Analysis of factors influencing public attention to masks during the COVID-19 epidemic-Data from Sina Weibo.

ABSTRACT

As we all know, vaccination still does not protect people from novel coronavirus infections, and wearing masks remains essential. Research on mask attention is helpful to understand the public's cognition and willingness to wear masks, but there are few studies on mask attention in the existing literature. The health belief model used to study disease prevention behaviors is rarely applied to the research on mask attention, and the research on health belief models basically entails the use of a questionnaire survey. This study was purposed to establish a health belief model affecting mask attention to explore the relationship between perceived susceptibility, perceived severity, self-efficacy, perceived impairment, action cues and mask attention. On the basis of the establishment of the hypothesis model, the Baidu index of epidemic and mask attention, the number of likes and comments on Weibo, and the historical weather temperature data were retrieved by using software. Keyword extraction and manual screening were carried out for Weibo comments, and then the independent variables and dependent variables were coded. Finally, through binomial logistic regression analysis, it was concluded that perceived susceptibility, perceived severity and action cues have significant influences on mask attention, and that the accuracy rate for predicting low attention is 93.4%, and the global accuracy is 84.3%. These conclusions can also help suppliers make production decisions.