**3. Results**

#### *3.1. Top 10 Hashtags Used*

Table 1 below provides an overview of the most used hashtags during this time, showing that "wearamask" (*n* = 34) and "maskssavelives" (*n* = 11) appeared among the most used hashtags.


**Table 1.** Top 10 hashtags.

The hashtag "kidlitformasks" referred to authors of children's books who used this hashtag to highlight the importance of wearing masks. Twitter users also shared images from a Japanese manga series "haikyuu" and at the same time encouraged others to wear a mask using the "#haikyuu" and 'hq' hashtags, and these appeared as the 7th and 8th most popular hashtags that were used. As the discussion during this time was global in nature, there also appeared to be relevant hashtags from around the world such as "マスク" which refers to "mask" in Japanese. The word " ビ ー ル" also appeared, which is the Japanese word for "beer". This was because users conversed about a humorous mask invention which allows a user to store beer inside the mask which can be consumed. The hashtag "andhrapradesh" appeared as a popular hashtag as this referred to a state in the south-eastern coastal region of India and news reports were shared at this time related to reports about the death of a young person who was allegedly "beaten to death" by police for not wearing a mask.

#### *3.2. Overview of Network Structure*

Figure 1 is a social network graph of the discussion taking place during this time. The largest group (group 1) is an isolates group which contained users who sent tweets that did not contain mentions. Overall, the group resembled a community network shape because there were many groups of users conversing about this topic. NodeXL clustered users into di fferent groups based on mentions. A community cluster indicated that many users were talking to each other across several groups. It was also interesting to see influential users (indicated by larger circles) scattered around the network, which indicated that the topic brought in a wide range of influential actors.

#### *3.3. Top 10 Users Ranked by Betweeness Centrality*

Table 2 highlights Twitter users that were influential during this time. The users in the network were ranked by betweenness centrality, which is a measure of centrality. These users would have acted as important bridges within the network. The follower's column refers to the amount of followers each user has. Many of the influential nodes within the network derived from ordinary citizens who became important bridges in the network. A number of popular culture figures also appeared among the most influential users within the network.


**Table 2.** Top 10 users ranked by betweenness centrality.

#### *3.4. Top 10 Users Most Mentioned*

Table 3 provides an overview of users that were most mentioned during this time which ranged from political figures, organizational accounts, and popular culture related accounts. Certain users may not have tweeted about masks but were mentioned frequently by other users. Examining the influential users from this time highlighted that the discussion had been focused in and around the United States. However, it must be noted that not all account mentions may have been relevant to medical masks, as the band Slipknot are known to wear non-medical masks.


**Table 3.** Top 10 users most mentioned.
