Back

What Does the Public Want to Know About The COVID-19 Pandemic? A Systematic Analysis of Questions Asked in The Internet

George, J.; Gautam, D.; Kesarwani, V.; Sugumar, P. A.; Malhotra, R.

2020-09-23 epidemiology
10.1101/2020.09.15.20192039 medRxiv
Show abstract

Background Quora is a popular question and answer (Q&A) website that enables people to connect with others and clear their doubts about the coronavirus disease (COVID-19). In this study, we analysed the content, type and quality of Q&As in Quora regarding this pandemic, and compared the information with that on World Health Organization (WHO) website. Methods We conducted a systematic search to include 964 questions in Quora. The tone of the question was categorized as either positive (questions with a primary intent to obtain information), negative (questions which represent panic or are related to misconception/false information) or ambivalent. The two most helpful answers of each question were graded for accuracy, authority, popularity, readability, and relevancy. Results 462 (48%) questions were classified as positive, while 391 (41%) were negative. Number of views were higher for negative questions (11421 vs 7300, p=0.004). Majority of the questions were on social impact (N=217, 23%), followed by politics (N=122, 13%) and disease management (N=96, 10%). Positive questions had more accurate, but less popular answers (p<0.05). Information related to 229 (28%) questions were present on WHO website, while partial information was present for 103 (11%) questions. Discussion Higher views with negative questions suggest that false and panic promoting information is more likely to get public attention. A substantial amount of questions was related to the present and future effects of COVID-19 on social and personal lives of the users which were not readily available on official health websites.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.