4. Discussion
Since the National Health Commission identified COVID-19 as a B type infectious disease officially, COVID-19 influenced the psychological states of people across China. This study collected active Weibo users’ data, and conducted sentiment analysis during 13–26 January, 2020. We used OER to acquire the psychological states, and found that Weibo users’ psychological conditions significantly changed under the outbreak of COVID-19.
The findings showed that people’s concerns by linguistic expression increased after January 20. We observe an increase in health and family, while a decrease in leisure and friend. Uncertainty of the upcoming situation causes cognitive dissonance and insecurity; this produces a feeling of mental discomfort, leading to Weibo’s activity oriented toward dissonance reduction and keeping security on health and family relationship [29]. According to the theory of BIS, people behave in a more reticent and conservative way when they feel threatened by disease [30]. Therefore, staying at home with family and reducing recreational activities seems to be a safer way to prevent illness. It also indicated that people begin to care more about their health and were more likely to seek social support from their families rather than getting together with friends, which suggested that people’ interests and attention were influenced by the restricted travel policy and self-isolation regulations from the health authorities and central government.
Affected by COVID-19, messages related to death and religion became salient after 20 January. Reports showed severity and potential mortality of COVID-19. Research confirmed that people tended to respond to emergencies such as stress or death in the way of religion, which can comfort tense moods and bring more positive emotions [31]. That is why people prayed for the county through religion or other beliefs, leading to the phrase that appeared most frequently on the Internet at that time: God bless China.
People showed more negative emotions (anxiety, depression, and indignation) and less positive emotions (Oxford happiness) after the declaration of COVID-19, which was supported by the theory of BIS, i.e., people did generate more negative emotions for self-protection [3,4]. These results are consistent to previous studies as well, which found that public health emergencies (e.g., SARS) triggered a series of stress emotional response containing a higher level of anxiety and other negative emotions [32,33]. Meanwhile, the confirmation that COVID-19 could be passed from person to person on 20 January, which was inconsistent with previous reports, lead to quite a number of people being unsatisfied with misinformation published from provincial governments (e.g., Hubei) and ineffective regulatory actions, causing an increase in indignation. However, it’s worth noting that the word frequency of positive emotions increased after 20 January, which seemed to be inconsistent with the theory of BIS. In fact, positive emotion includes words such as faith and blessing, which are more inclined to reflect group cohesiveness rather than pure personal emotions (e.g., happiness). Researchers found that group threats (e.g., natural disasters and epidemic diseases) made groups a community of interests, resulting in more beneficial behaviors and social solidarity, which indicated higher group cohesiveness [34]. For example, lots of provinces (e.g., Sichuan Province, Shandong Province, etc.) formed medical teams to help the Hubei province, which was the worst affected area. Many people donated money and supplies to Hubei Red Cross to support the control of COVID-19.
Furthermore, social risk judgement was higher and life satisfaction was lower after the declaration of COVID-19. It is consistent with the theory of BIS, which found that when social uncertainty increased, such as unknown etiology and ambiguous route of transmission, people developed the negative cognitive assessment (e.g., higher sensitivity of risk judgment or risk perception) so that they could discover potential infection sources in time and avoid infection [2,35]. Not only that, people’s fear of potential risk and lack of controllability caused by COVID-19 brought about higher risk judgement as perceived risk theory claimed [10]. Moreover, some preventive policies and regulations in terms of travel restriction and self-isolation made the quality of life worse, reflecting in lower life satisfaction.
The following briefly foregrounds some of the study’s implications for policy makers and clinical practitioners (e.g., social workers, psychiatrists, and psychologists) plan and fight against COVID-19. For policy makers: (1) develop a consistent policy and procedure for reporting the latest confirmed cases, recent death toll, and other data about the epidemic situation. For example, the surge of cases on February 12th did not mean that the situation has been out of control, but because of the new diagnostic criteria introduced. It is important to let people understand the data properly to reduce excessive stress responses (e.g., anxiety, depression, etc.) brought on by inappropriate perception. (2) Expand public awareness of continuous progress in decision-making measures. Since indignation may come mainly from mistakes and deficiencies in preventing and controlling the epidemic, it can effectively decrease indignation if public awareness and involvement are provided. (3) Ensure the supply of medical treatment service. It is critical to set up medical service to treat the disease, and let people know how to access it conveniently. People can get help in time if they are infected. It can improve people’s sense of control over risks, thereby avoiding excessive social risk perception. (4) Provide more in-door entertainment services to address good quality of life. People may be more willing to cooperate when their living and entertainment requirements are met, such as online shopping, entertainments, etc. For clinical practitioners: (1) adjust consultant configuration rationally and cooperate with each other. Psychological consultants should grasp the epidemic information correctly and conduct science popularization during counseling. Social workers can help solve practical problems in life. These actions can improve the sense of stability and relieve anxiety and depression. (2) Deliver necessary psychosocial therapy in various ways. Considering the particularity of self-isolation, relevant hotline counseling and online consulting should be applied in practice.
Several other points should be considered when generalizing this study’s findings. First, as Weibo users are mainly young people, the results may be biased to some extent. In addition, the current analysis is based on a weekly basis, with a relatively large granularity, which has certain influences on reflecting the changing trend of social mentality in a timely manner. In further studies, we will try to expand the range of sex and age and predict psychological traits in a finer granularity. Previous studies indicated that people tended to exaggerate attitudes and prejudices, especially when they felt more vulnerable to disease transmission [36]. It inspires us to try to build a prediction model which can predict people’s attitudes and beliefs against the virus through online Weibo data for further understanding of psychological impacts of public health emergencies.