are a very important part of communication with respect to the process of spreading rumours [
1,
2]. This is because, first, they account for 41.38% of the current total population in China according to the 2019 statistical yearbook [
3]. Second, older people have much social experience, and middle-aged people have a higher family status in China [
4]; therefore, other age groups tend to listen to both groups. Third, facing the flood of pandemic-related information, middle-aged and elderly groups who have low media literacy and relatively weak cultural qualities have low rumour-discrimination ability [
5,
6,
7,
8], which leads to being more susceptible to accepting rumours and to spreading rumours. Media literacy refers to people’s ability to choose, comprehend, question, evaluate, create, produce, and respond critically when confronted with information in various media [
9]. Cultural qualities are developed through education in the humanities and arts [
10], which is directly proportional to educational level, and rumour-discrimination ability refers to the ability to distinguish between right and wrong information. For example, due to the nuclear leak in Japan, there were rumours that the seawater contamination had led to salt shortage in China, which caused a salt scramble in 2011. According to an analysis by the public media, the middle-aged and elderly populations played a leading role in the spread of rumours and were also the main victims [
7]. During coronavirus disease 2019 (COVID-19), some widely circulated but unconfirmed news on social media, such as that salt brine mouthwash and smoked vinegar could prevent viruses, were eventually confirmed as rumours by public media and experts. According to a research in the US, people aged 45–65 share fake news three times as much as those aged 18–29 and those over 65 share fake news seven times as much as those aged 18–29 [
11]. A study shows that, in the context of the increasingly sound development of new media functions, the elderly population is enjoying short videos spread through WeChat (a chat software of Shenzhen Tencent Computer Systems Company Limited, Shenzhen, China), the authenticity of which cannot be verified. These contents were often identified as rumours [
12].
Fonzo and Bordia point out that the process of spreading rumours usually goes through three stages: generation, evaluation, and re-spreading [
13]. The concept of rumour re-spreading is relative to the initiation of the rumour. It emphasizes the role of the information audience in the dissemination process. The focus is on its diffusion behaviour after one hears the rumour [
14]. According to the existing literature, there are many variables affecting re-spreading, and the main research covers the following four aspects: the information itself, the disseminator, the individual audience, and the social environment [
7,
15,
16]. Allport and Postman note that the importance and ambiguity of information will contribute to the spread of rumours [
17]. Chours argues that, in addition to the importance and ambiguity of information that will cause the spread of rumours, individual intelligence, the level of knowledge, and moral values will also have an impact on the propagation of rumours [
18]. In addition, opinion leaders [
19], communication channels [
20], and trust in the rumour [
21] can also influence re-spreading of the rumours. There are many academic studies on the factors influencing rumour propagation. However, there is little research on the relationship between anxiety, rumour discrimination, opinion leaders, and trust in the process of rumour re-spreading as the influencing factors [
14].
In terms of subjects, the research on rumours of public health events in the existing literature mainly focuses on aspects of food safety [
22] and sudden occurrence [
23]. The existing research groups span across all ages, and there are fewer studies on elderly adults [
24]. In addition, the existing research on the influence of rumour re-spreading of various public health events has different factors [
7].
On 8 December 2019, the first case of a COVID-19 infection was found in Wuhan, Hubei Province. Subsequently, COVID-19, which is a communicable disease, has spread rapidly throughout Wuhan and other cities in China. The pandemic is rampant, and it is difficult for people to discern truth and rumours in a timely and accurate manner under conditions of panic and anxiety [
25]. The rumour that “Shuanghuanglian oral liquid (a famous Chinese traditional medicine, composed of honeysuckle, scutellarin, and forsythia and an antidote with the functions of relieving fever, cough, and sore throat) can inhibit COVID-19” has caused citizens across the country to quickly gather in local pharmacies to stock up on drugs [
26]. The spread of rumours related to the COVID-19 pandemic has not only exacerbated the possibility of the pandemic but also has stimulated public panic and caused chaos in social order [
12].
In this article, the middle-aged and elderly groups are used as the research object to help the public understand the rumour re-spreading behaviour of this group, which addresses the gap in the literature on rumour re-spreading in middle-aged and older populations. Meanwhile, in the process of rumour re-spreading, existing studies have found that anxiety level has an influential role in rumour propagation. However, the specific role of anxiety level in rumour propagation has not been discussed. Therefore, using the brine mouthwash rumour spread in COVID-19 as an example, we found that anxiety level influences rumour re-spreading behaviour in part through the degree of belief in the rumour. These findings complement research on the factors influencing rumour re-spreading and have important implications for the future design of public health policies.
1.2. Hypothesis Development
This paper’s theoretical foundations are rumour propagation theory and information diffusion theory.
Rumour propagation theory was first proposed by Allport and Postman and argues that rumours are the result of a combination of event importance and event ambiguity [
17]. When the environment and atmosphere are full of anxiety and uncertainty, it will increase anxiety and rumours are more likely to spread. Chours subsequently revised the theory, arguing that the spread of rumour is related to event importance plus event ambiguity and to public critical thinking of crowds [
18]. Public critical thinking is also an important factor that affects the spread of rumours, consisting of an individual’s intelligence, relevant level of knowledge, and moral values. Public critical thinking represents the insight into social events that comes from people’s intelligence or other relevant factors. Differences in personalities have a very different effect on the spread of rumours, which means that different moral values will affect the possibility that people believe and spread rumours.
Information diffusion theory suggests that the spread of most information is in the shape of an “S” curve. In the process of diffusion, early adopters provide necessary help for the subsequent popularity. These early adopters convince opinion leaders in their communities, often through interpersonal communication. The opinion leaders then spread their influence to audiences in their interpersonal communication sphere to allow more people to receive the information [
27,
28]. Moreover, information diffusion always takes place with the help of certain social networks. In the process of diffusion of information to the society, public communication through technology is effective in providing relevant knowledge and information. However, interpersonal communication is more direct and effective in persuading people to accept new information [
29].
Based on Chours’s study about critical thinking of crowds, the higher the individual’s intelligence and relevant knowledge is, the less likely the rumour is to spread; the better the moral accomplishment is, the less likely the rumour is to spread. In this paper, the ability to discern rumours is used to measure the individual’s intelligence and knowledge, and the perception of serious consequences of spreading rumours is used to measure the moral values.
Previous research has found that, in the face of rumours, the audience’s cognitive ability and emotional factors will have an impact on redistribution. In particular, anxiety is one of the most important manifestations of emotional factors [
30]. In early experiments, Anthony showed that anxiety is directly proportional to the number of rumours [
31], and in later empirical studies, Onook et al. also suggested that audience anxiety positively affects their willingness to re-spread [
32,
33,
34]. Therefore, the following assumptions are made:
H1: The degree of anxiety of the subjects is significantly positively related to their willingness to re-spread rumours.
The ability of rumour discrimination is to some extent an innovative concept in this paper. People’s behaviour relies on their own ability to think and judge, and a weaker ability to discern rumours limits this ability to some extent [
5]. In contrast, when an individual has enough relevant knowledge, such as if a person knows the origin of the virus and has protection knowledge, then they are less likely to spread rumours. Therefore, we propose the following hypothesis:
H2: Subjects’ ability to discern other rumours is significantly negatively related to their willingness to re-spread rumours.
People’s behaviour is regulated by social norms and their own values [
35]. If a person of high moral cultivation knows that what he or she hears is a rumour, then he or she will understand that the consequences of spreading may have bad consequences for the followers [
21], such as causing physical harm, sending the followers into panic, etc., and the perception of the severity of such consequences will prevent them from spreading the rumour. A related study also corroborated the idea that moral judgments are conducive to rumour-confrontation behaviour [
36]. Thus, we propose the following hypothesis:
H3: The participants’ perceptions of the serious consequences of spreading rumours are significantly negatively related to their willingness to re-spread rumours.
The news value is also known as the heat brought about by the spread of information. Opinion leaders can also increase news value by increasing the heat of the discussion [
37]. Opinion leaders are influential and active people in the process of information spreading and interpersonal interaction [
17]. In information dissemination, they can be nongovernment organizations, experts and scholars with a certain voice, or personal media accounts with a certain number of fans. To some extent, opinion leaders are more knowledgeable, talented, credible, and virtuous. According to the information diffusion model, we think opinion leaders can facilitate the re-spreading of rumours because they have higher levels of trustworthiness [
28]. Therefore, we propose the following hypothesis:
H4: Having an opinion leader or not has significant differences on participants’ willingness to re-spread rumours.
It has been shown that, when people believe rumours, they instinctively spread the topics they believe in. The more people believe a rumour, the more frequently they spread the rumour [
38]. Hua [
33] argued that the credibility of Internet rumours positively affects the audience’s willingness to redistribute rumours. In addition, people’s perceptions can have a driving effect on their behaviour. Trust is a person’s perception of things or opinions, which will have a propulsive effect in the course of action. In the process of rumour propagation, Liu argued that the degree of user trustworthiness plays a mediating effect in the influence of network density on the willingness to propagate rumours [
22]. Thus, we propose the following hypothesis:
H5: The degree to which subjects believe rumours is significantly positively related to their willingness to spread the rumours.
H6: Among the above factors that significantly affect participants’ willingness to spread rumours, the degree to which they believe rumours plays a mediating role.
Social network theory can infer that mass communication facilitates the spread of rumours, but interpersonal communication makes people believe them more. The former gets information from traditional media such as television, newspapers, or new media such as the Internet, and the latter gets information from familiar family and friends. Some studies have suggested that intimate social relationships have a positive effect on the willingness to spread rumours [
16,
20]. This research proposes the following hypothesis:
H7: The communication channels have significant differences on participants’ degree of approval of rumours and their willingness to re-spread rumours.
The overall conceptual model is shown in
Figure 1.